It’s our April break and we thought we’d do a cruise in addition to my brother’s wedding. Instead, here we are sitting on the floor in Logan.
— Stranded Logan Airport Passenger
All the objections that have been raised have been answered. The goal of the runway is to reduce the windrelated delays. I would be very surprised if we build this runway and it does not reduce wind-related delays. The science is telling us it absolutely will.
— Former Massachusetts Governor Paul Cellucci
That runway will only postpone 20 percent of delays until Logan fills up again in a matter of time. Meanwhile, the neighborhoods of Boston will be smothered in noise. — Boston Mayor Thomas Menino
…even according to [Massport’s] own environmental impact statement, under one scenario, Peak Period Pricing does more than the new strip to reduce projected delays. And simply by using the self-sorting logic of the market. In short, it’s a perfect conservative solution.
— The Boston Globe1
A recent study by the Airline Transport Association, an airline industry group, deemed U.S. airports to be “Approaching Gridlock.” In 2000, more than one in four flights—27%—were delayed, canceled or diverted, affecting approximately 163 million passengers. 2 And the problem is likely to get worse: The total number of yearly flights in the U.S. is projected to rise from almost 26 million in 2000 to 36 million in 2012.3 Consumer complaints to the Department of Transportation (DOT) quadrupled from 1995 to 2000 (from 6,000 to 23,000),4 with delays and missed connections being the most common source of frustration, comprising 40% of all complaints.5 Legislators have not been silent on the issue; after a recent report issued by the DOT detailed the rise in customer service problems, several bills were brought before Congress to address the issue, ranging from a bill making commitments made to customers enforceable by law to the passengers’ bill of rights legislation.6 With an average of 47.5 delays per 1,000 flights, Boston’s Logan Airport was ranked in 2001 as the fifth most significantly delayed airport in the country.7 Demand for the use of Logan was projected to increase from 479,000 annual operations (i.e., arrivals plus departures) in 2001 to anywhere from 510,000 to 656,000 annual operations in 2015, putting further pressure on local planning officials to find solutions for delays.
Doctoral Student George Batta prepared this case under the supervision of Associate Professor V.G. Narayanan. This case was developed from published sources. Special thanks are extended to Massport for its data. HBS cases are developed solely as the basis for class discussion. Cases are not intended to serve as endorsements, sources of primary data, or illustrations of effective or ineffective management. Copyright © 2001 President and Fellows of Harvard College. To order copies or request permission to reproduce materials, call 1-800-545-7685, write Harvard Business School Publishing, Boston, MA 02163, or go to http://www.hbsp.harvard.edu. No part of this publication may be reproduced, stored in a retrieval system, used in a spreadsheet, or transmitted in any form or by any means—electronic, mechanical, photocopying, recording, or otherwise—without the permission of Harvard Business School.
Purchased by: Khalid abuhassan [email protected] on February 27, 2014
Delays at Logan Airport
History of Logan Airport
Construction of an airfield at Jeffries Point, East Boston, was completed in September 1923, for use primarily by the Massachusetts Air Guard and the Army Air Corp. Named after Lt. General Edward Lawrence Logan, Boston’s airport ran its first commercial flight, between Boston and New York, in 1927. By the late 1930’s, demand for air travel grew to the point that American Airlines offered daily service between Boston and New York. Over the next few decades, more than 2000 acres were reclaimed from Boston Harbor in order to accommodate new runways and new hangars, and by the th
late 1950’s, Logan was the 10 busiest airport in the nation. In 1956, the state legislature created Massachusetts Port Authority (Massport) to operate and manage the airport and Boston’s port facilities, and an international terminal was completed in 1965 as the world moved into the jet age. Additional terminals, control towers, and runway extensions were completed over the years to accommodate the rapidly growing use of Logan.9 In response to Logan’s growth—which reached 27 million passengers, 900 million pounds of cargo, and 16,000 workers in 2000—Massport had planned a $1 billion Modernization Program, which involved renovating terminals, runways, and air traffic control facilities to better accommodate further expected increases in passengers and cargo and to help mitigate delay problems that have plagued Logan in recent years.
Causes of Delays
The primary cause of delays at Logan was adverse weather conditions. Around 70 to 75% of delays nationwide were weather-related,11 and Boston’s generally harsh climate made it particularly susceptible to weather delays; in fact, in adverse weather conditions, the percentage of aircraft delayed at Logan jumped from around 5% to 12%.12 Under normal circumstances, Logan airport could accommodate 118–126 operations an hour; under adverse weather conditions, however, capacity usually dropped down to 78–88 operations per hour.13 Delays generally occurred when adverse weather conditions required the use of runway configurations using less than three runways (see Exhibits 1–3). Fog and snow were often to blame for delays, but a more serious problem was strong northwest winds, which usually occurred during winter or after northeastern storm fronts blew through the area. Given Logan’s runway configuration, when northwest winds were moderately strong, only two runways could remain in operation; when northwest winds were severe, only one runway remained in operation (see diagrams).
In this latter case, total operational capacity dropped down to 40–60 operations per hour. Poor weather also required increased separation distances between aircraft, which also contributed to delays.14 Exacerbating the weather problems was Logan’s peculiar mix of airplanes. Forty percent of runway use at Logan airport was from small, non-jet airplanes, carrying between nine and thirty-four passengers15. Small craft held fewer passengers, flew more slowly, and had to maintain greater distances from the wind vortexes thrown off by larger craft.16 However, trends in U.S. aviation slowly mitigated this problem, as small, 9- to 19-seat turboprop planes were continually replaced by regional jet planes, which were 35- to 70-seat planes designed for flights flying between 150 and 800 miles.17 Regional jets composed 16% to 19% of Logan’s airplane mix in 2000, a proportion that was expected to grow over the next 15 years.
Overscheduling was another potential problem. If demand for runway use came close to normal capacity levels, even under good weather conditions, it resulted in delays. Overscheduling had generally not been a problem for Logan in recent years. With an increase in annual passengers from 27.4 million in 2000 to 37.5 million projected in 2015, however, overscheduling could prove to be a problem in the not-so-distant future. Another problem was Boston’s location as a far northern coastal city. As such, flights tended to originate and end at Logan rather than transition there to other destinations, which—among other things—meant that flights tended to arrive more uniformly over each hour rather than cluster in hourly “lumps” like in transitional, hub-type airports. Consequently, there was very little slack in the system for the airport to recover from delays. As a recent report by the Federal Reserve Bank of Boston had put it:19
Logan is unique in combining high levels of international, New England commuter, major domestic carrier, and cargo traffic. And almost 90 percent of Logan passengers start or end their journey there, making it one of the largest origin and destination markets in the country, a profile that tends to use airport services such as parking and customs more heavily than a hub-dominated airport. . . . Should arrival delays (the dominant kind) build due to fog, a hub airport stands a good chance of recovering to a normal schedule in the next hour. Logan has a different pattern, being at the end of many different flight streams. Planes arrive and depart continuously, and half of Logan’s takeoff and landing operations are small commuter and general aviation aircraft, almost twice the level of most major airports.
Solutions to Delays
Delays at Logan had been mitigated through the increased passenger utilization of regional airports. T.F. Green Airport in Providence and Manchester Airport together slowed the growth in utilization of Logan (see Exhibits 4–6). These airports attracted a large share of the short- to mediumhaul market,a and in doing so, not only recaptured passengers generated in their own areas but also absorbed demand from the periphery of Logan’s core service area20. From 1996 to 1999, only 27 percent of the region’s passenger growth occurred at Logan Airport and 73% occurred at the regional airports.21 Along with projected use of new high-speed rail lines—which may eventually allow commuters to travel from Boston to New York in two hours, from three and a half hours at present — regional airports will allow demand for Logan to reach 37.5 million annual passengers by 2015, rather than 2010 as originally projected.22 Massport has also attempted to grow the use of Worcester airport by encouraging major airlines to use it as a layover point en route to New York, Atlanta, and Orlando, in lieu of Logan.23
A New Runway
One measure advocated by Massport, former Massachusetts Governor Cellucci, and then Acting Governor Swift, was to expand capacity at Logan by building a new runway. According to a statutory environmental impact report (EIR) filed by Massport and the Federal Aviation Administration (FAA), Unidirectional Runway 14/32 would be located along the southwestern edge of the airport, and all arrivals and departures would fly southeast over Boston Harbor. Massport intended the runway to be used mainly by smaller planes, thus diverting larger planes to existing runways. The main motivation for the new runway was to increase Logan’s capacity during adverse weather conditions, ensuring that at least two runways would be in operation at all times. According to Massport officials, the new runway would cost about $100 million to build.
However, several community groups rallied behind Boston Mayor Thomas Menino to speak out against the new runway. Communities in South Boston felt that the runway would hamper the prospects for a waterfront development district in that area. More than $20 billion of public money had been committed to developing a 1,000-acre stretch of land for both residential and commercial use, and hopes remained high that the development would be a catalyst for urban renaissance in the area.24 Critics believed that the new runway, by diverting larger planes to an older runway that requires flying directly over the waterfront district, would deaden the attractiveness of the area for both residential and commercial users. In fact, the Massport/FAA EIR concluded that an area including the South Boston waterfront would likely face noise levels up to 65 decibels in magnitude if the new runway were in operation.
Though regional authorities promised to pay for soundproofing of homes in the area to soften the noise generated by the new runway, community groups were still opposed. As one group put it, “. . . the success of the South Boston Seaport District will be in its ability to evolve as a mixed-use district for living and working – not as a pedestrian-unfriendly environment with planes roaring 500–600 feet overhead.”25 Community activists also viewed the runway as a back-door means of increasing good weather capacity at the airport. Massport made claims to the contrary, stating that: !
The relatively low average passenger load factors (i.e., percentage of airplane seats filled) experienced by airlines that utilize Logan is proof that flight schedules are not being significantly depressed due to delays. As a result, there will likely be little demand for “extra” good weather capacity at Logan.
With the new runway, delays will only be mitigated during unpredictable weather conditions. As a result, airlines will not be able to anticipate—and thus adjust their flight schedules for— times when delays currently experienced will be mitigated by the new runway.26
Foes of the runway also got more fodder for their fight when a study completed in May 2001 by an independent aviation consultancy (funded, in fact, by Massport, at the request of the Community Advisory Group, an anti-runway group representing 27 Logan-area communities) claimed that most regional jets, which were making up an ever larger proportion of Logan’s landing mix, would usually be incapable of using the runway, thus mitigating its potential effectiveness in reducing delays. The report also criticized the EIR authors for employing unreasonable assumptions in generating runway capacity numbers for Logan. Whereas the EIR projected 118–126 operations per hour under good weather conditions, the consultancy’s report claimed that capacity really hovered around 98 operations per hour.27
The FAA, which had to approve the expansion, was expected to make a final decision on the new runway in the fall of 2001. In addition, Massport attempted to lift a 1976 court injunction against the project; in response, mayor Mayor Menino had hired a top Boston law firm to thwart Massport’s attempts.28 Community groups and Menino had instead pushed for the introduction of demand management techniques as an alternative to Logan expansion.
Peak-period pricing was another possible solution. Utilized when capacity constraints arise in some form and periods of high usage follow clear patterns, peak-period pricing involved charging users higher rates during periods of high capacity utilization in order to smooth runway demand over the course of the day. Peak-period pricing had been used in setting electricity and phone rates, highway toll pricing, and by airlines themselves when allocating airplane seats. In its truest form, airport peak-period pricing involved three things: access was not to be controlled by slots or any other type of administrative rationing mechanism; the charge was relative to the amount of runway demand in each period; and charges were applied to both arrivals and departures.
No U.S. airport had yet implemented peak-period pricing in its purest form, although three New York area airports (LaGuardia, Newark, and John F. Kennedy) had imposed minimum peak-period landing fees since 1968. Pure peak-period pricing was successfully implemented at Sydney, Australia’s Kingsford Smith Airport in January, 1991. During two peak periods (8 a.m. to 9 a.m. and 6 p.m. to 7 p.m.), fixed landing fees jumped 780%, and during the four “shoulder” hours occurring before and after the peak hours, fees jumped 700%.30 This policy had a very dramatic effect, resulting in a 28% reduction in peak-period flights, while actually increasing passenger seating capacity31 (see Exhibit 7).
Logan did implement a form of demand management for a short period in 1988, whereby all planes incurred a flat fee of $91, in addition to a weight-based fee of $0.45 per thousand pounds.b The plan significantly discouraged the use of smaller aircraft, since fixed landing fees were spread over smaller seating capacity. After implementing the new pricing scheme, Logan jumped from being the 12 best U.S. airport for on time performance to the second best, although industry consolidation in the regional carrier industry also played an important part in effecting this result.32 Logan was forced to discontinue the program after only six months, however, when the DOT deemed the pricing scheme discriminatory against airlines that primarily used smaller aircraft. The DOT did, however, claim that peak-period pricing would be an acceptable demand management tool, since it was not solely aimed at diverting smaller planes out of Logan.
Peak-period pricing at Logan would likely entail a $150 fixed landing fee during the consistently busiest hours of the day at Logan. Massport, however, contended that peak-period pricing would not represent an effective demand management tool, since Logan’s operations did not generally come near to capacity (see Exhibits 8 and 9). Massport did proffer that as the use of Logan grew over the next 15 years, peak-period pricing may very well represent a solution to delay problems (see Exhibit 10). In fact, under scenarios where Logan utilization grew to 45 million users and the proportion of non-jet aircraft retained its current character, the peak period could stretch from 7 a.m. to 11 p.m., encompassing 88.7% of weekly operations.
Regional carriers and Boston’s satellite communities spoke out against the use of demand management. Regional carriers (such as Cape Air and New England Air) predicted major job losses from the proposal, as higher landing fees caused a rise in ticket prices and forced them to restrict or even cancel operations. Communities in Cape Cod and other parts of New England also feared that the resulting loss of air transportation would cause hardship for residents who relied on frequent, affordable access to Boston for business, medical care, and other purposes.34 In fact, the authors of the Massport /FAA EIR projected substantial flight cancellations for airlines that primarily used 9- to 19-seat aircraft.
In response, Massport floated a proposal to exempt 16 New England communities from some peak-period fees, whereby any planes arriving or departing from airports in communities like Bangor, Maine; Manchester, New Hampshire; or Nantucket, Massachusetts would escape any levies for two roundtrip flights during each peak period. However, since many of the smaller planes that were responsible for congestion at Logan airport originated from these surrounding communities, critics claimed that allowing this exemption would attenuate the usefulness of peakperiod pricing as a demand management tool.