Fare and Location Competitiveness Analysis of Major Hub Airports in East Asia

Fare and Location Competitiveness Analysis of Major Hub Airports in East Asia Hyoungku YEO Ph.D Direct general policy communities Management officer Ministry of land, transportarion and Maritime Affairs, Jungang-dong, Gwacheoun 427-712 South KOREA Fax : +82-2-504-6825 E-mail : hkyeo@moct.go.kr Kyungwoo KANG Professor Transportation Engineering University of Hanyang Sa-3dong, Sangrok-gu, Ansan 426-791 South KOREA Fax : +82-31-406-6290 E-mail : kyungwoo@hanyang.ac.kr Wookag KOOK Ph.D Transportation Engineering University of Hanyang Sa-3dong, Sangrok-gu, Ansan 426-791 South KOREA Fax : +82-31-406-6290 E-mail : trkw@lycos.co.kr Abstract: Main airlines naturally dominate the Incheon International Airport. The government executes various trade policies in order to protect main airlines. This paper analyzes the distances between hub airports in Asia and the distances of spokes, as well as relationships with other variables, travel fare, and monopoly power. The purpose is to recognize the position of the Incheon International Airport through analyzing fare and location differences between Incheon International Airport and other hub airports by market. The result appears that Incheon Airport was ranked 4th in fare competitiveness and location among East Asia airports, where the most competitive cities were Singapore in terms of fare competitiveness and Tokyo in terms of location competitiveness. Strategic and political schemes from the government are needed by Incheon Airport. Key Words: Hub-spoke system, Incheon Airport, Fare, Location 1. INTRODUCTION Recently, East Asia has been grappling with various changes in air transportation due to changes in international air transportation brought about by deregulation and free competition, which have been implemented rapidly by the United States and Europe. Therefore, each country has been attempting to build competitive hub-airports. Currently, South Korea has been dealing with these changes by attempting to construct a main hub of East Asia air transportation with Incheon International Airport, based on a Hub and Spoke System.

However, the current situation is challenged by the current system of competition in order to make inroads into the air transportation market of neighboring countries, if one country becomes a hub, then neighboring countries become spokes. However, hub airports must be led by airline companies that have complex and accurate forecasts of demand for air travel, and, in intense competition, the government assists with various back-up services, infrastructure support, and favorable policies to airlines. Each government constructs and extends an airport by positively emphasizing infrastructure investment for the hub airport. Thus, this tends to create only artificially complete hub airports because of excessive competition between countries. Therefore, each country and city makes an effort to achieve competitive superiority by constructing competitive hub airports. However, each country would do this uselessly if the country does not understand the nature of competition among airports. Excessive investment and deficit operations would have negative effects on the airline industry and related industries. Actually, some of the most recently opened airports are in trouble because of a lack of demand for air travel, excessive investment in infrastructure, and deficit operations. As a result, they have made an effort to increase the rate of operation. Specifically, because South Korea that has many geopolitical and social restrictions and difficulties in budget operation, these problems are very important to Incheon International Airport, which contributes to national competitive advantage. Therefore, the strong merits of Incheon International Airport should be emphasized and the weak points should be corrected. From this point of view, this research analyzes how Incheon International Airport is ranked among airports in Asia and what the strengths and weaknesses of Incheon International Airport are. Section 2 summarizes the major findings of previous studies on hub airports in Asia. Section 3 presents the theoretical background and the model used in this paper. Section 4 presents data and the main results. Section 5 summarizes and discusses the main policy implications. 2. PREVIOUS STUDIES OF HUB AIRPORTS IN ASIA Borenstein (1989) was the first to tackle the problem of airport dominance. Borenstein relates the fare paid for a trip to the cost inducement of the flight and related characteristics. He used relative prices for scaling, and the logarithm of the price of an airline on a route relative to the price of another airline on the same route as the dependent variable. By scaling down to the competitor on the same route, Borenstein was able to reduce the model to the airline s specific differences. The variable reflecting hub dominance is the average of an airline's share of daily passengers originating at the two endpoints of the route, weighted by the proportion of passengers on the route who originated at each endpoint. He finds the effect of this variable on fare differences to be significantly positive, indicating dominant positions of airlines in their hubs. Borenstein (1991) again uses the differences on the same routes as a scaling method. This time, however, the share of traffic on a route is the dependent variable. Again, Borenstein finds significantly positive results, indicating that airlines exercise some market power over their hubs. He also tried to isolate the effect of Computer Reservation Systems. He found that the effect is small and statistically insignificant.

Berry (1990) estimates a structural model, using airport presence as an explanatory variable in both cost and demand equations. Airport presence is defined as the number of top 50 cities served by an airline out of a given city. Berry finds that airport presence has a positive effect on demand while simultaneously lowering costs. Although Berry interprets airport presence as product differentiation, which, therefore, is different from Borenstein's market power approach of hub dominance, their results point in the same direction. Evans and Kessides (1993) use a fixed effects estimation method to test whether airport and route dominance, both measured by respective market shares, have a significant effect on fares. They found that airport dominance confers substantial pricing power upon the airline, which is consistent with earlier results. However, they also found that the isolated impact of route dominance on fares is insignificant. Berry, Carnall, and Spiller (1996) extend Berry's earlier work on the influence of airport presence to costs and demand. Using an unobserved product characteristic to capture restrictions placed on a ticket, and thus implicitly dividing the market between tourists and business travelers, they found that airport presence has a positive effect on demand and a negative effect on costs. The cost-reducing effect of airport presence can be interpreted as the obstacle to profits that comes from the economic rationale of the hub and spoke system. The effects are labeled as the hub premium, and whether product differentiation, market power, or both interfere with the Hub Premium Effect has not yet been proven. Marin (1995) was the first to address the issue in the European context, by estimating both market shares and price equations, in both regulated and deregulated segments. Marin finds that the effect of airport presence is not significant in a regulated environment, but significantly negative in a deregulated environment. This implies that, in contrast the U.S. situation, European airlines in a deregulated environment tend to exploit the cost reducing effect of airport presence in order to compete on prices. Marin explains the difference between his European results and earlier U.S. estimates by pointing out significant differences in the causes of market power. Because of these differences, FFPs hardly existed in Europe and the Hub-and-Spoke system had not yet gained as much ground as it has in the U.S. aviation industry. Robert Windle (1999) showed the mathematical relationship between the airline competition and profit rates that are faced by low-cost airlines. He designs a model the explains fares with five categories of variables such as distance squared, population, dummy variables for slot controlled routes, vacation routes, and eleven quarter dummy variables. Four of five route categories showed that the influence of distance is statistically significant, and three of five categories showed that the distance squared was significant. This indicates that encouraging the introduction of low-cost airlines increases consumer welfare, and should therefore be government policy. Mark G. Lijesin (2000) analyzed the advantages of European hub airports. There was a statistically significant relationship between fare and distance. Timothy H. Hannan (1997) explains the roles of market share inequality, the number of competitors, and HHI (Herfindahl-Hirschman Index) with bank deposit and loan rates. Despite the limitations of the HHI, it is widely used as a measure of competition. Among the limitations of the HHI, one limit is that it gives equal weight to market share inequality and

number of competitors. This problem is solved using HHI analysis through the reciprocal of competitors and dispersion of product. 2 V 1 ( HHI = ) N + N (1) Robert Windle used HHI consisting of passenger of airlines that used specific airports. The research on the international public transportation system is relatively uncommon. Keeling (1995), Rimmer (1996) and Matsumoto (2004) analyzed the international public air transportation system. Murayama (1991 a,b) analyzed the public air transportations systems in Canada and U.S. Matsumoto (2003) analyzed Asia and Park (1995) analyzed Japan and Korea. The most recent research, Matsumoto (2007), estimated a model using passenger and freight data in Asia, Europe, and America using a basic gravity model (GDP, population, distance). The result revealed differences among cities using a city dummy variable, but a basic gravity model analysis could not reflect monopoly power and changing fares by a monopolist hub airport. Thus, a international monopolistic hub airport was not reflected by a point of difference among the cities. Taaffe (1962), for example, attempted to apply the gravity model to air passenger flows in the U.S. to examine spatial organization. Harvey (1951) and Richmond (1955) estimated simple gravity models using population and distance variables to predict passenger numbers. Lansing & Blood (1958) and Lansing et al. (1961) analyzed other variables such as income, education level, the accumulation level of enterprises, and measures of city characteristics, such as location advantages and climate. Howrey s (1969) studies have also focused on the supply aspect by introducing fare, time, and service frequencies. Long (1970) used variables for business passengers, tourist passengers, and cargo. The chief objective of the research is to analyze international air network structures within and among Asian, European, and American regions from the standpoint of urban systems, and to reveal the degree of air traffic density for major cities worldwide, using a gravity model. 3. THORETICAL BACKGROND AND MODEL DEVELOPMENT Mark G. Lijesen s (2000) analysis rests on the assumption that fares are cost-related. From this, we assume that costs are distance-related. When we combine these assumptions, we find that there should be a relationship between fares and distance. This research changed the

numerical formula with the exception of variables connected with airlines for the airport analysis for fare and distance of Asian hub airports. Figure 1 Hub-and-Spoke System Consider a simple hub and spoke network as shown in Figure 1. In this network, we define H as the hub, A, B, and C as close spokes, and X, Y, and Z as distant spokes. Assuming distance-related fares, we define the natural logarithm of the fare of a flight from hub H to destination as: log fare = a + b logdist (1) HZ HZ HZ HZ HZ fare : fare from hub H to destination dist : distance from airport H to airport Z HZ Rfare Rdist HHI fare fare AHZ = (2) dist dist HZ AHZ = (3) HZ x / emf x / emf 2 = å ( ) (4) å x : number of passengers at airport emf : mileage of service distance (each main airline) Matsumoto (2007) proposed an analytical method in which the location competitiveness of the airport used a basic gravity model (GDP, population, and distance). T ij a b ( GiG j ) ( PP i j ) = A (5) c ( R ) ij where, G : GDP P : population T : total trip distance

R : distance between cities The equation to express differences among cities, including the city dummy variable, is: T ij a b dd1 dd2 dd3 GiG j PP i j e e e c ( Rij ) ( ) ( ) = A (6) In this research, a model was developed to reflect monopolistic nature of the airport, HHI, and fare. T ij a b dd1 dd2 dd3 HHI j GiG j e e e c ( Fareij ) ( ) ( ) = A (7) where HHI x ( ) x 2 = å å expresses the monopoly and x is passenger occupancy among continental airlines. Passenger Air Tariff (Worldwide Fares) was used fare. 4. ANAYSIS COMPETITIVE POWER OF INCHEON AIRPORT 4.1. Analysis Fare Competitiveness 4.1.1. Data and Basic Statics We selected six Asian origins and nine non-asian destinations. Nine locations in this data set turned out to be intercontinental hubs for their main airline. When we selected the origins and destinations, we tried to combine main hub airports. The airports used in the analysis are summarized in Table 1. We needed the main airline selection to receive the necessary service distance passenger mileage when we produced the HHI variable. Fare and distance data were retrieved from Passenger Air Tariff data, and passenger numbers and frequency data came from Air Statistics 2005. Mileage data was retrieved from the internet homepage for each main airline. Table 1 Origins, hubs and destinations in the sample From (abbreviation) : A Via (Airline) : H To (abbreviation) : Z Incheon: ICN Korean Air Los Angeles: LAX Narita: NRT Japan Airlines San Francisco: SFO New York: JFK Beijing: PEK Air China Chicago: ORD Hong Kong: HKG Cathay Pacific Frankfurt: FRA Paris: CDG Singapore: SIN Singapore London: LHR Taipei: TPE China Airlines Sydney: SYD Rome: FCO

All airports (of East Asian origin) chosen for analysis show service frequencies that ranked within the top 25. The reason that service frequency was chosen for the standard was because frequency is more important than passengers. The destinations of the flights were chosen because passengers were more important than frequency in this case. The hub and destination airports in the analysis are summarized in table 2. Table 2 Number of observations by hub and destination ICN NRT PEK HKG SIN TPE Total LAX 5 5 4 5 5 4 28 SFO 5 5 4 5 5 4 28 JFK 5 5 4 5 5-24 ORD 2 2 - - - - 4 FRA 6 6 5 6 6 5 40 CDG 5 5 5 5 5-30 LHR 5 5 5 5 5-30 SYD 6 6 5 6 6 5 40 FCO - - 5 6 6-17 Total 39 39 37 43 43 18 219 4.1.2. Analysis Result on Fare Competitiveness As shown in Table 3, the variable Log Rdist is not directly proportional with fare, but is a factor of 0.8. It can know that as the fare from the distance from the origin to the hub airport does not increase, and pays less than the distance. This means that competition between Asian airports in establishing hub airports is primarily reducing the fare or price. Variable Log dist_hz is negative. If the distance from hub to destination airport decreases, either the fare from origin via hub to destination decreases or the fare from hub to destination increases. The former means that airport competition is influx into hub airports, similarly Log Rdist, and the latter shows fare clearly increase by increase of distance. Variable HHI_HZ and HHI_AZ are negative. If the monopoly power of the hub and origin airport is stronger, the fare from the origin via the hub to the destination either decreases or the fare from hub to destination increases. First, if the degree of monopoly power of a hub airport is large, the fare should be reduced by the increased influx of passengers. The influx of passengers can also increase the fare to the destination, which also increases competitive advantage. Similarly, if the degree of monopoly power of an origin airport is large, then the fare via the hub should be reduced to develop the hub airport. Furthermore, the coefficient of variable HHI_AZ is smaller than variable HHI_HZ. This means that the decrease in fares based on the degree of monopoly power of origin airports is larger than that of hub airports. In other words, competition between Asian airports to develop hub airports is becoming intense.

Table 3 Regression results for log Rfare Variable Log Rdist Log dist_hz HHI_HZ HHI_AZ coefficient (t-values) 0.80 (10.22)*** -0.08 (-1.64)* -1.53 (-1.54)* -4.53 (-4.0)*** *** indicates significance at the 5% level * indicates significance at the 10% level 4.2. Analysis Location Competitiveness Location competitiveness of Incheon airport is analyzed with fare instead of distance, because fare by distance is not directly proportional in the previous fare analysis. Additionally, including HHI reflects monopoly power instead of population. 4.2.1. Data and Basic Statics Table 4 Data to analyze location competitiveness Data Source Trip ICAO OFOD DATA (2005) GNP Korea National Statistical Office (2005) Population Korea National Statistical Office (2005) Distance IATA WORLD WIDE FARES (2006) fare IATA WORLD WIDE FARES (2006) Data used to analyze location competitiveness are: Trips among cities (ICAO OFOD, 2005), GNP and population (Korea National Statistical Office, 2005), distance and fare (IATA, SITA, Passenger Air Tariff WORLD WIDE FARES, June, 2006) Analysis airports are Asia (6), Europe (9), America (7). Table 5 shows these.

Table 5 Analysis cities Continent city continent city Beijing Manchester Hong Kong Moscow Europe Singapore Munich Asia Taipei Paris Europe Tokyo Seoul Amsterdam Copenhagen Frankfurt Istanbul London America Chicago Los Angeles New York San Francisco Sao Paulo Toronto Vancouver Figure 2 shows the number of passengers in the air network in East Asia-East Asia. In East Asia-East Asia, Hong Kong had the most passengers at 1.8 million; Seoul had the next most passengers at 1.6 million. Other cities were Osaka, Taipei, and Beijing. The city with the fewest passengers was Beijing. In terms of number of trips between cities, the order was Hong Kong-Taipei, Seoul-Tokyo, and Bangkok-Singapore, at about 2.6 million. Figure 3 shows number of passengers in the air network in East Asia-Europe. In East Asia-Europe, Tokyo had the most passengers at 3 million, and Bangkok, Hong Kong, and Singapore had similar numbers with 2.2 to 2.6 million. In terms of number of trips between cities, the order was Singapore-London and Tokyo-London at about 0.95 million, and Tokyo-Moscow at about 0.9 million. Hong Kong 5868 5921 307 289 299 295 774 616 Osaka 2029 2114 157 160 365 385 727 736 Shanghai 2547 2465 6 9 7 2 8 5 Beijing 1604 1640 3 3 9 5 4 9 196 217 6 6 7 7 6 8 4 7 5 1 138 134 5 5 639 626 845 858 280 289 537 545 9 9 4 6 6 6 6 3 Manila 8 1 3 3 6 4 2225 2 7 2156 356 881 378 870 1092 1055 202 188 1300 1318 336 319 3 3 6 6 9 6 256 290 Singapore 4756 4810 331 328 10 11 3 3 8 7 9 9 5 5 3 5 4 2 462 460 Taipei 1939 1937 Bangkok 4716 4720 406 392 Kuala Lumpur 2381 2347 569 574 129 130 636 613 6 6 3 1 173 174 Seoul 5360 5294 1301 1275 6 6 9 7 251 249 Tokyo 4746 4767 8 8 4 4 Figure 2 Passenger air network in East Asia

Istanbul Copenhagen Frankfurt Manchester Moscow Amsterdam Munich London Paris Kuala Lumpur Hong Kong Seoul Shanghai Tokyo Bangkok Beijing Osaka Manila Singapore Taipei Figure 3 Passenger air network in East Asia-Europe Figure 4 shows the number of passengers in the air network in East Asia-America. In East Asia-America, Tokyo had the most passengers at 3.9 million. Hong Kong had 2.2 million passengers and Seoul had passengers 1 million. In terms of number of trips between cities, the order was Tokyo-Los Angeles with 1.05 million, Tokyo-New York with 0.95 million, and Tokyo-San Francisco with 7.9 million. Chicago 888 891 5 4 2 2 3 2 New York 1900 1916 12 0 459 464 73 74 273 276 190 183 Vancouver 1494 1512 627 624 Sao Paulo 228 274 4 3 2 8 7 8 7 7 2 1 6 8 2 9 1 0 1 3 Bangkok 42 38 L A 2283 2097 Beijing 402 445 1 4 4 6 147 159 8 8 8 9 2 2 0 0 7 3 Hong Kong 314 316 1159 1049 3 3 3 3 Osaka 133 133 226 0 San Francisco 1525 1647 2 2 7 6 6 0 2 2 2 1 542 514 2 2 3 2 5 4 9 9 2 0 Seoul 703 901 Manila 159 147 470 479 82 77 0 134 34 32 69 78 133 136 233 223 19 21 141 135 394 398 0 1 2 0 3 2 193 0 5 195 181 173 Shanghai 322 171 Singapore 279 273 Toronto 1659 1686 Tokyo 1956 1959 5 4 8 6 Taipei 82 77 Figure 4 Passenger air network in Asia-America

4.2.2. Analysis Result on Location Competitiveness Table 6 shows the results of analysis of GNP, population, and distance. The estimated parameters for GNP and population appear positive but distance appears negative. The parameter on distance appears to be the largest, over twice as large as GNP. This result is similar as Matsumoto s (2007), but the estimated parameter on population is insignificant. This is because air transport trips are not generated by large population cities, for example, Beijing, but by small cities, for example, Hong Kong and Singapore. Table 6 Result of GNP, population, distance constant GNP population distance R-square Parameter 12.319 0.300 0.025 0.752 t-value 12.153 9.034 0.627 10.701 0.307 Table 7 shows the results of analysis of monopoly power, population, and fare. The estimated parameter on GNP and monopoly power appears positive but population and fare appear negative. The estimated parameter on population is insignificant. The model that includes monopoly power shows a decreasing effect of fare and GNP. Additionally, the R-squared value was increasing. Table 7 Result of monopoly power, GNP, population, fare constant Monopoly GNP population fare R-square power Parameter 17.751 0.255 0.077-0.031 0.570 t-value 18.333 12.63 2.474-0.957 10.674 Table 8 Result of monopoly power, GNP, fare constant Monopoly GNP fare R-square power Parameter 17.089 0.252 0.079 0.566 t-value 25.223 12.632 2.560 10.632 0.591 0.593 Table 8 shows the results of analysis, except for population because it was insignificant. In the estimated result, it appears that the constant was the largest, because air trips were reflected by many factors: development of airplane technology, changes in air passenger types (business and leisure), changes in available operation capacities of airports, and transfer passengers. 4.3 Comparison of Hub-Airport Competitiveness in East Asia We also added dummy variables and analyzed the fare difference between markets. The results of fare and location difference analysis from the comparisons of the Incheon International Airport and Asian hub airports can be shown in Figure 5.

The results of analysis of fare and location competitiveness at Incheon airport is shown in Table 9. Figure 5 Fare and location that pass difference comparisons of Incheon International Airport and other airports Concerning the result of analysis of competition of fare and location in East Asia hub airports, Incheon airport is ranked 4th among East Asia hub airports. Tokyo and Singapore hold a dominant position in competition based on pricing and location over Seoul. Price competition in Beijing is better than Seoul and location competition in Singapore is also better. Thus, Incheon airport is falling behind the competition in East Asia. Incheon International Airport should be developed with strategic and differential support, infrastructure, investment planning, and improvements to operational organization as a coordinated effort to become competitive in making inroads into the air transportation markets of neighboring countries in East Asia. Table 9 Comparison of fare and location competitiveness at Incheon airport Ranking Fare Location 1 Singapore Tokyo 2 Beijing Singapore 3 Tokyo Hong Kong 4 Seoul Seoul 5 Taipei Beijing 6 Hong Kong Taipei 5. CONCLUSION We are emphasizing that Incheon Airport has an advantage when it comes to construction and operation over other competitor airports in East Asia, including a strong demand for air travel (35% of global market), potential energy resources, proximity to 43 cities with populations over 1 million within 3.5 flight hours, ability to operate 24 hours without noise, excellent expansion potential, and location within a special economic zone.

Although this research shows that Incheon airport is less competitive in fares and location, it does not mean that the analysis of the merits of Incheon airport was incorrect. Fare is determined by demand and the relative revenue of other airports revenue. Additionally, location must also be considered in the context of neighboring countries. Therefore, emphasizing prepared strategic alternatives to overcome and maximize these merits is crucial because the merits of Incheon are not special and in some cases, are worse than the comparable merits of competitor airports. Thus, schemes to hub Incheon airport must not involve a discussion of established levels of service and must not be satisfied with the status quo. Moreover, the government must establish available medium and long-term roadmaps with a clear understanding of Incheon s current competitive position as well as domestic and foreign responses. Strategic and political plans are required. REFERENCES You Gyoung Eui (1997) The Fare Policy for Competitiveness Guarantee Hub-Airport, JOURNAL OF SEJONG UNIVERSITY AEROSPACE INDUSTRY RESEARCH Vol 63, pp.30-49. Lee Sung Chang (1996) Marketing Strategy of Hub-Airports, Aviation Industry Policy Research, pp.67-80. Lee Hyung Yong, A Study of the Opportunities and Threats of Incheon International Airport as a Hub Airport, JOURNAL OF SEJONG UNIVERSITY AEROSPACE INDUSTRY RESEARCH Vol 63. An air Statistics (worldwide) (2005), KOREA CIVIL AVIATION DEVELOPMENT ASSOCIATION. Berry, S. (1990) Airport presence as product differentiation, American Economic Review, vol. 80, pp. 394-99. Berry, S., M. Carnall and P.T. Spiller (1996) Airline hubs: costs, markups and the implications of customer heterogeneity, NBER Working paper 5561. Borenstein, S. (1989) Hubs and high fares: dominance and market power in the U.S. airline industry, Rand Journal of Economics, vol. 20, pp. 344-65. Borenstein, S. (1991) The dominant firm advantage in multi-product industries : evidence from the U.S. airlines, Quarterly Journal of Economics, 106, pp. 1237-66. Brueckner, J.K. and P.T. Spiller (1994), Economies of traffic density in the deregulated airline industry, Journal of Law and Economics, vol. XXXVII, pp. 379-415. Caves, D.W., L.R. Christensen and M.W. Tretheway (1984) Economies of density versus economies of scale : why trunk and local service airline costs differ, Rand Journal of Economics, vol. 15(4), pp. 471-89. Gerber, Peter. (2002) Success factor for the privatization of airports-and airline perspective, Journal of Air Transport Management, 8, 26-36. Goetz, A.R. and C.J. Sutton (1997) The geography of deregulation in the U.S. airline industry, Annals of the Association of American Geographers, 87(2), pp. 238-63. Evans, W.N. and I.N. Kessides (1993) Localized market power in the U.S. airline industry, Review of Economics and Statistics, vol LXX(1), pp. 66-75. Hannan, T.H. (1992) The Functional Relationship between Prices and Market Concentration : the Case of the Banking Industry, in W. Weiss, D. B. Audretsch, and J. J. Sigfried (eds.)

Hannan, T.H. (1997) Market share inequality, the number of competitors and the HHI : an examination of bank pricing, Review of Industrial Organization, vol 12, pp. 23-35. Hansen, M. (1990), Airline competition in a hub-dominated environment: an application of noncooperative game theory, Transportation Research, vol 24B, pp. 27-43. Levine, M.E. (1987) Airline competition in deregulated markets: theory, firm strategy and public policy, Yale Journal on regulation, vol 4(2), pp.393-494. Marin. P.L. (1995) Competition in European aviation: pricing policy and market structure, The journal of industrial economics, vol XLIII (2), pp. 141-59. Mark G. Lijesen, Piet Rietveld, Peter Nijkamp. (2000) Do European Carriers dominate their Hubs?, Tinbergen Institute Discussion Paper Matsumoto, Hidenobu. (2003) Hubness of Asian major cities in terms of international air passenger and carfo flows, The Korean Transport Policy Review 10, pp.103-123. Matsumoto, Hidenobu. (2004) International urban system and air passenger and cargo flows: some calculation, Journal of Air Transport Management 10, pp239-247. Matsumoto, Hidenobu. (2006) International air network structure and air traffic density of world cities, Transportation Research Part E. Passenger Air Tariff (2005), IATA & SITA Robert Windle, Martin Dresner (1999) Competitive responses to low cost carrier entry, Transportation Research Part E 35, pp. 59-75. Taaffe, E.J. (1962) The urban hierarchy: an air passenger definition Economic Geography 38, 1-14. www.airport.or.kr www.moct.go.kr