Understanding Airline and Passenger Choice in Multi-Airport Regions (2013)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

73 The research team reviewed articles from journals and on search engines including the following: • Academic and industry journals in airport (and transpor- tation) planning, management, and economics and • Search engines such as EBSCO HOST, Google Scholar, Academic Search Premier, and SciVerse ScienceDirect. Key search words consisted of variations of the following word combinations: airport choice, airport selection, air passenger choice, air passenger behavior, airline choice of airport, airport choice factors, multi-airport regions, pas- senger airport choice, traveler choice of airport, air passen- ger preferences, airport choice in a multiple-airport region, and airport competition, among others. The team’s focus was on articles published after 2005, unless a previously published study was of particular impor- tance. Articles on passenger choice factors were given prece- dence over airline choice articles due to the complexity of passenger choice factors in multi-airport regions. Unrelated to either airline or passenger choice factors, but also poten- tially of interest, were statistical models related to analysis of choice factors. Consequently, the team included articles that discuss the appropriateness of statistical models for analyzing airport choice factors. Studies with data collection from both international and U.S. locations were examined. Consistent with expectations, the study team found that the passenger choice factors of airfare, flight frequencies, and access time are most often studied. Choice factors determined by passenger type or purpose of travel were also topics of interest to researchers. Concerning frequently studied loca- tions, data are used from the San Francisco Bay area in mul- tiple reports, with other studies using passenger data collected from international locations. This appendix provides a review of relevant literature identified by the study team. Adler, N.; Berechman, J. (2001). Measuring airport quality from the airlines’ viewpoint: An application of data envelopment analysis. Journal of Transport Policy, 8 (3), 171–181. Abstract: The main objective of this paper is to develop a model to determine the relative efficiency and qual- ity of airports. This factor seems to have a strong effect on the airlines’ choice of hubs. Previous studies of airport quality have used subjective passenger data whereas in this study, airport quality is defined from the airlines’ viewpoint. Accordingly, the researchers solicited airlines’ evaluations of several European and non-European airports by means of a detailed ques- tionnaire. Statistical analysis of the median score has shown that these evaluations vary considerably, relative to quality factors and airports. The key method used in this study to determine the relative quality level of the airports is data envelopment analysis (DEA), which has been adapted through the use of principle component analysis. Of the set of West-European airports analyzed, Geneva, Milan, and Munich received uniformly high, relative efficiency scores. In contrast, Charles de Gaulle, Athens, and Manchester consistently appear low in the rankings. Albers, S., et al. (2005). Strategic alliances between airlines and airports—Theoretical assessment and practical evidence. Journal of Air Transport Management, 11 (2), 49–58. Abstract: Strategic alliances are now widespread. This paper shifts the focus from alliances among airlines toward strategic alliances involving passenger airlines and airports. Following a conceptual path analyzing motives, potential benefits and problems, potential fields of cooperation are identified along with three basic classes of airline–airport alliances. Capacity-based, marketing-based, and security based cooperation models A P P E N D I X B Literature Review

74 are assessed with regard to benefits for the participating airline and airport partners. This expands the existing literature that has largely neglected the airline–airport relationship and its potential for developing their respective competitive strategies. The case of the alli- ance between Lufthansa and Munich airport serves as an illustration. Alder, T., et al. (2005). Modeling service trade-offs in air itinerary choices. Transportation Research Record 1915: Journal of the Transportation Research Board, 20–26. Abstract: The application of a mixed logit approach using stated-preference survey data to the development of itinerary choice models is described. The models include the effects on itinerary choices of airline, airport, aircraft type, fare, access time, flight time, scheduled arrival time, and on-time performance. The empiri- cal results demonstrate the importance of explicitly accounting for traveler preference heterogeneities by using segmentation by trip purpose, interaction effects involving frequent flyer status, and random parameter specifications. Explicitly including preference hetero- geneity by using the mixed logit specification results in significant statistical improvements and important coefficient differences as compared with using a stan- dard fixed-parameter logit model. The calculated mar- ginal rates of substitution show the relative importance that travelers assign to key service variations among itineraries. All service features that were included in the model had significant values to travelers, and the values were affected, as would be expected, by the traveler’s frequent flyer status. Although current reservation and ticketing services provide information to prospec- tive travelers on most of these itinerary features, most services do not report on-time performance, which, however, can be an important selection criterion for travelers. Barrett, S. D. (2004). How do the demands for airport services differ between full-service carriers and low-cost carriers? Journal of Air Transport Management, 10 (1), 33–39. Abstract: There has been a considerable increase in the share of air traffic within Europe that is carried by LCCs. This paper explores the nature of the demand function for the services of such carriers and contrasts it to that of the more traditional European airlines. It pays particular attention to the links that airlines have with airports and how that will need to change in the future with the growth of the LCCs. Basso, L. J.; Zhang, A. (2008). On the relationship between airport pricing models. Transportation Research Part B: Meth­ odological, 42 (9), 725–735. Abstract: Airport pricing papers can be divided into two approaches. In the traditional approach the demand for airport services depends on airport charges and on congestion costs of both passengers and airlines; the airline market is not formally modeled. In the vertical- structure approach instead, airports provide an input for an airline oligopoly and it is the equilibrium of this downstream market that determines the airports’ demand. The study proves, analytically, that the tradi- tional approach to airport pricing is valid if air carriers have no market power, i.e., airlines are atomistic or they behave as price takers (perfect competition) and have constant marginal operational costs. When carriers have market power, this approach may result in a surplus mea- sure that falls short of giving a true measure of social surplus. Furthermore, its use prescribes a traffic level that is, for given capacity, smaller than the socially opti- mal level. When carriers have market power and conse- quently both airports and airlines behave strategically, a vertical-structure approach appears a more reasonable approach to airport pricing issues. Bazargan, M.; Vasigh, B. (2003). Size versus efficiency: A case study of U.S. commercial airports. Journal of Air Transport Management, 9 (3), 187–193. Abstract: The paper presents a productivity analy- sis using data envelopment analysis (DEA) of 45 U.S. commercial airports selected from the top 15 large, medium, and small airports. Financial and operational data, such as aircraft movements, number of airport gates, the annual number of enplaned passengers and runway capacity, are used. Initially, a DEA is deployed to analyze the efficiency and performance measures of airports within each group by comparing and cross- referencing them with each other. The analysis is then extended to identify those airports that are not efficient and are thus dominated by other airports that are more efficient. Blackstone, E., et al. (2006). Determinants of airport choice in a multi-airport region. Atlantic Economic Journal, 34 (3), 313–326. Abstract: The Civil Aeronautics Board was disman- tled on the premise that competition and the threat of entry would restrain airline prices. If consumers do not search for low fares, then the threat of entry will have

75 little impact. The entry of a low-fare carrier will real- locate flyers within but not between airports. Telephone survey data were used to estimate probit models for the use of BWI, Newark International, JFK International, and Philadelphia International Airports to evaluate the effect of low fares on consumer behavior. In airport usage, age and gender do not matter. Although survey participants reported that airfare is an important con- sideration, actual searching for a low fare was unimport- ant. The availability of nonstop flights, wait at check-in, income, and distance from home were important. Carlsson, F.; Lofgren, A. (2006). Airline choice, switching costs and frequent flyer programmes. Applied Economics, 38 (8), 1469–1475. Abstract: Switching costs are costs that customers face when switching from one firm to another. In markets such as the airline market where repeated purchases are common, switching costs may be substantial. In this paper, the switching costs are estimated for domestic airline routes in Sweden between 1992 and 2002. In addition, the determinants of these switching costs are tested for, in particular, to what extent factors such as frequent flyer programs and flag carriers have an effect on switching costs. A substantial switching cost is found. Although a large part of this calculated switching cost can be attributed to perceived quality differences, it is also found that frequent flyer programs contribute a non-negligible part of the switching cost. The paper ends with a brief discussion on the welfare consequences of switching costs, where the connection between habit formation and switching costs is discussed. Ciliberto, F.; Williams, J. W. (2009). Limited access to airport facilities and market power in the airline industry. Journal of Law and Economics. Available at SSRN: http://ssrn.com/ abstract975955. Abstract: This paper investigates the role of limited access to airport facilities as a determinant of the hub premium in the U.S. airline industry. The researchers used original data from competition plans that airports are required to submit to the Department of Transpor- tation in compliance with the Aviation Investment and Reform Act for the 21st Century. Information on the availability and control of airport gates, leasing arrange- ments, and other restrictions limiting the expansion of airport facilities was collected. The paper finds that the hub premium is increasing in the ticket fare, and that control of gates is a crucial determinant of this premium. Limits on the fees that airlines can charge for subleasing their gates lower the prices charged by airlines. Finally, control of gates and restrictions on sublease fees explain high fares only when there is a scarcity of gates relative to the number of departures out of an airport. Derudder, B., et al. (2010). A spatial analysis of multiple air- port cities. Journal of Transport Geography, 18 (3), 345–353. Abstract: This paper presents a detailed empirical description of airport connectivities in four major multiple-airport cities (London, New York, Los Angeles, and San Francisco). The analysis draws on data derived from a previously largely untapped information source, the so-called Marketing Information Data Transfer (MIDT). This dataset contains information on actually flown transnational routes, which allows for a thorough assessment of the chief connectivity characteristics of specific airports. Combined with information derived from several other sources, our results point to func- tional divisions among airports, both in terms of their geographical scale (e.g., national, regional, and inter- national airports) and their specific role in the airline network (e.g., origin/destination versus hub airports). The implications of the results are discussed, and some avenues for future research are considered. Dresner, M. (2006). Leisure versus business passengers: Simi- larities, differences, and implications. Journal of Air Transport Management, 12 (1), 28–32. Abstract: As low-cost air carriers increase their mar- ket share, the percentage of leisure to total passengers will increase. Data from an airport passenger survey are analyzed to document differences and similarities between leisure and business passengers. Surprisingly, the two groups of passengers are quite similar in terms of their reasons for choosing to fly from the airport sur- veyed, their parking requirements, and the number of bags they checked. These similarities indicate that air- line and airport managers may not be obliged to make significant adjustments to their operations to account for the changing passenger mix. Fuellhart, K. (2007). Airport catchment and leakage in a multi-airport region: The case of Harrisburg International. Journal of Transport Geography, 15 (4), 231–244. Abstract: This paper presents a spatial analysis of the market area of Harrisburg International Airport (MDT) in south-central Pennsylvania using a zipcode-level

76 spatial database of a sample of airport customers. Using a geographic information system (GIS) and regres- sion techniques, a description of MDT’s market area is presented in relation to various demographic and geographic variables. The results show clear patterns of possible airport substitution—particularly between MDT and BWI. These results are corroborated with a simple route-level regression analysis showing relative passenger levels at MDT versus BWI in relation to fare differences and other factors. Fuellhart, K. (2003). Inter-metropolitan airport substitu- tion by consumers in an asymmetrical airfare environment: Harrisburg, Philadelphia and Baltimore. Journal of Transport Geography, 11 (4), 285–296. Abstract: Airfares vary significantly over space, and can even vary substantially between airports in relatively close proximity to one another. With the spread of vari- ous Web tools, consumers are armed with more infor- mation than ever to assess fare and service differences between competing airlines and competing airports. This leads to the possibility of airport substitution for particular routes. Linear regression models are devel- oped that suggest, despite the 70 to 90 mile distance, that passenger substitution may be occurring from Harrisburg and Philadelphia to Baltimore based as a result of differential fares, low-fare service, and other factors. Gelhausen, M. C. (2011). Modelling the effects of capacity constraints on air travellers’ airport choice. Journal of Air Transport Management, 17 (2), 116–119. Abstract: This paper analyzes the effects of limited capacity on air travelers’ airport choice. The analysis is based on a market segment specific airport choice model that accounts for limited capacities. The region of Stuttgart in Germany serves as a case study to exam- ine the impact of limited airport capacity on air travel- ers’ airport choice. Air travelers’ choice depends on the supply of flights and accessibility of the airports in their choice set as well as on their preferences and willing- ness to pay. To elaborate the effects of limited airport capacity, scenarios relating to the capacity situation at airports serving the air travel demand of the Stutt- gart region are analyzed. This paper reveals the mutual dependence among airports. Capacity constraints at one airport cause spill-over effects and thus influence air travel demand served at other airports. In some cases this may even lead to new capacity constraints elsewhere. Gelhausen, M. C. (2009). The influence of limited airport capacity on passengers’ airport choice in a decentralised airport environment. Journal of Airport Management, 3 (4), 366–383. Abstract: This paper examines the impact of limited airport capacity on the airport choice of individual air travelers. The quantitative analysis is based on a nested logit model, enhanced to allow for capacity constraints at airports and to improve model applicability in the real world. The paper starts by describing the main idea of the model and goes on to consider airport choice behavior in the Cologne region in a capacity-constrained decentralized airport environment. To elaborate the impact of limited airport capacity on passenger choice, three different scenarios are analyzed. In this manner, it is possible to illustrate the complex distributional changes in airport choice by market segment, trip ori- gin, and trip destination. This research aims to show the mutual dependence among airports operating in a decentralized environment in which some airports do not have the capacity to meet their full demand potential. In such an environment, capacity constraints at one airport may lead to spill-over effects and thus influence air travel demand served elsewhere. In some cases, this may even lead to new capacity constraints at these airports. Gillen, D.; Lall, A. (2004). Competitive advantage of low-cost carriers: Some implications for airports. Journal of Air Trans­ port Management, 10 (1), 41–50. Abstract: In this paper, the sources of competitive advantage of LCCs such as Southwest, Ryanair, and easyJet are identified. Many have looked to these carri- ers’ operational efficiency as their source of advantage, but the choice of business model with point-to-point service provides the strategic advantage and the opera- tional effectiveness complements this choice. The vertical relationships between processes are based on the simplic- ity of service. This leads to simplicity of processes and simplicity of organization. These points are illustrated with a discussion of Southwest and how it organizes its “turns” for flights. The team organization and the simplicity of information flows result in greater rela- tional coordination. This contrasts with airlines such as Ryanair that seek lower costs through lower prices. The paper argues that the Southwest model is not generic and duplication is difficult because of system coordi- nation, whereas the Ryanair model can be more easily duplicated. This results in first mover advantages for carriers such as Ryanair and their willingness to engage

77 in long-term contracting for key assets, such as airport access. These differences in achieving operational effi- ciency have different implications for airports, which include bargaining power and risk exposure. An airport with a dominant single LCC is subject to more risk and low bargaining power. Gözen, B.; Chandra, B. (2004). A parameterized consider- ation set model for airport choice: an application to the San Francisco Bay Area. Transportation Research Part B: Method­ ological, 38 (10), 889–904. Abstract: Airport choice is an important air-travel- related decision in multiple-airport regions. This paper proposes the use of a probabilistic choice set multi- nomial logit (PCMNL) model for airport choice that gen- eralizes the multinomial logit model used in all earlier airport choice studies. The paper discusses the proper- ties of the PCMNL model, and applies it to examine airport choice of business travelers residing in the San Francisco Bay area. Substantive policy implications of the results are discussed. Overall, the results indicate that it is important to analyze the choice (consideration) set formation of travelers. Failure to recognize consideration effects of air travelers can lead to biased model param- eters, misleading evaluation of the effects of policy action, and a diminished data fit. Gupta, S., et al. (2008). Air passenger preferences for choice of airport and ground access mode in the New York City metropolitan region. Transportation Research Record 2042: Journal of the Transportation Research Board, 3–11. Abstract: In current practice, regional models are limited in their capability to analyze policies involv- ing changes and improvements to airports (and their services) and ground access transportation. Typically, airports are treated only as employment centers or as special generators. Important and distinct features of air passenger travel affecting trip distribution and mode choice are rarely modeled explicitly. This paper presents the development of a joint airport and ground access mode choice model for the New York City met- ropolitan region based on an extensive survey of airport users. Unlike travel to and from most U.S. cities, air pas- sengers flying to and from the New York region face a nontrivial choice of airports and ground access modes (including premium transit options). A nested logit model was formulated with airport choice at the upper level and ground access mode choice at the second level; however, a multinomial logit model was found to be statistically preferable. Results indicate that air passen- ger travel behavior is significantly different for business and nonbusiness travelers. Overall, willingness to pay for trips to and from the airport is much higher than for regular intra-city trips. Average yield, access time, and access cost are the most important determinants of air passenger’s choice; demographics and trip characteris- tics are also significant. The developed tool was used for a comprehensive study of airport development alterna- tives in the New York region and is seen as the platform for additional data development and model extensions for future studies of air passenger service planning in the New York mega region. Hess, S. (2010). Evidence of passenger preferences for specific types of airports. Journal of Air Transport Management, 16 (4), 191–195. Abstract: Studies of air travel choice behavior increas- ingly make use of data collected through stated choice surveys. This paper puts forward the hypothesis that when making their choices in such surveys, respondents may complement the information presented to them by additional attributes. Specifically, the paper looks at characteristics linked to airport size and breadth of ser- vice, as well as the proximity to a respondent’s home. The findings in a discrete choice analysis suggest that, all else being equal, respondents prefer larger to smaller airports while having a preference for the airport clos- est to their home. This could suggest that even though respondents associate a higher likelihood of delay and other inconveniences with larger airports, there is a perception that if things go wrong (e.g., flight cancel- lations), the backup options at larger airports (e.g., replacement aircraft) are superior to those at small or regional ones. Hess, S. (2004). An analysis of airport-choice behaviour using the Mixed Multinomial Logit model. ERSA conference papers. European Regional Science Association. Abstract: This paper describes part of an ongoing study of airport choice for passengers departing from the San Francisco Bay area. The aim of the present paper is to test for the prevalence of taste heterogeneity across travelers, using the mixed multinomial logit (MMNL) model. Our results indicate the presence of significant levels of heterogeneity in tastes, especially with respect to the sensitivity to access time, characterized by sig- nificant (deterministic) variation between groups of travelers (business/leisure, residents/visitors) as well as random variation within groups of travelers. Our analysis reinforces earlier findings showing that business

78 travelers are far less sensitive to fare increases than lei- sure travelers, and are willing to pay a higher price for decreases in access time (and generally also increases in frequency) than is the case for leisure travelers. Finally, the results show that the random variation between business travelers in terms of sensitivity to access time is more pronounced than that between leisure travelers, as is the case for visitors when compared to residents. Hess, S.; Polak, J. W. (2006). Airport, airline and access mode choice in the San Francisco Bay area. Papers in Regional Sci­ ence, 85 (4), 543–567. Abstract: This article presents an analysis of air travel choice behavior in the San Francisco Bay Area. The analysis extends existing work by considering the simultaneous choice by passengers of a departure air- port, airline, and access mode. The analysis shows that several factors, most notably flight frequency and in- vehicle access time, have a significant overall impact on the attractiveness of an airport, airline, and access mode combination, while factors such as fare and aircraft size have a significant effect only in some of the popula- tion subgroups. The analysis highlights the need to use separate models for resident and non-resident travelers, and to segment the population by journey purpose. The analysis also shows that important gains can be made through the inclusion of airport-inertia variables, and through using a non-linear specification for the mar- ginal returns of increases in flight frequency. In terms of model structure, the results suggest that the use of the different possible two-level nested logit models leads to modest, yet significant, gains in model fit over the cor- responding multinomial logit models, which already exhibit very high levels of prediction performance. Hess, S.; Polak, J. W. (2006). Exploring the potential for cross- nesting structures in airport-choice analysis: A case study of the Greater London Area. Transportation Research Part E: Logistics and Transportation Review, 42 (2), 63–81. Abstract: The analysis of air passengers’ choices of departure airport in multi-airport regions is a cru- cial component of transportation planning in many large metropolitan areas, and has been the topic of an increasing number of studies over recent years. This paper advances the state of the art of modeling in this area of research by making use of a cross-nested logit (CNL) structure that allows for the joint representation of inter-alternative correlation along the three choice dimensions of airport, airline, and access mode. The analysis uses data collected in the Greater London Area, which arguably has the highest levels of inter-airport competition of any multi-airport region; the authors of this paper are not aware of any previous effort to jointly analyze the choice of airport, airline, and access mode in this area. The results of the analysis reveal signifi- cant influences on passenger behavior by access time, access cost, flight frequency, and flight time. A struc- tural comparison of the different models shows that the cross-nested structure offers significant improvements over simple nested logit (NL) models, which in turn outperform the multinomial logit (MNL) model used as the base model. Hess, S.; Polak J. W. (2005). Accounting for random taste heterogeneity in airport choice modeling. Transportation Research Record 1915: Journal of the Transportation Research Board, 36–43. Abstract: The findings from a disaggregate analysis of the choice of airport, airline, and access mode for busi- ness travelers living in the San Francisco Bay Area, Cal- ifornia, are presented. Aside from formulation of the multidimensional choice process, the main objective is to explore random taste heterogeneity among deci- sionmakers in their evaluation of the attractiveness of the different alternatives. The results indicate that this heterogeneity is present in tastes relating to in-vehicle access time, access cost, and flight frequency. Account- ing for this heterogeneity leads to gains in model fit but, more important, leads to important insights into the differences in behavior across decisionmakers and avoids the bias introduced into trade-offs when fixed coefficients are used in the presence of significant lev- els of heterogeneity. In terms of substantive results, the models also reveal a significant impact of changes in out-of-vehicle access time, indicate a preference for service on jet over turboprop flights, and show that experience plays an important role in air travel choice behavior. Hess, S., Polak J. W. (2005). Mixed logit modelling of airport choice in multi-airport regions. Journal of Air Transport Man­ agement, 11 (2), 59–68. Abstract: This paper presents an analysis of the choice of airport by air travelers departing from the San Fran- cisco Bay area. The analysis uses the mixed multinomial logit model, which allows for a random distribution of tastes across decisionmakers. To our knowledge, this is the first application using this model form in the analy- sis of airport choice. The results indicate that there is significant heterogeneity in tastes, especially with respect

79 to the sensitivity to access time, characterized by deter- ministic variations between groups of travelers (busi- ness/leisure, residents/visitors) as well as random variations within groups of travelers. The analysis rein- forces earlier findings showing that business travelers are far less sensitive to fare increases than leisure travel- ers, and are willing to pay a higher price for decreases in access time (and generally also increases in frequency) than is the case for leisure travelers. Finally, the results show that the random variation between business trav- elers in terms of sensitivity to access time is more pro- nounced than that between leisure travelers, as is the case for visitors when compared to residents. Hess, S., et al. (2007). Modelling airport and airline choice behaviour with the use of stated-preference survey data. Transportation Research Part E: Logistics and Transportation Review, 43 (3), 221–233. Abstract: The majority of studies of air travel choice behavior make use of revealed preference (RP) data, generally in the form of survey data collected from departing passengers. While the use of RP data has cer- tain methodological advantages over the use of stated- preference (SP) data, major issues arise because of the often low quality of the data relating to the unchosen alternatives, in terms of explanatory variables as well as availability. As such, studies using RP survey data often fail to recover a meaningful fare coefficient, and are gen- erally not able to offer a treatment of the effects of air- line allegiance. In this paper, we make use of SP data for airport and airline choice collected in the United States in 2001. The analysis retrieves significant effects relating to factors such as airfare, access time, flight time, and airline and airport allegiance, illustrating the advan- tages of SP data in this context. Additionally, the analysis explores the use of non-linear transforms of the explan- atory variables, as well as the treatment of continuous variations in choice behavior across respondents. Ishii, J., et al. (2009). Air travel choices in multi-airport mar- kets. Journal of Urban Economics, 65 (2), 216–227. Abstract: Study of how air travel consumers departing from a multi-airport region trade-off across airport and airline supply characteristics. Researchers empiri- cally investigate this trade-off by estimating a weighted conditional logit model of airport–airline choice, using survey data on travels departing from one of three San Francisco Bay Area airports and arriving at one of four airports in greater Los Angeles in October 1995. Non- price characteristics like airport access time, airport delay, flight frequency, the availability of particular airport–airline combinations, and early arrival times are found to strongly affect choice probabilities. The study calculates marginal effects and counterfactual scenarios to compare the values of these characteristics among each other and across traveler type. To exam- ine the robustness of the conditional logit model, the researchers estimate a mixed logit model and find that the results are similar. The researchers attribute the similarity to our strictly defined travel market and to our distinction between leisure and business travelers, thus controlling for two important sources of consumer heterogeneity. The paper considers the implications of empirical findings on vertical integration between air- lines and airports, on the effectiveness of “airport dom- inance,” and on the competitive effect of entry by LCCs. Jiang-tao, L. (2008). Airport choice in multi-airport regions: An empirical study for Chinese Metropolitan Area. Inter­ national Conference on Intelligent Computation Technology and Automation, 2, 329–332. Abstract: In this paper, a multinomial logit (MNL) model is constructed to predict airport choice in a multiple-air- port region and estimated using passenger data from a Chinese metropolitan area. Four explanatory variables were investigated, namely, access time to the airports of choice, airline service (mainly flight frequencies) at the regional airports, airfare, and a passenger’s experi- ence with an airport. In agreement with previous work, it was found that flight frequency is one of the signifi- cant predictors of airport choice. However, estimation results indicate that not access time but airfare is another important predictor in the competition between airports in a developing country’s region. Travelers in develop- ing countries have higher airfare elasticity than those in developed countries, while travelers in developed coun- tries have higher access time elasticity than those in devel- oping countries. In addition, a passenger’s experience is significant in the airport choice behavior in both devel- oped and developing countries. This would indicate that passengers who have used an airport will tend to continue to use the same airport, all other factors being equal. Lian, J. I.; Ronnevik, J. (2011). Airport competition— Regional airports losing ground to main airports. Journal of Transport Geography, 19 (1), 85–92. Abstract: Regional airports in Norway are losing mar- ket shares to nearby main airports on flights to the national capital, Oslo, and on international travel via Oslo. Travelers are willing to spend several hours extra

80 driving to a larger airport in order to take advantage of lower fares and more convenient airline services. Traffic leakage from regional airports is high when the service from the regional airport is indirect and fare differences are large. Public service obligation (PSO) tenders set maximum fares on the regional legs, but do not cover through travel from regional airports that involve com- mercial legs. Traffic leakage is particularly evident in the leisure segment. Leakage levels tend to increase as com- petition is intensified at main airports, but the evidence is rather mixed. Logistic curves of airport market shares have proven to be useful when comparing spatial varia- tions in leakage levels. Loo, B. P. Y. (2008). Passengers’ airport choice within multi- airport regions (MARs): some insights from a stated-preference survey at Hong Kong International Airport. Journal of Trans­ port Geography, 16 (2), 117–125. Abstract: Passengers’ airport choice in multi-airport regions (MARs) is of great interest to transport research- ers, local governments, airport authorities, and airline companies. This paper analyzes the airport choice of passengers departing from Hong Kong International Airport (HKIA) to 15 destinations in different parts of the world. The results, based on stated-preference (SP) data, show that airfare, access time, flight frequency and the number of airlines were the most important airport level of service (LOS) attributes. In contrast, the num- ber of airport access modes, access cost, airport shop- ping area, and queue time at check-in counters were not statistically significant. The segmentation analyses reveal important subtle variations in airport preferences among different market segments. The findings provide valuable insights on a less well-researched MAR—the Hong Kong-Pearl River Delta (HK-PRD) MAR. Luken, B. L.; Garrow, L. A. (2011). Multiairport choice mod- els for the New York Metropolitan Area : Application based on ticketing data. Transportation Research Record 2206: Jour­ nal of the Transportation Research Board, 24–31. Abstract: This study examines the potential to use online ticketing data to model airport choice for domestic flights originating in one of the three major airports located in the New York City area. Results indicate that airport accessibility and LOS influence airport choice. Results also suggest that capacity constraints—reflected in sold-out flights and higher fares—may lead to more switching across airports as the flight departure dates approach. This underscores the importance of incorpo- rating the actual flights available and the actual prices seen by consumers at the time that they ticket into multi-airport choice models. Marcucci, E.; Gatta, V. (2011). Regional airport choice: Con- sumer behaviour and policy implications. Journal of Trans­ port Geography, 19 (1), 70–84. Abstract: The analysis of origin airports in multi- airport regions has a well-established tradition in transportation and regional economics. The main goal of the paper is to estimate the importance of the differ- ent attributes that determine origin airport choice. In this paper, a stated-preference approach was adopted to study this problem and evaluate the effects of pos- sible policy interventions. A detailed segmentation of the sample studied according to the socioeconomic variables that prove statistically relevant when inter- acted with the attributes used to characterize airport choice was performed. Moreover, in order to test for the presence of heterogeneity in agents’ preferences the researchers estimate several mixed logit models with different specifications, including heteroscedasticity and error component. With respect to previous studies the researchers developed and extended the traditional SP approach by also analyzing the role and relevance of attribute cut-offs in this research field. The policy simulations produced are based on the estimation of airport-specific attributes. The study concentrates on a multi-airport region in central Italy where four com- peting airports are located. Matisziw, T. C.; Grubesic, T. H. (2010). Evaluating locational accessibility to the US air transportation system. Transporta­ tion Research Part A: Policy and Practice, 44 (9), 710–722. Abstract: Although there are hundreds of airports that support commercial air passenger traffic in the United States, not all areas are equivalently served by the commercial air transportation system. Locations in the United States differ with respect to their level of access to the commercial air network and their overall acces- sibility within the system. Given the complexity of the domestic commercial air passenger network and sup- porting infrastructure, past research has only been able to provide a limited assessment of locational accessibil- ity within the United States. To address these complexi- ties, this paper proposes a new metric that incorporates measures of access to air transport as well as accessibil- ity within air transportation networks. Using a com- prehensive dataset on scheduled airline service, the developed approach is then applied to the U.S. domes- tic commercial passenger air transportation network to

81 explore geographic differentials in accessibility. Results suggest marked differences between core-based statisti- cal areas throughout the United States. Nicole, H. (2004). The upside of using an inconvenient airport. The Wall Street Journal ­ Eastern Edition, 244 (5), D1–D4. Abstract: Reports that flyers are using smaller, less convenient airports to save time and money. Describes advantages of using non-major airports (such as cheaper tickets and avoiding long security delays) statistics on passenger volume at major airports, effect of discount airlines on the popularity of smaller airports, and chal- lenges of utilizing smaller airports. Pathomsiri, S.; Haghani, A. (2005). Taste variations in airport choice models. Transportation Research Record 1915: Journal of the Transportation Research Board, 27–35. Abstract: A mixed multinomial logit model for analyz- ing choice of departure airport in a multiple-airport system (MAS) is presented. The model aims to capture random taste variations across passengers in response to airport LOS through a set of random coefficients. A case study is carried out for the Baltimore, Maryland- Washington, D.C., MAS. The 1998 Air Passenger Sur- vey database is used to estimate the model. The results indicate significant taste variations in response to flight frequency and airline fare even within smaller segments by both trip purpose and residency status. Analyses of the model provide several insightful results, such as dis- tribution of perceived LOS and time value. In addition, the model is used to simulate the impact of interesting scenarios on market share. Substantial policy implica- tions for airport management are also provided. Pels, E., et al. (2009). Low-cost airlines and airport competi- tion. Transportation Research Part E: Logistics and Transporta­ tion Review, 45 (2), 335–344. Abstract: An important question from the viewpoint of competition analysis in the air transport industry is the extent to which low-cost airlines operating from a secondary airport compete with full-service airlines serving a main airport in a multiple-airport region. This paper addresses the issue of the competition between full-service and low-cost airlines serving adjacent air- ports in Greater London using econometric estimation of demand structure (own- and cross-price elasticities). The analysis follows the method in (Pels, E., Nijkamp, P., Rietveld, P., 2000. Airport and airline competition for passengers departing from a large metropolitan area, Journal of Urban Economics, 48 (1), 29–45; Pels, E., Nijkamp, P., Rietveld, P., 2003. Access to and competi- tion between airports: A case study for the San Francisco Bay area, Transportation Research Part A: Policy and Prac­ tice, 37 (1), 71–83). It is based on the nested logit model we use to capture three key dimensions of passenger choice: airfare, surface-access costs, and frequency. The researchers obtained estimates of the own- and cross- price elasticities, which was the focus of the researchers’ interest. On the basis of understanding of the industry dynamics the paper found these estimates, especially of the cross-price elasticities, to be on the low side. Pels, E., et al. (2001). Airport and airline choice in a multiple airport region: An empirical analysis for the San Francisco Bay Area. Taylor and Francis Journals, 35 (2), 1–9. Abstract: In this paper a nested logit model is used to describe passenger preferences concerning airports and airlines. A statistical model for the passengers’ sequen- tial choice of airport and airline is calibrated. It appears the nested multinomial logit model, with airports as the common elements in the nests, is statistically pref- erable to the standard multinomial logit model. Fre- quency and access time to the airport are all significant. Separate models are estimated for business and leisure travelers, but there appear to be only small divergences. Pels, E., et al. (1999). Airport and airline competition for pas- sengers departing from a large metropolitan area. Journal of Urban Economics, 48 (1), 29–45. Abstract: In this paper, an airport and airline choice model, based on a nested multinomial logit model, is developed to investigate both airport competition and airline competition in a metropolitan area with mul- tiple departure airports. The model can be used to analyze the effects of an improvement in accessibility of a specific airport in a metropolitan area. It is shown analytically that if the frequency elasticity of demand is smaller than 1, unique airfare-frequency and passenger- charge equilibria exist. Next, symmetric equilibria are derived analytically; their properties are also examined. Finally, asymmetric equilibria are derived numerically, while their properties are discussed as well. Redondi, R., et al. (2011). Hub competition and travel times in the world-wide airport network. Journal of Transport Geog­ raphy, 19 (6), 1260–1271. Abstract: The aim of this work is to measure the com- petition between airport hubs based on an analysis of

82 travel times in the world-wide airport network. By con- sidering the minimum travel time required to connect each pair of airports, it is possible to create new measures of hub competition, separating the effects of hub position and temporal coordination. This analysis was carried out at the global level, considering all 232 airports with more than 3 million seats yearly offered in departure flights in 2008, and also in relevant geographic markets. The results show a high level of competition among the most important world airports, but the major airports of Europe have a geographical advantage in relation to world markets over the major American and Asian air- ports. Also shown is that airports located on different continents often compete for the same origin–destination markets. Geographical position appears to be the most important variable explaining hub performance. Sec- ondary hubs show a higher degree of specialization toward specific markets. Ricondo & Associates Inc. (2011). Official Statement—City of San Jose Airport Revenue Bonds 2011A. Report of the Inde­ pendent Airport Consultant, City of San Jose. Abstract: Uses regional survey data from 2009 study of Bay Area residents conducted for the Regional Airport Planning Committee of the Association of Bay Area Governments to capture airport activity market share. Results indicate that passenger demand is influenced by flight availability, airfares, proximity to residence, airport accessibility, and airport reliability. Suzuki, Y. (2007). Modeling and testing the “two-step” deci- sion process of travelers in airport and airline choices. Trans­ portation Research Part E: Logistics and Transportation Review, 43 (1), 1–20. Abstract: This paper develops and estimates a nested logit model of airport–airline choice that incorporates the “two-step” decision process of air travelers. The model assumes that a traveler first eliminates certain choice alternatives that do not satisfy his/her minimum acceptable standards (first step), and then chooses the utility-maximizing alternative from the set of screened choice alternatives (second step). The model is cali- brated by using the survey data collected in the U.S. (Central Iowa). The results imply that the “two-step” choice model may fit the observed data significantly better than the conventional “one-step” choice models. Suzuki, Y., et al. (2003). Airport choice, leakage, and experi- ence in single-airport regions. Journal of Transportation Engi­ neering, 129 (2), 212–218. Abstract: Airport leakage refers to the phenomenon where travelers in small, single-airport regions avoid using their local (nearest) airports and prefer to use the more distant but larger metropolitan airports. Past studies of airport choice did not consider airport leak- age tendencies of air travelers in single-airport regions, nor considered how the choice probabilities of indi- vidual travelers are affected by the goodness or badness of the travelers’ experiences with one or more of the candidate airports. This paper extends the research on airport choice by considering the airport leakage ten- dencies of travelers in single-airport regions, and by incorporating variables that capture the individuals’ heterogeneity of airport experience. The results indi- cate that single-airport area travelers are more likely to leak to larger metropolitan airports when their trip purpose is leisure than when it is business, and that travelers are more likely to choose the airports in which they gained good experiences than those which they have never used or had bad experiences. Tam, M. L., et al. (2011). The impact of travel time reliability and perceived service quality on airport ground access mode choice. Journal of Choice Modelling, 4 (2), 49–69. Abstract: This study makes two contributions to exist- ing airport ground access mode choice models. The first is an assessment of travel time reliability on air passenger airport ground access mode choice decisions. Revealed preference questions were asked to determine the safety margin allowed for ground access journey to airports. The larger the safety margin allowances, the less reliable the passenger perceived the mode to be. Stated-preference questions were also used to determine the impact of travel time reliability on mode choice decisions. The second contribution of this research is the incorporation of air passenger perceived service quality in the calibration of the airport ground access mode choice model. With the use of the survey data, the effects of safety margin allowances, travel time reli- ability, and perceived service quality on ground access mode choices to Hong Kong International Airport are quantified by a multinomial logit-type mode choice model. For strategic planning, the calibrated model can be used by the airport authority and various transport operators for evaluating the changes in the service attri- butes on modal split pattern in international airports, hence improving the access mode services. Tierney, S.; Kuby, M. (2008). Airline and airport choice by passengers in multi-airport regions: The effect of Southwest Airlines. Professional Geographer, 60 (1), 15–32.

83 Abstract: The business strategy of Southwest Airlines (SWA) features low fares and direct flights between major cities. To minimize aircraft turnaround times, SWA favors smaller, urban-fringe airports over larger, more congested airports. The researchers surveyed pas- sengers flying to the multi-airport regions of Boston- Providence and Baltimore-Washington to assess how many and what types of passengers choose their less convenient airport and why. Maps of final destinations illustrate a reverse traffic shadow favoring smaller air- ports served by SWA. Motives for choosing less conve- nient airports include cheaper fares, fewer delays, and easier ground transport. Logit analysis confirms that leisure travel, traveling with family, and frequent flyer membership, significantly affect the choice of a less convenient airport. Tsamboulas, D. A.; Nikoleris, A. (2008). Passengers’ will- ingness to pay for airport ground access time savings. Transportation Research Part A: Policy and Practice, 42 (10), 1274–1282. Abstract: There are cases when passengers are willing to pay a premium to reduce travel time, in particu- lar when the trip has to be made. This paper aims to provide insight into factors that determine passen- gers’ willingness to pay to reduce travel time for their ground access to an airport. A method is developed that comprises two steps: the identification of the pas- sengers with zero willingness to pay and, from the rest, the estimation of the additional price they are willing to pay to reduce their travel time. For the first step, a probit model was formulated, and for the second, a lin- ear regression model. To this purpose, data have been collected employing stated preference from passengers at the Athens International Airport. It has been found that a high percentage of passengers have zero will- ingness to pay, and of the remaining ones those using public transport have a significant willingness to pay to reduce access travel time. The method and the models are structured in such a way that their transferability to any airport environment is possible, thus providing a useful tool for decisions relating to airport ground access measures. Warburg, V., et al. (2006). Modeling demographic and unob- served heterogeneity in air passengers’ sensitivity to service attributes in itinerary choice. Transportation Research Record 1951: Journal of the Transportation Research Board, 7–16. Abstract: Modeling passengers’ flight choice behavior is valuable to understanding the increasingly competitive airline market and predicting air travel demand. Stan- dard and mixed-multinomial logit models of itinerary choice for business travel are estimated on the basis of a stated-preference survey conducted in 2001. The results suggest that observed demographic- and trip-related dif- ferences are incorrectly manifested as unobserved het- erogeneity in a random-coefficient mixed logit model that ignores the demographic- and trip-related charac- teristics of travelers. Among demographics, gender and income level have the most noticeable effects on sensitiv- ity to service attributes in itinerary choice behavior, but membership in a frequent flyer program, employment status, travel frequency, and group travel also emerge as important determinants. However, residual heteroge- neity is significant because of unobserved factors, even after accommodating sensitivity variations due to demo- graphic- and trip-related factors. Consequently, substi- tution rates for each service attribute show substantial variations in the willingness to pay among observation- ally identical business passengers. Warnock-Smith, D.; Potter, A. (2005). An exploratory study into airport choice factors for European low-cost airlines. Journal of Air Transport Management, 11 (6), 388–392. Abstract: LCCs are an increasingly important part of the European aviation industry. Airport choice is a cru- cial factor in determining their success or failure. While research has been conducted into airport choice fac- tors, their relative rankings have not previously been investigated. This paper addresses this through an exploratory survey of eight European low-cost airlines. The paper finds that demand for low-cost services is the most important choice factor, with aeronautical charges ranked fourth. Further analysis reveals differ- ent requirements depending on airline characteristics. This implies that airport managers need to tailor their service offering to individual low-cost airlines rather than treating the sector uniformly. Wei, W.; Hansen, M. (2006). An aggregate demand model for air passenger traffic in the hub-and-spoke network. Transportation Research Part A: Policy and Practice, 40 (10), 841–851. Abstract: This paper builds an aggregate demand model for air passenger traffic in a hub-and-spoke network. This model considers the roles of airline service vari- ables such as service frequency, aircraft size, ticket price, flight distance, and number of spokes in the network. It also takes into account the influence of local passengers and social-economic and demographic conditions in

84 the spoke-and-hub metropolitan areas. The hub airport capacity, which has a significant impact on service qual- ity in the hub airport and in the whole hub-and-spoke network, is also taken into consideration. The study’s demand model reveals that airlines can attract more connecting passengers in a hub-and-spoke network by increasing service frequency than by increasing aircraft size in the same percentage. This research confirms the importance of local service to connecting passengers, and finds that, interestingly, airlines’ services in the first flight leg are more important to attract passengers than those in the second flight segment. Based on data in this study, it also was found that a 1% reduction of ticket price will bring about 0.9% more connecting passengers, and a 1% increase of airport acceptance rate can bring about 0.35% more connecting passengers in the network, with all else equal. These findings are helpful for airlines to under- stand the effects of changing their services, and also use- ful for quantifying the benefits of hub airport expansion projects. This paper concludes with an example as an applica- tion to demonstrate how the developed demand model could be used to valuate passengers’ direct benefit from airport capacity expansion. Wilken, D., et al. (2007). Airport choice in Germany: New empirical evidence of the 2003 German air traveller survey. Journal of Airport Management 1 (2), 165–179. Abstract: The paper deals with the quantitative rela- tionship between the number of air travelers in a region and the airports chosen in Germany in 2003. The paper presents results of an analysis of airport choice behav- ior of total air passenger demand in Germany, based on data from the German air traveler survey conducted at 17 international and 5 regional airports. Approximately 210,000 passengers were interviewed about their trip origin, destination, choice of travel mode to the airport, purpose of their journey, and further related journey and personal attributes. As a result of analyzing these data, the distribution of airports chosen by passen- gers coming from different regions in Germany can be shown in relation to the journey purpose and des- tination. Based on these data, logit models have been calibrated for each market segment to forecast airport choice in relation to the accessibility and attractive- ness of airports. The research in the present paper is intended to describe the formulation, estimation, and application of an airport choice model by market seg- ment for use in a subsequent paper. Windle, R.; Dresner, M. (1995). Airport choice in multiple- airport regions. Journal of Transportation Engineering, 121 (4), 121–132. Abstract: A logistic model is constructed to predict airport choice in a multiple-airport region and esti- mated using passenger data from the Washington, D.C./ Baltimore area. In agreement with previous work, it was found that airport access time and flight frequencies were significant predictors of airport choice, although, as might be expected, decreased access time and addi- tional flight frequencies were more important to the business traveler than to the nonbusiness traveler. Addi- tional estimations indicated that when only those pas- sengers within reasonable proximity of more than one airport were included in the estimation, the significance of access time decreased and that of flight frequencies increased. Additional variables for a passenger’s experi- ence with an airport were then included in the model and were significant. This would indicate that passen- gers who have used an airport will tend to continue to use the same airport, all other factors being equal. Wooi, L. O., et al. (2010). Note on the determinants of airline choice: The case of Air Asia and Malaysia Airlines. Journal of Air Transport Management, 16 (4), 209–212. Abstract: Logit analysis is employed on primary data from departing air passengers at the Penang Interna- tional Airport, Malaysia, to examine the determinants of airline choice between incumbent Malaysia Airlines and low-cost Air Asia. With the exception of educa- tional level and ethnicity, other socio-demographic characteristics do not play a statistically significant role in determining airline choice. Instead, behavioral fac- tors such as concerns over schedules and fares, routes, booking methods, and purpose of journey are found to be predictors of airline carrier choice. Zhang, Y.; Xie, Y. (2005). Small community airport choice behavior analysis: A case study of GTR. Journal of Air Trans­ port Management, 11 (6), 442–447. Abstract: The issue of airport selection attracts con- siderable attention. However, most studies focus on using advanced discrete choice models to analyze air- port choice behavior in metropolitan areas with sev- eral closely located, competing airports. This paper addresses passengers’ choice behavior in selecting between local small community airports and a more distant major commercial airport. It looks at factors affecting air travelers’ airport choice behavior in cities

85 with small community air service. Data relating to the Golden Triangle Regional Airport in Mississippi is used in logistic regressions to identify the key factors that influence air travelers’ airport choices. Ticket price, experience with Golden Triangle Regional Air- port, and flight schedule were found to be the strongest effects. Zhang, A., et al. (2010). Revenue sharing with multiple air- lines and airports. Transportation Research Part B: Method­ ological, 44 (8–9), 944–959. Abstract: This paper investigates the effects of conces- sion revenue sharing between an airport and its airlines. It is found that the degree of revenue sharing will be affected by how airlines’ services are related to each other (complements, independent, or substitutes). In particular, when carriers provide strongly substitut- able services to each other, the airport has incentive to charge airlines, rather than to pay airlines, a share of concession revenue. In these situations, while rev- enue sharing improves profit, it reduces social welfare. It is further found that airport competition results in a higher degree of revenue sharing than would be had in the case of single airports. The airport–airline chains may nevertheless derive lower profits through the revenue-sharing rivalry, and the situation is similar to a Prisoners’ Dilemma. As the chains move further away from their joint profit maximum, welfare rises beyond the level achievable by single airports. The (equilib- rium) revenue-sharing proportion at an airport is also shown to decrease in the number of its carriers, and to increase in the number of carriers at competing air- ports. Finally, the effects of a “pure” sharing contract are compared to those of the two-part sharing contract. It is found that whether an airport is subject to competi- tion is critical to the welfare consequences of alternative revenue-sharing arrangements.