Competition and Entry Deterrence in the International Flight Market: Evidence from Korean Regional Airport

Abstract

This study examines carriers’ entry of the Korean international routes at Daegu International Airport from 2010 to 2018. We investigate whether carriers’ entry into the international flight market is deterred by pre-existing carriers. The competition level of the six months before entry has a much higher contribution to the decision to launch route service. The lower the market concentration six months ago, the less likely a new carrier entry in the current period; competitive and increasingly crowded routes are hindered by entry barriers, deterring new entrants to the international flight market. The shorter the flight distance and the larger the air passenger market, the higher the probability of entering a new airline carrier. New entrants to non-stop point-to-point route structures are more likely on short-haul routes and routes operated by at least one low-cost carrier.

KEYWORDS:

• Airline competition
• market entry
• entry deterrence
• low cost carrier
• Korean aviation industry

Acknowledgement

The author is grateful to the Editor, Associate Editor, and an anonymous referee for their helpful comments.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Correction Statement

This article has been corrected with minor changes. These changes do not impact the academic content of the article.

Notes

1 Daegu Airport is a base airport of TWB, and Cheongju Airport is a base airport of ESR.

2 This seems to be the effect of reducing the curfew time. The coffee time, which was eight hours from 10 pm to 6 am the following day before 2014, was adjusted to five hours from midnight to 5 am the following day, increasing the number of available slots. The time difference has also been reduced, making it easier to launch new routes or increase flights.

3 Spring Airlines is an LCC operating from Shanghai Airport. Spring Airlines has successfully acquired relatively high prices and high seat occupancy without competing for price cuts by increasing flight frequencies or launching new flights mainly on routes to densely populated cities with high air demand.

4 Since entering the market, airfare levels have declined by 5.1% (Air China), 3.4% (China Eastern), and 6.2% (Hainan airlines), respectively.

5 All types of airlines have been able to operate large aircraft, and entry restrictions have been reduced as the age limit on aircraft has been removed since the revision of the Aviation Act in May 2008.

6 Most of the preceding studies on the U.S. airline industry use data provided as a sample by the United States DOT DB1A Database, which received 10% of the total itinerary from the airline.

7 Fu et al. (2015) analyze the effect of the new launch of an LCC with competitiveness on the profitability of each FSC’s inland route operation.

8 For routes that are already saturated, it may be challenging to get slots for new flights, and for routes with fierce competition, it may be challenging to get additional flights or departures during prime time. Unlike airports that have reached a saturation point, such as Jeju International Airport or Gimhae International Airport, Daegu Airport has relatively ample airport capacity. Therefore, it is necessary to analyze the airport with slot saturation to analyze the determinants of route entry.

9 In 2013, the Korea-Japan routes suffered a blow due to the strained relationship between Korea and Japan and the weak yen. Due to the deterioration of relations with China over the Terminal High Altitude Area Defense (THAAD) deployment, the number of flight routes between Korea and China was reduced in 2016. The boom in the aviation industry due to the increase in air travel demand in 2017 did not last long, and boycotts against Japanese goods took a toll on the Korea-Japan routes in the summer of 2019.

10 The assumptions for the model in which entry is determined based on the air network structure and route characteristics are as follows (Boguslaski et al., 2004; Oliveria, 2008; Homsombat et al., 2011; Fu et al., 2015a; Wang et al., 2017). An airline enters the market only when revenue from route operation is generated. Therefore, it is possible to infer the operating profit of air transport services, an unobservable latent variable from the airline's entry decision.

11 Since the international air passenger market is vulnerable to changes in global external economic conditions, including the rise and fall of the won, it is necessary to consider external economic conditions, unlike the domestic air passenger market. In the case of Japanese routes, the hardship phase due to the worsening of relations between Korea and Japan leads to a decrease in travel demand, and the relationship between Korea and China over the deployment of THAAD on the Korean Peninsula leads to suspension of routes between Korea and China, leading to the deterioration of route operation profits.

12 The profitability of air operations is also a factor influencing the decision to enter new routes. An airline's unit yield indicates unit price (income per 1 RPK) by dividing operating revenue by paid passenger kilometers (RPK, paid passenger x transportation distance). However, like the operating profit rate indicator, the unit revenue indicator cannot be used as an explanatory variable in the regression model because operating profit data by airline, month, and the route cannot be obtained. Therefore, we define the passenger load factor variable for each route to be used as an indicator of route operation profitability.

13 Year 2019 is excluded, as the recession in airline industry is caused by the COVID-19 pandemic. The flight frequency scheduled in December 2019 was 1110, which was decreased by 24.6% compared to December 2018. The impact of COVID-19 has been huge to the aviation industry, causing an 89.9% decrease in year 2020’s flight frequency departing from Daegu International Airport compared to 2019 levels. The operated international flights at Daegu airport dropped by 96.3% year-on-year in 2021 due to ongoing COVID-19 pandemic. A nearly empty Daegu International Airport operated a total of 68 international flights in 2021, which was unprecedented collapse.

14 The Associate Editor raises a concern for potential omitted variable bias. However, no disaggregated data on airline carrier’s revenue and cost at the airport-month level is available. Given the dearth of either specific route- or airport-level data, analysis using aggregate data from all domestic routes have received much attention. Specifically, this study aims to build data on Korean international routes at the regional airport level. As suggested by previous studies, the passenger load factor as an indicator for airline operation profitability is also used to minimize the limitations caused by the availability of data. The load factor for each airport is calculated the number of passengers divided by total seats supplied. Please see footnote 12 for a detailed explanation.

15 We also consider the total number of seats supplied to be an airport density variable. However, the pairwise correlation between the average load factor and the total number of seats supplied is 0.6384 and statistically significant at the 5% level. The existence of the large pairwise correlations may be sufficient for detecting multicollinearity. Therefore, we proceed with the chosen variables.

16 Gimpo International Airport, Jeju International Airport, and Gimhae International Airport, which are major domestic airports, are operated from 11 pm to 6 am the next day, the curfew time. In other regional airports, a 24-hour operation is possible, or airport operating hours have been extended in some cases. The curfew time is not set at Cheongju International Airport, and the curfew time (10 pm to 5 am the next day), which was 7 h at Daegu International Airport, has changed to five hours (midnight to 5 am the next day) since July 2014. After being reduced to five hours, international routes were expanded, and additional flights were assigned. We control for the effect of curfew time using time-fixed effects.

17 The tests of the hypothesis that the dummy variables have no collective influence on the dependent variable at the 5% significance level are conducted for year dummy variables and month dummy variables, respectively. According to the F-statistics and the p values associated with the tests of overall significance, both the year and month dummy variables are collectively significant.

Additional information

Notes on contributors

Joo Yeon Sun

Joo Yeon Sun is an assistant professor of department of economics at Dankook University. Her primary focus is on product differentiation and airline markets, having published works on how policy changes can alter demand and firms' responding strategies through non-price competition. Her research also encompasses topics such as international trade policy and sports economics.