Click for next page ( 30


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 29
30 APPENDIX Literature Review This literature review was conducted as part of the effort to density.15 Aguirregabiria and Ho (2008) report evidence in a create a tool to assist airports with anticipating changes in air study that the cost of entering new connecting routes declines service due to external shocks (particularly fuel price changes) with a carrier's scale of operations at airports and that hub- that have important implications for airport development bing serves as a deterrent to entry by potential competitors. and finance. Because the mandate is practical, particular atten- Several studies have attempted to explain the market entry tion was focused on empirically based literature that attempts patterns of low-cost carriers (LCCs), who generally do not set to model quantitatively air traffic flows. up hub and spoke systems. Boguslaski et al. (2004) finds that Southwest initially entered dense, short-haul markets and later entered longer-haul markets, partly motivated by network Overview effects. Ito and Lee (2003) report similar results, and also find that LCCs tend to enter markets with above-average prices. Exhibit A-1 provides an overview of study objectives and the Oliveira (2008) presents evidence that Gol Airlines, an LCC major findings of the empirical literature we have reviewed. in Brazil, engaged in entry strategies similar to those of South- Many of these studies attempt to explain the structure of west. Both Oliveira (2008) in a study of the U.S. market and the commercial airline industry--how the airline network Alderighi et al. (2004) find that full-service carriers lower prices system evolved, the nature of completion among carriers in response to market entry by LCCs. including strategic entry into markets, and the role that Virtually all of the studies listed in Exhibit A-1 define prod- scale and density economies play. Two issues have received ucts as route-specific trips between airport pairs. In this sense, special attention: the emergence of the hub and spoke sys- these studies address, at least indirectly, the issue of modeling tem, and strategic entry by low-cost carriers like Southwest airport-specific traffic patterns. However, most of these stud- Airlines. ies do not model traffic volumes (either the number of flights Arguments have been made in the literature that the hub or the number of passengers) explicitly. For example, those and spoke system confers both cost and demand-side advan- studies that focus on carrier entry patterns typically model tages to carriers. Berry (1990), for example, notes that hub discrete outcomes (i.e., an airline either does or does not offer and spoke systems reduce the number of round-trips needed service at a particular airport).16 While carrier presence and to transport a given number of passengers and, given economies traffic volumes are related, it is not always possible to distin- of flying large planes, can produce cost savings sufficient to guish one from the other because of variations in aircraft size, overcome the costs of flying more miles (on connecting load factors, and flight frequency. One exception is Borenstein spokes). At the same time, he argues that hubbing is a form and Rose (2003) who model the effects carrier bankruptcies of product differentiation that allows airlines to offer ser- have on airport-specific service levels. They find no significant vices for which passengers are willing to pay premiums. The bankruptcy effects on service levels at large and small airports, demand-side advantages include superior gate and ticketing services, higher flight frequency, and frequent flyer programs. 15 Spoke density confers cost advantages in that it allows carriers to use larger In a later study, Berry et al. (1997) find evidence that hubbing planes that have lower costs per seat than smaller aircraft. 16 Tamer and Ciliberto (2007) and Sugawara and Omori (2008) make probabilis- airlines are able to charge fare premium to relatively price- tic estimates of carrier service entry into specific airports. Morrison and Winston inelastic (business) travelers for such services, and that hub- (1995) make similar estimates at the route level (entry and exit from specific bing confers cost advantages related to economies of spoke airport pairs).

OCR for page 29
31 Exhibit A-1. Literature summary--study objectives and major findings. Study Study Objective Major Findings Alder et al. (2008) Assess European transport Investments in rail infrastructure will improve social infrastructure investments. welfare. Aguirregabiria & Effects of demand, costs, and Cost of entry into a route declines with scale of Ho (2008) strategic factors on adoption of airline's operations at connecting airports. Also, hub hubspoke networks. spoke networks deter strategic entry by rivals. Alderighi et al. Response of full-service carriers to Incumbent carriers lower fares for both business and (2004) entry of low-cost carriers in leisure travelers when low-cost carriers enter markets. Europe. Berry (1992) Effects of airlines' scale of Within-market competition limits the number of operations on profits, as indicated entering firms, even if airport access restrictions are by entry decisions. eased. Berry et al. (1997) Estimate the effects of hubs on Hubbing airlines' ability to raise fares limited mainly to airline costs and price markups. price-inelastic travelers. Find evidence of economies of spoke density. Berry (1990) Test hypothesis that airport Airport presence by carriers increases demand for air presence (e.g., better service travel and explains, in part, pricing practices by related to hubspoke system) affect hubbing airlines. demand as well as costs. Boguslaski et al. Explain entry patterns of Initially, Southwest entered dense short-haul markets, (2004) Southwest Airlines. then entered long-haul markets, partially motivated by network effects. Borenstein & Rose Estimate the effect of airline No substantial effects of bankruptcies on large and (2003) bankruptcies on air service. small airports, but some impacts on medium-sized airports. Goolsbee and Analyze how incumbents respond Incumbents decrease fares substantially on threatened Syverson (2008) to threat of entry (by Southwest). routes. Ito & Lee (2003) Identify market characteristics Low-cost carriers enter dense markets with above- affecting entry of nonstop, low- average prices; entry no longer limited to short- and cost carriers. medium-haul markets. Lederman (2003) Investigate the effects of frequent Frequent flyer programs affect airline demand and flyer programs and product pricing strategies. Low-cost carrier entry is a form of differentiation on airline demand product differentiation. and pricing. Morrison and Explain route entry and exit Carrier entry decisions depend on own and other Winston (1995) decisions of U.S. carriers from carriers' hub status, expected fare, and presence of 19881992. Southwest. Exit decisions are influenced similarly, but carriers more likely to exit long-haul markets. Oliveira (2008) Explain entry patterns of low-cost Initially, Gol focused on high-density, short-haul carrier Gol Airlines in Brazilian markets, but then diversified into longer-haul markets. domestic market. Pai (2007) Identify the determinants of Aircraft size and flight frequency increase with market aircraft size and flight frequency population, income, and runway length. on airline routes. Sugawara & Omori Model airline entry decisions. Predict entry probabilities for two airlines at new (2008) Shizuoka airport. Tamer & Ciliberto Investigate impacts of firm Competitive effects of low-cost carriers are different (2007) characteristics on market structure from large airlines and are increasing in airport of U.S. airline industry. presence, and repealing Wright Amendment would increase markets served out of Dallas Love by 20%. Yan et al. (2008) Explain point-to-point network Main network effects are airport and regional presence, effects and entry patterns of and substitutability of markets. Southwest Airlines. and small effects on medium-sized airports. Pai (2007) mod- markets with a small number of sellers, each of whom may influ- els traffic volume measured as flight frequency and finds that ence the decisions of the other sellers). In this setting, passen- frequency increases with market population, income levels, gers are assumed to be so-called "utility maximizers" and firms and maximum airport runway length. engage in strategies that they believe are consistent with profit maximization. The demand facing any single carrier depends on Demand-Side Modeling the pricing, output/capacity, and market entry decisions of its rivals. Indeed, several authors make explicit assumptions about Before discussing the details of demand-side modeling, a the nature of the strategic "games" that rivals play in markets.17 brief digression on market structure is worthwhile. It is fair to say that there is a consensus in the recent literature that domes- 17As we explain later in this review, some authors incorporate assumptions tic air carriers participate in "oligopolistic" markets (meaning about strategic gaming explicitly in their econometric models.

OCR for page 29
32 Exhibit A-2 summarizes market/product definitions and and some, for example, Berry (1990) and Aguirregabiria and demand-side control factors that are used in the studies that Ho (2008), specify structural models in which these factors have been reviewed. Most of the studies define a "product" as appear in both demand and cost equations. a non-directional one-way or round-trip route between air- port pairs. Aguirregabiria and Ho (2008) define a product as Supply/Cost Modeling a round-trip, but distinguish direction. The demand-side control variables generally fit into three Exhibit A-3 describes the flight cost/supply factors and cost categories: controls for buyer (passenger) characteristics, con- economy measures used in the reviewed studies. Perhaps the trols for site (origin/destination) characteristics, and controls most important feature of supply-side modeling is the absence for product differentiation. Two commonly used types of of cost data that can be linked to route-level demand-side data. controls for buyer characteristics are: Moreover, no study that was reviewed controlled explicitly for fuel costs. Passenger income in airport market areas--measures used Because of the lack of data, researchers have generally in the literature include average per capita income for adopted one of two strategies for controlling for carrier costs: city/airport pairs, per capita GDP at the departing airport, the minimum and maximum per capita GDP in city pairs, Impute costs from fully specified structural models and changes in state-level income and employment. Include proxies or instrumental variables as controls for Number of potential passengers in airport market areas-- costs measures include average population for city pairs, and the geometric mean of population at market endpoints. Two studies, Aguirregabiria and Ho (2008) and Berry et al. (1997) adopt the first strategy. Both specify full structural Also, some researchers have attempted to capture differ- models, assume strategic behavior on the part of air carriers, ential pricing strategies by airlines by distinguishing from and find market equilibria as solutions to N-person games. business (relatively price-inelastic) travelers and leisure (rel- They then compute imputed costs as the difference between atively price-elastic) travelers. Berry et al. (1997) and Lederman observed prices and optimal (profit-maximizing) markups, (2003) model differential pricing explicitly by assigning pas- which are independent of costs.18 sengers to "business" and "leisure" groups from fare distribu- Most of the studies reviewed adopt the second strategy and tions observed in the samples they use. Boguslaski et al. (2004) control for cost variables through the use of proxy variables. include controls for the fraction of leisure travelers in their These proxy variables include: model. Finally, Pai (2007) controls for the percentage of man- agerial workers in airport market area workforces. Trip distance Several studies distinguish origin/destinations characteris- Hub presence, measured as hub size or dummy variables tics by controlling for so-called vacation sites. For example, Ito indicating the existence of hubs and Lee (2003) include dummy variables for Sunbelt states; Pai Airport congestion (e.g., average delay, slot constraint indi- (2007) includes dummy variables for Las Vegas and Orlando; cators, airport volume) Yan et al. (2008) include dummy variables for Nevada and Maximum runway length Florida trips, and Berry et al. (1997) include mean temperature New carrier verses legacy carrier indicators differences between city pairs. Several authors recognize and attempt to control for product Also, Oliveira (2008) uses city-specific fixed effects to control differentiation in their studies. The following are commonly for cost differences across airports. used controls: Some studies (particularly those focused on entry deci- sions) also include, as supply-side variables, indicators of the Nonstop verses connecting flights degree of competition at airports. Several compute Herfindahl- Hub presence, captured as dummy variables or measures Hirschman indices at airports to control for competition levels, of hub size and Goolsbee and Syverson (2008) and Boguslaski et al. (2004) Trip length include dummy variables for Southwest Airlines' presence at Flight frequency between airport pairs airports as an indicator of entry threat potential. As noted earlier, these factors are also likely to affect carrier costs in addition to affecting service quality, and hence demand. Some authors have characterized these factors as demand-side 18 The optimal markup depends only on price elasticity, and not the level of controls; others interpret them as cost/supply-side controls; marginal cost.

OCR for page 29
33 Exhibit A-2. Literature summary--demand modeling. Study Market and Product Definitions Demand Control Factors Alder et al. (2008) Business and leisure trips on hub Trip time, transport alternatives. spoke and low-cost air, and rail transport. Business and leisure differential pricing. Aguirregabiria and Directional round-trip between cities. Hub size at origindestination and connecting Ho (2008) airports, distance, and nonstop flight indicator. Alderighi et al. City pair trips for various passenger Per capita GDP in area of departing airport. (2004) subclasses (promotional, discounted economy, unrestricted). Berry (1992) Dependent variable is entry into city Market characteristics, proxies for profitability, pair markets. Market characteristics, include distance, population (product of city pair proxies for profit, include distance, populations), tourist site indicator, and measures of population (product of city pair airport presence. populations), tourist cite indicator, and measures of airport presence. Berry et al. (1997) Directional round-trip between city Trip distance, direct flight indicator, airport pairs. Distinguish between high- and congestion indicator, population of end-point cities low-elasticity passengers. (geometric mean), temperature difference between city pairs (tourism indicator), flight frequency proxy. Berry (1990) Round-trip itineraries between city Population (product of city pair populations), trip pairs. distance, airport presence (number of top 50 cities served by airline from airport). Boguslaski et al. City pair trip, regardless of direction. Density (daily number of passengers on all flights), (2004) geometric mean of population in city pair, per capita income at origin and destination, maximum fraction of leisure travelers among the city pairs, trip distance. Borenstein and Rose Two different measures of service: Seasonal and time-period fixed effects, changes in (2003) total nonstop domestic flights to and state-level employment, and changes in state-level from airport; total number of income. domestic locations served nonstop from airport. Goolsbee and Airport to airport trip. Demand controls not identified. Syverson (2008) Ito and Lee (2003) Round-trip and one-way itineraries. Route density (average daily number of passengers carried by all passengers), distance, population at endpoint cities, per capita income at endpoint cities, "vacation" cite indicator (sunbelt states). Lederman (2003) Carrier-specific round-trips. Airline-route fixed effects, airline-quarter (time) fixed effects, fare distributions (percentiles), hub presence, airline flight shares. Morrison and Carrier-specific route between two Slots, distance, density, relative fares, population Winston (1995) airports. and real per capita income at origin and destination. Oliveira (2008) Non-directional origin and City-specific dummy variables intended to capture destination routes aggregated to city geographic idiosyncrasies such as income, wealth, levels. and propensities for business and leisure travel, trip distance. Pai (2007) Dependent variables are aircraft size Percentage of households with income greater than and flight frequency between airport $75,000, percentage of managerial workers in labor pairs. force, percentage of population under age 25, in airports' MSAs; route distance, leisure travel indicator (Las Vegas and Orlando). Sugawara and Route between two airports. Population at airports. Omori (2008) Tamer and Ciliberto Non-directional trip between two Average population, average per capita income, (2007) airports. average rates of income growth at market endpoints, distance to closest competing airport, trip distance, and distance form market endpoints to the geographic center of the United States. Yan et al. (2008) Airport pair routes. Distance between airports, average population, average per capita income, and vacation site (Nevada and Florida).

OCR for page 29
34 Exhibit A-3. Literature summary--supply/cost modeling. Study Flight Cost/Supply Factors Hub/Spoke Density Economies Alder et al. (2008) Function of great circle distance and number of seats Not measured. for short and long haul. Aguirregabiria and Costs not modeled explicitly. Hub size, trip distance, Estimate economies of hub size. Ho (2008) nonstop, and airline-specific effects; airport effects. Model distinguishes variable flights costs, fixed flight costs, and entry costs imputed from price markups. Alderighi et al. Trip distance, Herfindahl-Hirschman index Not measured. (2004) computed over all full-service carriers serving market, presence of low-cost carrier in market. Berry (1992) Costs not modeled explicitly. Distance between city Not measured explicitly, but pairs, airport presence used as proxies. measures of airport presence (city pair market shares, number of routes served out of airport) included in models. Berry et al. (1997) Costs not modeled explicitly (computed as Spoke density economies imputed difference between fares and markups). Cost from differences between fares and instruments include airport congestion, segment markups. distance, and trip frequency proxy. Berry (1990) Costs not modeled explicitly. Distance between city Airport presence used as proxy for pairs, airport presence (number of top 50 cities hub density. served by airline from airport), and instruments for new versus legacy carriers used as proxies. Boguslaski et al. Costs not modeled explicitly. Supply-side proxies Not measured directly, but measures (2004) include number of cities served at trip endpoints, of Southwest presence at airports Southwest share of O&D passengers, and several interpreted as measures of network indicators of competiveness including presence of effects. competing hub and Herfindahl-Hirschman indices at end point cities. Borenstein and Rose Costs not modeled explicitly. Market share of airline Not measured. (2003) filing for bankruptcy included as supply-side variable. Goolsbee and Costs not modeled explicitly. Southwest presence at Not measured. Syverson (2008) airports included as supply-side entry threat variables. Ito and Lee (2003) Costs not modeled explicitly. Supply-side indicators Not measured. include hub presence, delays (dummy variable for 10 airports with highest delays), multiple airport cities, Herfindahl-Hirschman indices for endpoint cities. Lederman (2003) Costs not modeled. Not measured. Morrison and Costs not modeled. Other carriers' presence at Not measured. Winston (1995) airports included as supply-side entry-threat variables. Oliveira (2008) Costs not modeled explicitly. Gol presence at Not measured directly, but city- airports included as supply-side entry threat specific dummy variables interpreted variables. City-specific dummy variables interpreted as proxies for network effects. proxies for cost and air travel service support differences across airports. Pai (2007) Costs not modeled explicitly. Supply-side variables Not measured directly, but number include number of nearby airports, maximum of destinations served and runway length, airport delays, and slot constraint proportion of passengers with indicator. connecting flights used as hub presence proxies. Sugawara and Costs not modeled explicitly. Distance used as a Not measured. Omori (2008) measure of travel cost. Availability of high-speed train used as air travel alternative. Tamer and Ciliberto Costs not modeled explicitly. Geographic distance Not measured directly, but cost (2007) between airlines' closest hub and market endpoints proxy variable used to control for used as proxy for cost. hub effects. Yan et al. (2008) Costs not modeled explicitly. Supply-side proxies Not measured directly, hub presence include Herfindahl-Hirschman index (maximum of variables used to control for hub airport pair), airport volume (maximum of airport effects. pair), and dummy variables for full-service hub presence.

OCR for page 29
35 Econometric Methods and distributions of model error terms, estimates of parame- ters are then drawn iteratively (using simulation estimators) Exhibit A-4 identifies the econometric methods employed until the values of observed variables (e.g., prices) can be in the recent literature. Generally, these methods can be clas- retrieved. Estimating these models is typically very computa- sified into the following three groups: tionally intensive. Multivariate regression models Discrete choice models--logit and probit estimators Concluding Remarks Structural models--simulation estimators Demand-side models in the recent literature are relatively The choice of estimators depends primarily on model speci- rich, primarily because data on passenger, site, and product fications. characteristics can be married with detailed, route-specific The multivariate regression models have been employed to DOT data on U.S. domestic air travel. The cost or supply-side estimate reduced form models when the dependent variable modeling in the literature is much less rich because of lack of of interest is continuous. For example, Borenstein and Rose data. Most researchers have resorted to controlling for costs (2003) use this technique to explain traffic volumes at airports through proxies and instruments. Two of the structural models at which bankruptcies have occurred. Goolsbee and Syverson reviewed impute detailed cost estimates as differences between (2008) is primarily interested in explaining variations in air observed prices and optimal markups. However, neither of fares, and Pai (2007) models two continuous variables-- these models incorporates the effects of exogenous shocks aircraft size and flight frequency. such as changes in fuel prices. Many authors employ logit and probit models that are suit- Many of the studies reviewed attempt to explain the evolu- able for use when the dependent variable of interest is discrete. tion of the structure of airline markets, and several of these Many of the studies reviewed have used these estimators to focus on market entry decisions. While these models provide model market entry decisions including, for example, Berry useful insights, they fall short as tools for modeling airport- (1992), Boguslaski et al. (2004), Ito and Lee (2003), Morrison specific traffic and revenue streams. While carrier entry (and and Winston (1995), and Oliveira (2008). exit) decisions are linked to airport traffic volumes, modeling Several studies, including Aguirregabiria and Ho (2008), these is not sufficient to predict airport traffic flows. Also, Berry (1992), Berry et al. (1997), and Sugawara and Omori most of these studies employ discrete choice estimators (logit (2008), employ structural models in their work. These mod- and probit). These models are well suited for identifying pat- els have been developed, in part, out of empirical work in the terns of behavior for populations (i.e., the industry as a whole), field of industrial organization. In these models, consumers but typically have weak predictive power for individual obser- are assumed to behave consistently with utility maximization, vations (i.e., specific airports). and firms attempt to maximize profits while playing strategic The structural models are the most sophisticated of (oligopolistic) games. Given assumptions about the structure those reviewed. These models are capable of dealing with Exhibit A-4. Literature summary--econometric methods. Study Econometric/Statistical Methods Alder et al. (2008) Nested multinomial logit model. Aguirregabiria and Ho (2008) Recursive pseudo maximum likelihood estimator [See Aguirregabiria and Mira (2007)]. Alderighi et al. (2004) Multivariate regression model. Berry (1992) Probit model, simulation estimator. Berry et al. (1997) Simulation estimator. Berry (1990) Simulation estimator. Boguslaski et al. (2004) Probit model. Borenstein and Rose (2003) Multivariate regression model. Goolsbee and Syverson (2008) Multivariate regression model. Ito and Lee (2003) Probit model. Lederman (2003) Nested logit model. Morrison and Winston (1995) Probit model. Oliveira (2008) Amemiya Generalized Least Squares (AGLS); probit model. Pai (2007) Multivariate regression models. Sugawara and Omori (2008) Bayesian estimation using Markov chain Monte Carlo Simulation. Tamer and Ciliberto (2007) Multinomial logit model. Yan et al. (2008) Spatial probit model.

OCR for page 29
36 endogeneity and strategic behavior and in two cases have per- Borenstein, Severin (1990). "Airline Mergers, Airport Dominance, and mitted researchers to make inferences about underlying cost Market Power," American Economic Review, 80(2): 400404. Borenstein, Severin and Nancy Rose (2003). Do Airline Bankruptcies structures. However, estimating and using these models is very Reduce Air Service? Cambridge, MA: National Bureau of Economic computationally intensive. This drawback would appear to rule Research. (NBER Working Paper W9636) out these types of models as good candidates for practical tools Goolsbee, Austan and Chad Syverson (2008). "How Do Incumbents for predicting airport-specific traffic flows. Respond to the Threat of Entry? Evidence From the Major Air- lines," Quarterly Journal of Economics, 123(4): 16111633. References Ito, Harumi and Darin Lee (2003). Incumbent Responses to Lower Cost Entry: Evidence From the U.S. Airline Industry. Brown University, Adler, Nicole; Chris Nash and Eric Pels (2008). High-Speed Rail and Air Department of Economics. (Working Paper 2003-22) Transport Competition. Amsterdam: Tinbergen Institute. (TI Dis- Lederman, Mara (2003). Airline Strategies in the 1990s: Frequent Flyer cussion Paper 2008-103/3) Programs, Domestic and International Partnerships, and Entry by Low- Aguirregabiria, Victor and Chun-Yu Ho (2008). A Dynamic Oligopoly Cost Carriers. Ph.D. thesis, Massachusetts Institute of Technology, Game of the US Airline Industry: Estimation and Policy Experiments. Department of Economics. University of Toronto, Department of Economics. (Working Morrison, Steven A. and Clifford Winston (1995). The Evolution of the Paper 337) Airline Industry. Washington, DC: The Brookings Institution. Aguirregabiria, Victor and P. Mira (2007). "Sequential Estimation of Oliveira, Alessandro (2008). "An Empirical Model of Low-Cost Carrier Dynamic Discrete Games," Econometrica, 75: 153. Entry," Transportation Research Part A: Policy and Practice, 42(4): Alderighi, Marco; Alessandro Cento, Peter Nijkamp and Piet Rietveld (2004). The Entry of Low-Cost Airlines: Price Competition in the Euro- 673695. pean Airline Market. Amsterdam: Tinbergen Institute. (TI Discus- Pai, Vivek (2007). On the Factors That Affect Airline Flight Frequency and sion Paper 2004-074/3) Aircraft Size. University of California-Irvine, Department of Eco- Berry, Steven (1990). "Airport Presence as Product Differentiation," nomics. (Working Paper 070803) American Economic Review, 80(2): 394399. Sugawara, Shinya and Yasuhiro Omori (2008). Bayesian Estimation of Berry, Steven (1992). "Estimation of a Model of Entry in the Airline Entry Games With Application to Japanese Airline Data. University Industry," Econometrica, 60(4): 889917. of Tokyo, Faculty of Economics. (CIRJE Working Paper F-556) Berry, Steven; Michael Carnall and Pablo Spiller (1997). Airline Hubs: Tamer, Elie and Frederico Ciliberto (2007). Market Structure and Multi- Costs, Markups and the Implications of Customer Heterogeneity. Yale ple Equilibria in the Airline Industry. Social Science Research Network University, Department of Economics. (Revision of NBER Working Working Paper. Paper W5561) Yan, Jia; Xiaowen Fu and Tae Oum (2008). Exploring Network Effects of Boguslaski, Charles; Harumi Ito and Darin Lee (2004). "Entry Patterns Point-to-Point Networks: An Investigation of the Spatial Entry Pat- in the Southwest Airlines Route System," Review of Industrial Orga- terns of Southwest Airlines. Washington State University, School of nization, 25(3): 317350. Economic Sciences. (Working Paper 2008-21)