FTC Staff Comment Before the Federal Aviation Administration Concerning a Proposed Congestion Management Rule

In the Matter of Congestion Management
Rule for LaGuardia Airport
Docket No. FAA–2006–25709
Notice No. 08–04
Comment of the Staff of the Bureau of Economics
of the Federal Trade Commission1
Summary
The staff of the Bureau of Economics of the Federal Trade Commission (FTC)
appreciates this opportunity to provide its views on the economic issues raised in the
Federal Aviation Administration (FAA) Supplemental Notice of Proposed Rulemaking
regarding the “Congestion Management Rule for LaGuardia Airport.”2 In this public
notice, the FAA has discussed two broad auction-based approaches that are designed to
provide a more efficient allocation of take-off and landing rights to air carriers and to
provide incentives for the carriers to use capacity more efficiently. Although this
proposal potentially raises some interesting legal and regulatory issues, our comment
focuses on the economics underlying this proposal.
Currently, congestion at LaGuardia is controlled by limiting the number of slots, or rights
to land or take-off, allowed during each hour of operation. The two alternative auction-
based approaches proposed for allocating a limited number of these slots reflect the
FAA’s desire for market forces to determine the most efficient uses for the limited
capacity available at LaGuardia. At present, take-off and landing fees are time-invariant
and based solely on the weight of the aircraft. 	Accordingly, these fees are unrelated to
the demand for takeoffs and landings. We believe that the imposition of slots and slot
auctions represent one promising approach to relieve congestion at the busiest airports
and efficiently allocate scarce airport capacity, but that careful attention needs to be paid
to the design and testing of candidate auction mechanisms before they are implemented.
Furthermore, alternative policies such as congestion-based take-off and landing fees,
which create a similar change in air carrier incentives, should also be considered.
1 This letter expresses the views of the Federal Trade Commission’s Bureau of Economics. The letter does
not necessarily represent the views of the Federal Trade Commission (Commission) or of any individual
Commissioner. The Commission has, however, voted to authorize us to submit these comments.
2 Federal Aviation Administration, Congestion Management Rule for LaGuardia Airport, Docket No.
FAA–2006–25709; Notice No. 08–04, 73 FED. REG. 20846 (April 17, 2008) (“Proposed Rules”) at
http://edocket.access.gpo.gov/2008/E8-8308.htm. Also note that a similar proposal has been made to
alleviate congestion at two other New York area airports, JFK and Newark. See Federal Aviation
Administration, Congestion Management Rule for John F. Kennedy International Airport and Newark
Liberty International Airport, Docket No. FAA–2008–0517; Notice No. 08–05, at
http://www.dot.gov/affairs/aviation080516/JFK_EWR_NPRM_final.pdf
1

Interest and Experience of the Federal Trade Commission
The FTC is an independent administrative agency charged with maintaining competition
and safeguarding the interests of consumers.3 The staff of the Federal Trade
Commission, typically upon request from federal, state, or local governmental bodies,
comments on a variety of regulatory proposals that may affect competition, consumers, or
economic efficiency. In the course of this work, as well as in antitrust and consumer
protection research, nonpublic investigations, and litigation, the staff applies established
principles and recent developments in economic theory to competition and consumer
protection issues.
The staff of the Bureau of Economics has a longstanding interest in issues involving
airport regulation, as well as a general interest in issues involving competition and
regulation in the airline industry. This interest has been reflected in comments submitted
by the Bureau of Economics staff in previous FAA administrative proceedings on slot
allocation and transfer methods and in previous FAA proceedings involving other aspects
of airline competition and regulation.4 In addition, the staff has issued research reports
on slot allocation and airline deregulation.5
Proposed Rule Changes
Under the proposed rule, the FAA would auction a limited number of landing and take-
off slots but grandfather the majority of current operations at LaGuardia Airport. A
limited number of slots throughout the day would be selected by the FAA and affected
airlines to be auctioned. The ownership of slots resulting from the auctions would
terminate ten years from the first auction. The auctions are to be staggered within the
first five years of the rule, and since all slot ownership terminates at the same date, this
results in slot tenures of six to ten years. Two options are given for the proposed rule: (1)
auctions where the proceeds are used for LaGuardia investment and that are accompanied
3 Federal Trade Commission Act, 15 U.S.C. ' 45.
4 See the comments of the staff of the Bureau of Economics of the Federal Trade Commission, Study of the
High Density Rule, FAA Docket No. 27664 (November 23, 1994); High Density Traffic Airports: Slot
Allocation and Transfer Methods, FAA Docket No. 25758 (November 15, 1991); Slot Allocation
Alternative Methods, FAA Docket No. 24110 (July 27, 1984). Comments on other aspects of airline
competition and regulation include, Elimination of Airport Delays, FAA Docket No. 24206 (August 30,
1994); Discussion Authority for Agreement to Shift Schedules, Department of Transportation, Docket No.
44634 (February 17, 1987); and Charges for the Use of Metropolitan Washington Airports, FAA Docket
No. 25204 (April 13, 1987). See also Comments of the Bureaus of Economics, Competition, and
Consumer Protection of the Federal Trade Commission, Massport Program for Airport Capacity Efficiency
(February 29, 1988).
5 See D. Koran and J. Ogur, Airport Access Problems: Lessons Learned from Slot Regulation by the FAA,
Bureau of Economics Staff Report to the Federal Trade Commission (May 1983), and J. Ogur, M. Vita, and
C. Wagner, The Deregulated Airline Industry: A Review of the Evidence, Bureau of Economics Staff
Report to the Federal Trade Commission (January 1988).
2

by the retirement of a number of slots and (2) auctions of a greater number of slots but
where the proceeds go to the airline that held the slot originally. According to the notice,
the number of slots to be auctioned under each option is equivalent to 7 or 18 daily
roundtrips, respectively.
The proposed rule has two major procedures: the selection of slots to be auctioned and
the auction itself. Based on the schedule that was in place during the second week of
January 2007, the slot selection process begins by determining the number of operations
each airline will lose. The slots corresponding to the operations an airline will lose are
called “limited slots.” For both options, each airline is “grandfathered” up to 20
operations, which creates a baseline of the number of operations in the second week of
2007 minus twenty. Under option one, the number of slots a carrier will lose is ten
percent of the baseline. Under option two, the number of slots a carrier will lose is
twenty percent of the baseline.
Once the number of operations each airline will lose is determined, the time window for
each of these limited slots must be established, which occurs in three steps. During the
first step (only for option 1), any hour window with more than 75 operations will have
excess operations retired. Airlines are randomly selected to have an operation retired
during each of these windows until the number of operations is reduced to 75. No airline
will lose more than one operation in this step. During the second step, the airlines
designate the time window for one-half of their limited slots. During the third step, the
FAA determines the time window for each of the remaining limited slots. These
assignments to each time window are determined in a random process that also attempts
to spread the limited slots throughout the day. However, this three-step process may not
spread slots up for auction evenly throughout the day.
The final step in the slot selection process is determining the year in which each of an
airline’s limited slots will “revert” and be subject to auction. Each airline’s limited slots
are assigned a reversion year of either 0, 1, 2, 3, or 4 where a slot of reversion year x will
be auctioned x years after the start of the program. These assignments are arranged so
that each airline’s limited slots have an average reversion year of 2.
After the slots to be auctioned (or retired) are selected, auctions are conducted for the
limited slots at 0, 1, 2, 3, and 4 years after the start of the program, depending on each
limited slot’s reversion year. Slots that are obtained through the auctions will remain in
the purchasing airlines possession until ten years after the start of the program and are
called “unrestricted slots.” The exact auction format and procedure has yet to be
announced, but the notice refers to an ascending clock auction. In this auction, the
auctioneer announces the prices for each slot. Next, the airlines submit which slots they
would purchase at those prices. Then, the auctioneer raises the prices of those slots for
which demand was greater than supply. This process iterates until the prices are such that
demand is equal to supply.
In addition to the primary auction of unrestricted slots, the proposal also discusses the
FAA’s desire to create a secondary market for all slots. Due to concerns about
3

incumbents having incentives to exclude entrants, the FAA proposes a secondary market
in which bidders are initially anonymous and cash bids are submitted for slots. Once a
slot has been won, the parties to the transaction have the option to renegotiate the deal to
include non-cash assets, such as other slots. Negotiation over these non-cash assets
would require dropping the anonymity. If no alternative agreement is reached, the deal
proceeds under the terms of the cash-only bid.
Economic Analysis of the Proposal
Background
In general terms, congestion arises from a basic economic problem of a failure to
correctly allocate a scarce resource. In this context, time on the runway is the scarce
resource. Congestion is caused by a failure to balance supply and demand efficiently. At
airports without slot controls, runway time is allocated based on an airplane’s position in
a queue. This leaves airlines free to schedule as many flights as they want at any time.
Airlines may choose to schedule a flight when there is no available runway capacity
especially since some of the queuing costs are borne by other airlines and their
passengers. At some airports, this results in over-scheduling.
To solve the congestion problem, a mechanism has to be implemented to allocate the
scarce runway time. Relieving congestion alone does not necessarily accomplish an
optimal outcome; ideally the allocation should be efficient. A resource is employed
efficiently if it is used in the most valuable manner possible. In this context, an efficient
allocation of runway time is one that takes into account both the number of consumers
served as well as the value different consumers place on traveling at congested times. It
is unlikely that a regulator could determine the value of runway time to all potential
users, and assign the resource accordingly. One approach to reduce congestion in an
efficient manner is to price runway time using some mechanism that matches demand
with supply.6 Economic efficiency requires that when an airport is congested the price
for the use of limited runway capacity must rise to ensure that this resource is allocated to
the consumers who value it the most. With an increase in price, those who benefit most
from using the congested airports (and therefore have the highest willingness to pay) will
be the ones who use it, while those whose relative benefit is lower will use an alternative.
The current proposal suggests using auctions to set these prices for a limited number of
slots.
6 It should be noted that with the network effects and interdependencies of valuations across bidders that
are likely in these auctions, as discussed below, there is no guarantee that a market clearing set of prices
can be found by a non-combinatorial auction. For a discussion of this in an auction environment, see
Bykowsky, M., Cull, R. and Ledyard, J., “Mutually Destructive Bidding: The FCC Auction Design
Problem.” Journal of Regulatory Economics; 17:3 205-228, 2000. Note, however, that congestion pricing
would also fail to find an efficient market-clearing set of prices in the same situations that would cause
non-combinatorial auctions to fail.
4

Potential Benefits of the Proposal
We agree that the proposal represents a step towards a more reasonable approach to
improve the efficiency of the allocation of scarce resources at LaGuardia Airport.
Efficiency of the distribution of landing and take-off slots is a concern because the typical
solution to runway congestion, administratively assigned slots, is likely inefficient. An
auction of landing and take-off slots is a mechanism that could increase the efficiency of
the slot distribution, if successfully implemented.
Historically, the FAA has solved the worst congestion problems in a likely inefficient
manner by administrative assignment of landing and take-off times (slots). For an
allocation of slots to be efficient, each slot should be assigned to the airline that places
the highest value on it. One airline may place a higher value on a slot than another airline
because it can serve a greater number of customers with the slot or can serve customers
who place a higher value on the flight that the slot will enable. The current system is
most likely inefficient because typically these assignments are based on some measure of
historical usage and do not necessarily reflect any airline’s value of the slot.
Auctions have the potential to produce an efficient distribution of slots. Auctions use a
competitive bidding process to determine which airline receives which slot. Because the
airline that values the slot the most can be expected to bid the most for the slot, the
expected outcome of the auction is likely more efficient than the current allocation.7
The higher prices for scarce runway time that would result from an auction may cause
airlines to use larger planes (“up-gauge”) or to change routes to accommodate higher
value customers. Up-gauging and route changes may occur as auctions increase the
efficiency of the distribution of slots since slots are allocated to the airlines who value
them the most. An airline that is able to serve a greater number of customers by up-
gauging or by changing routes will value a slot more than another airline that cannot do
so and therefore should win the slot in an auction.
In addition to enhancing efficiency, auctions may also increase the competitiveness of the
market by replacing historically-based slot allocations with market-based slot allocations.
Historically-based slot allocations are a relatively anti-competitive mechanism because
they lock in the market structure and increase the difficulty of entry. To the extent that
slot allocations will still largely be based on historical precedent under this proposal, we
feel that a robust secondary market as envisioned in the proposal could mitigate the
inefficiencies inherent in such an assignment. We have commented previously that
7 See Whalen, W.T., Carlton, D., Heyer, K., Richard, O. “Proposal For A Market-Based Solution to Airport
Delays.” Economic Analysis Group Discussion Paper, 07-14, October 2007. However, as they note, one
potential source of value of a slot is that it prevents a competitor from using the slot, with potential negative
implications for competition. For a parallel argument in the patent context, see Gilbert, R. and Newbery,
D. "Preemptive Patenting and the Persistence of Monopoly," American Economic Review, 72(3), 514-26.
Given the very limited number of slots to be auctioned under this proposal, consolidation of market power
is probably not a concern in the primary auction, but care should be taken to make sure the secondary
market for slots does not create a mechanism for a carrier to purchase market power.
5

anonymity can encourage more competitive bidding in auctions, and believe the same is
true in this secondary market.8
Although it may be difficult to design a perfectly efficient allocation mechanism, we find
the notion of a limited step toward a market-based approach a reasonable way to attempt
to increase the efficiency with which existing resources are used at LaGuardia Airport.
The use of auctions to find an efficient allocation is based on sound economic reasoning.
However, we would like to raise some concerns about the limitations of the proposal, the
likelihood of the need for adjustments, and the value of trying congestion pricing in
addition to auctions.
Limitations of the Proposal
We believe a well designed mechanism can increase the efficiency of slot allocations, but
it is not possible to comment on the auction design because the proposal only lays out
broad features the FAA would like to see included in the mechanism, such as an
endorsement of package bidding and a bulletin-board based secondary market. As
evidenced by the experience with auctions (as explained below in “Lessons from
Spectrum Auctions”), an auction’s performance depends critically on the details of its
design. Not only should the design be tailored to the specific features of the industry, but
the design should be vetted publicly. A poorly designed mechanism, as evidenced by the
California electricity markets, can lead to disastrous results. At the conclusion of a
thorough investigation of the problems in that market, which included rolling blackouts,
the Federal Energy Regulatory Commission concluded “that significant supply shortfalls
and a fatally flawed market design were the root causes of the California market
meltdown.”9 Therefore, one cannot assume that just any market-based mechanism will
produce efficient outcomes, especially in situations where some actors may have some
market power. Below, we discuss features of the proposal that may affect its
performance as well as some of the important characteristics of the market that should be
taken into account when designing the auction.
The proposal to auction slots at LaGuardia is limited in scope, which is prudent given the
FAA’s and the airlines’ inexperience with the use of auctions to reallocate slots, the
FAA’s goal to implement the new mechanism by the end of the year, and the FAA’s
desire to minimize the disruption of operations at the airport that any reallocation of slots
might impose. However, the limited nature of the auctions also limits the degree to
which they may improve the allocation of scarce slot resources. Both of the alternative
8 See the comments of the staff of the Bureau of Economics of the Federal Trade Commission, In the
Matter of Auction of Advanced Wireless, FCC Docket No. 06-30 (June, 2006) at
http://www.ftc.gov/os/2006/03/ReplyoftheFTCBureauofEconomicsOnFCCAWSAuctionAUDocket06-
30.pdf.
9 Final Report On Price Manipulation In Western Markets: Fact-Finding Investigation 63
of Potential Manipulation of Electric and Natural Gas Prices, Docket No. PA02-2-000, Prepared
by the Staff of the Federal Energy Regulatory Commission (March 2003), at ES-1, available at
http://www.ferc.gov/industries/electric/indus-act/wec/enron/info-release.asp.
6

auction options proposed by the FAA involve a very small percentage of the total slots
available and allocated to the airlines.10 Accordingly, the vast majority of slots will not
be allocated by a market mechanism. In addition, the slots available for auction are
biased toward the least valued slots because at least half of the slots to be auctioned are
chosen by the airlines. Along with limiting the degree to which the auctions can be
expected to enhance the efficiency of slot allocations, this latter feature may also act to
diminish the potential competitive position of any airline that desires to initiate or expand
service at LaGuardia through the purchase of slots at auction. To the extent that this
proposal simultaneously introduces the primary auction, a secondary market, and a
reduction in the total number of slots available, it may be difficult to identify the policy
change that was responsible for any observed outcome, whether it be positive or negative.
For instance, if airfares for flights through LaGuardia increase substantially after the
proposal has been implemented relative to other flights, it may be difficult to determine if
the magnitude of the increase was primarily the result of the reduction in the total number
of slots, the airlines having to pay more for prime runway time, or reasons unrelated to
the proposal.
One feature of the airline industry that may have an important impact on the outcome of
these auctions is the existence of network effects. Network effects occur because the
value of a flight depends on how it is coordinated with other flights. Examples of
network effects are that connecting flights need to be coordinated and that planes need to
be scheduled so that they may serve the next flight at the destination. When airlines are
determining their bids in the auctions, they must take into consideration all the
complexities introduced by these network effects as well as the likely actions of
competitors. Designing auctions that are likely to result in efficient allocations in the
presence of network effects such as these is not a simple task. Since an auction design
may not fully be able to account for these complexities, the auctions may fail to produce
an efficient outcome. For example, an airline may not obtain slots vital to its national
schedule. Thus, implementation of an auction ill-suited to the specifics of the demand
conditions of this market may not provide the degree of network efficiency that one
might otherwise expect from a market-based allocation. The proposal suggests that the
auction design should allow an airline to bid for packages of slots so that it can attempt to
jointly purchase combinations of slots whose value to the airline depends on ownership
10 The FAA proposal defines a slot as “the operational authority assigned by the FAA to a carrier to
conduct one scheduled arrival or departure operation at LaGuardia on a particular day of the week during a
specific 30-minute period.” The proposal caps operations at LaGuardia at 75 scheduled slots (take-offs or
landings) per hour, which are to be allocated to the airlines, plus 3 unscheduled slots per hour. The caps
would apply during the period 6 a.m. to 9:59 p.m. Monday through Friday and 12 noon through 9:59 p.m.
on Sunday. Accordingly, the proposal creates 6,750 scheduled slots per week, 1200 per day Monday
through Friday and 750 on Sunday. Under Option 1, the number of slots will be reduced by 2% over five
years; under Option 2, there will be no reduction in the number of slots [see the SNPRM at 20848 for a
summary of the two proposals]. Yet, the FAA anticipates that each of five annual auctions will make
available only 14 (Option 1) or 36 (Option 2) slots [SNPRM at 20855]. Even if the number of slots
anticipated to be auctioned is 14 or 36 per day (as implied by the accompanying discussion in the SNPRM)
rather than per week (as implied by the definition of a slot in the SNPRM), the number of slots to be
auctioned each year still represents a very small fraction of the total number of slots to be allocated to the
airlines.
7

of the other slots in the package. If slots are used to control congestion and package
bidding is implemented effectively, auctions should enhance efficiency vis à vis non-
market-based administrative allocations of slots by ensuring that an airline that values a
slot the most is likely to obtain it.
Should one of the two proposed options be implemented, the FAA may be tempted to
view its experience at LaGuardia as a basis for implementing additional slot auction
programs at congested airports outside of the New York area, for example at Chicago’s
O’Hare International Airport.11 However, establishing slots and slot auctions at
additional airports may cause significant problems that will not be observed when
auctions are limited to just LaGuardia and may worsen problems that do arise. 12 The
network-related problems previously discussed in the context of auction design may
worsen should airports with flights serving LaGuardia also allocate slots through
auctions. Every flight from one auction-allocated airport to another would require
coordinated slots at both airports for both take-off and landing. Designing an auction
mechanism that allows bidders to coordinate slots at LaGuardia and at other airports
serving LaGuardia may prove to be difficult, especially given the small number of slots
to be auctioned. To a certain extent, the secondary market for slots may lessen the
coordination problems caused by slot auctions at multiple airports. However,
competitive concerns may lessen the willingness of an airline to sell a slot to a
competitor.
Lessons from Spectrum Auctions
The Federal Communications Commission began auctioning licenses to use portions of
the wireless spectrum in 1994 using a novel auction approach, the simultaneous multiple
round (SMR) auction. This auction format had not been implemented previously outside
of experimental economics laboratories.13 The auction rules used by the FCC have
evolved over time; both in response to perceived problems with previous auctions, and to
account for varying demand characteristics for the resource being sold. Over the years,
the auction rules have been adjusted based on concerns about defaults on installment
payments, bid signaling, and slow auction speed.14 Most recently, the FCC has been
exploring the use of package bidding to increase the efficiency of auctions in which
bidders have strong complementarities between licenses.15 Given the strong
11 Since scheduled flights between any two major New York airports (LaGuardia, JFK, and Newark) are
unlikely, the implementation of slot auctions at these three airports will not be as problematic as the
implementation of slot auctions at airports with flights scheduled to and from LaGuardia.
12 Of course, irrespective of how slots may be allocated, any use of slots to control congestion at multiple
airports may impose network effects because of the need to coordinate take-off and landing times.
13 See Kwerel, E.R. and Rosston, G.L., “An Insiders’ View of FCC Spectrum Auctions.” Journal of
Regulatory Economics. 17:3. 253-289, 2000.
14 Ibid at 275.
15 See Connolly,. M. and Kwerel, E., “Economics at the Federal Communications Commission: 2006-
2007.” Review of Industrial Organization. 31:107-120, 2007.
8

complementarities between landing slots and take-off slots at other airports, FAA would
be well advised to keep this in mind should the auction mechanism be expanded to
airports other than LaGuardia. It is possible that running a sequence of separate auctions
for slots at individual airports could result in inefficient allocations.
Results from auctions for third generation wireless licenses across Europe in 2000 and
2001 also demonstrated that the performance of auctions can depend crucially on the
rules of the auction and the characteristics of the pool of potential bidders. By most
accounts, the first such auction which took place in the UK was quite successful, where
success can be measured by the extent of participation or the revenue per capita generated
by the auction. Italy, the Netherlands, and Switzerland all roughly stuck with the UK
auction design, and raised much less revenue per capita. Germany and Austria modified
the rules, but the results were mixed; Germany’s auction resulted in high per-capita
revenues, but Austria did not.16 Obviously, these countries are different in many
respects, including per-capita income and number of incumbent mobile providers, but the
varying results seem likely to be a signal that careful attention needs to be paid to auction
design, as has been argued by many economists.17
Auctions and Congestion Pricing
In addition to the use of slots and slot auctions, congestion pricing is another mechanism
capable of controlling congestion and inducing an efficient use of available runway
resources. The FAA has recently considered implementing congestion pricing.18 Both
congestion pricing and auctions have the potential to solve congestion more efficiently
than administratively assigned slots, and both are more flexible in terms of market
structure and are more likely to lead to a more competitive marketplace than historical-
usage based slots. Under ideal, perhaps unrealistic, conditions, in which airport
authorities and airlines had perfect information, the equilibrium prices and allocations in
an efficient auction and in well executed congestion pricing would be identical. Under
more realistic conditions that are less than ideal, each system has advantages and merits
consideration.
Auctions have several advantages over congestion pricing due to the way they handle
uncertain and changing information. Auctions require that the correct quantity of slots be
16 See Klemperer. P., “How (Not) to Run Auctions: the European 3G Telecom Auctions.” European
Economic Review 2002, 46, 829-845 and van Damme, E., “The European UMTS-auctions.” European
Economic Review 2002, 46, 846-858.
17 Ibid.
18 Department of Transportation, Policy Regarding Airport Rates and Charges, Docket No. FAA-2008-
0036, 73 FED. REG. 3310-3316 (Jan. 17, 2008). In addition on August 21, 2001, the FAA issued a request
for public comment on several mechanisms to address congestion. The mechanisms considered by the
FAA were slot auctions, congestion pricing, peak period pricing, and flat fees. See “Notice of Market-
based Actions to Relieve Airport Congestion and Delay,” OST-2001-9849. More recently, the FAA
proposed giving airport operators more latitude to let landing fees respond to congestion, “Policy
Regarding Airport Rates and Charges”, Docket No. FAA-2008-0036, RIN2120-AF90.
9

set whereas congestion pricing requires that the correct price be set. It is more reasonable
that airport authorities know the correct quantity of slots than the correct congestion
price. As market conditions change, the correct quantity of slots will not vary but the
correct congestion price will, so any difficulties with determining the correct price will be
repeated, while in any additional auctions the bidders will take the new conditions into
consideration.
Congestion pricing however offers several advantages over auctions with regard to
efficiency and competition. Congestion pricing may have an advantage in producing an
efficient allocation of slots because it allows air carriers to adjust their schedule more
frequently. Any slot allocation, either by administratively assigning slots, auctioning
slots, or by assigning slots by some other means will be relatively inflexible compared to
congestion pricing due to the increased difficulty of schedule adjustments. The increased
flexibility afforded by congestion pricing may increase efficiency by allowing airlines to
more easily change their schedules to reflect changing market conditions.
Congestion pricing may also have an advantage with regard to competition. First,
congestion pricing may lead to a more competitive market because its flexibility is more
conducive to entry. When congestion is controlled by the use of slots, entry can only
occur if capacity is increased or an administrative or market-based reallocation of slots is
undertaken, such as when an auction is conducted. Moreover, with the use of slots,
potential entry is limited by the degree to which capacity is increased or by the number of
slots being reallocated or auctioned. Under congestion pricing, an entrant is able to enter
at any time or scale. Given that entry often requires experimentation by entrants, the
scheduling flexibility of congestion pricing seems more favorable to entry. Second, slot-
based congestion controls limit the competitive response of those airlines that do have
slots. Existing competitors can only gain a limited number of customers through price
competition because they too have a limited number of slots. Further, congestion pricing
eliminates the notion of property rights to slots – a notion that allows incumbent airlines a
measure of control over entry, while auctions serve to continue it. However, we should
note that absence of property rights creates the possibility of a tragedy of the commons,
wherein excessive operations degrade the quality of service of all airlines due to some
continued congestion. It is unclear whether the net impact of the potential increased
competition and the possibility of continued congestion is positive or negative for
consumers.
10

Conclusion
We find the proposal to be a reasonable, measured introduction of auctions to improve
the efficiency of congestion management and to gain experience using auctions for
allocating slots. The proposal does not provide great detail about the auction design, and
as the spectrum auctions demonstrate, tailoring the rules to fit the application is crucial;
the airline industry is sufficiently unique, complex, and important to warrant auction
designs specifically tailored to it. We think it is important that any candidate auction
designs be thoroughly analyzed and tested, perhaps in an experimental setting. In light of
the tradeoffs between auctions and congestion pricing, it is likely socially beneficial for
both solutions to be given a trial. We therefore support the current proposal to consider
the implementation of auctions, but also encourage consideration of a congestion pricing
test.
Respectfully submitted,
Michael R. Baye
Director Bureau of Economics
Federal Trade Commission
11