Airline industry: Effects of alliance formation on prices and welfare

Airline industry: Effects of alliance formation on prices and welfare

Ivan Vukcevic

Student number: 0583642

11 April 2010

Bachelor thesis Prof. dr. Jeroen Hinloopen

University of Amsterdam Faculty of Economics and Business

Airline industry: Effects of alliance formation on prices and welfare

2

Table of Contents I. Introduction ......................................................................................................... 3  II. Industry characteristics....................................................................................... 4  III. Models............................................................................................................... 7  IV. Discussion....................................................................................................... 12  V. Empirics ........................................................................................................... 14  VI. Conclusion ...................................................................................................... 17  Bibliography ......................................................................................................... 19 

Airline industry: Effects of alliance formation on prices and welfare

3

I. Introduction The airline industry has traditionally been characterised by strict government regulation, consisting of price control and entry barriers. Airlines enjoyed exclusive rights given by their own governments to operate on a certain market under strictly defined rules. However, once the process of deregulation started, they quickly found themselves competing intensively for every passenger. The increased competition has forced a considerable number of companies out of the market, and competitive pressure on the remaining ones increased substantially. Carriers were forced to search for ways to dampen the impact of these developments, primarily by reducing the level of competition on the market. Forming strategic alliances is one way to achieve this goal. However, while alliance formations could be an efficient way to soften the effects of market liberalization from the airlines point of view, they could also hamper the competition and have negative impact on prices and welfare, from the consumers’ point of view. Since these formations could be seen as types of mergers to a certain extent, it is very feasible to assume that lesser number of firms in the market, or higher number of “colluded” firms, could increase prices and decrease welfare. Therefore, it is important to analyze the effects of airline alliances and assess whether such formations raise competition concerns. This paper analyses the aforementioned problem. It starts by presenting other industry aspects on which alliances could have an impact. Then, the two models that deal with price effects of alliance formations are presented and discussed. Finally, the paper concludes by testing whether the theoretical predictions match up with the empirical evidence. The outline of the paper is as following. In order to get familiar with the topic of airline alliances and the industry in general the next section presents an overview of different effects that alliances can have; such as increasing barriers to entry and cost reduction. In the third section the two models and their results will be presented. The fourth section will present a discussion of the results and highlight few shortcomings in the analysis. In the fifth section empirical evidence is presented and discussed. Finally, the sixth section concludes.

Airline industry: Effects of alliance formation on prices and welfare

4

II. Industry characteristics In this section the literature review will be presented. In order to get a better understating of what airline alliances are and what their significance in the airline business is, one should be aware of the different ways in which they could influence the industry. In this part few articles that analyze influence on different industry aspects, such as cost reduction and level of competition, are presented. Since the deregulation of the airline industry had started and the conditions for alliance formation were created, economists have tried to explain different effects that the alliances have on the industry. Much research has been conducted, ranging from rather simplified models to econometric analyses, in order to determine the effects that these formations have on welfare, prices, entry accommodation and other economically interesting aspects. In their publications Nolan, Ritchie and Rowcroft (2001) and Lin (2005) use a simple simulation airline network model to show that effects of alliances on welfare and entry deterrence are substantial. On the other hand, the results presented by Goh and Yong (2006) in their econometric analyses of alliance formation effects on cost structure suggest that, although statistically significant, their actual impact is rather small. Finally, some research results imply that the effects of alliances are ambiguous. According to Chen and Gayle (2007) competitive effects of alliances depend on market structure and companies profiles. Different outcomes of these papers can be used to better understand various effects that alliances may have. Deregulation of the airline business certainly is one of the finest examples of industry liberalization in recent history. Although the process appears to be rather constant and with only few shocks, the final shape of the market is still far away. Recent changes in the industry’s economic and regulatory environment have induced formation of supranational alliances that will have a major influence on the form of the industry in the future. Probably the most obvious effect of alliance formation on the airline industry is in the cost structure area. Airline carriers cost structures are very complex, and are characterized by high fixed and relatively low marginal costs. Nolan, Ritchie & Rowcroft (2001) mention number of different costs that airline companies incur, such as rent or lease of planes

Airline industry: Effects of alliance formation on prices and welfare

5

(companies rarely buy planes), buying airport slots, arranging ground handling of aircraft and employee salaries (pp. 121-122). These expenses are very high individually, and together they make the airline industry difficult to enter. However, by engaging in alliances companies are able to agree on common materials, on-board items and aircraft used, all of which will decrease production costs and reduce overall costs of purchasing. Furthermore, once in an alliance, airlines would be able to share gate space at international airports. This should lead to large cost reduction because one set of ground crew and equipment is sufficient to facilitate all member flights (pp. 119-121). Membership can thus decrease total costs, make companies more competitive and thereby have a positive impact on their profits. However, in empirical research on the impact of alliances on airline coststructure Goh and Yong (2006) found somewhat contradictory results. They use a truncated third-order translog model in order to estimate the effects of alliances on cost structure. The subject of the analysis are code-share alliances, the most common form of alliances in which member airlines are allowed to use each other's designator code and, usually, also synchronize flight schedules in order to create the feeling that the services are operated by the same airline. Their stand point is that by becoming a member of an alliance airlines could gain important advantages such as allowing passengers to avoid inter-line transfers, obtaining a marketing edge since computer reservation systems usually display flights without inter-line transfers first and achieving improved efficiency by rationalizing network structures. Due to alliances member airlines are able to avoid excesses and duplications and to use factor inputs in a more efficient way (p. 836). Nevertheless, they conclude that, although alliance formation does have a statistically significant impact on costs regardless of the number of alliance partners, the impact of these effects appears to be small and unimportant (p. 854). Another important aspect of alliance formation is that it can serve as a tool for entry deterrence. Lin (2005) argues that although entrant entering the market can obtain positive profits in the first period, because of the alliances between previously present airline companies those profits may decline and turn negative

Airline industry: Effects of alliance formation on prices and welfare

6

in subsequent periods. That is, existing airlines could use alliances as a credible threat of entry deterrence. It appears safe to say that such a role of the alliances should be high on the agenda when governments encourage competition in liberalized airline markets (pp. 7-8) The effects of alliances go beyond internal cost structure changes and possibility of raising entry barriers. They also have a major impact on the equilibrium position; in other words on prices and competitive structure. In another analysis of the code-share alliances, Chen and Gayle (2007) found that its impact on prices and competition depends on features of potential partners and the existence of competition to become a partner. Putting it differently, the characteristics of airlines bidding to become an alliance members and the likelihood of that actually happening will have a significant impact on the postalliance market structure. Depending on the level and form of competition the alliance might lead to (i) marginal cost pricing, i.e. lowest prices and highest consumer surplus; (ii) lower final prices for consumers although not as low as in the marginal cost pricing case; and (iii) higher final prices for consumers (p. 1057). In other words, the alliance formation will have an effect on prices but the nature of that effect is ambiguous; it depends on the particular pre-alliance market structure. So far we have seen different effects that airline alliance formation could have on the industry. Many different research results imply that the economical game played in this industry is rather difficult to assess. However, the focus in this paper will mostly be on the impact that alliances could have on prices in the market. More precisely, I will present and discuss results of two theoretical models, by Brueckner (2001) and Brueckner & Whalen (2000). They predict two types of effects that alliances could have: positive effects, arising because of cooperative pricing and realizing the effects that price increase may have on total revenues in the market, and negative effects that reduced competition and increased possibility of collusion are likely to create. To be more precise, the first model, based on a hub-and-spoke network with two hubs and two airlines, predicts that: i) there will be a price decrease in the interline market because

Airline industry: Effects of alliance formation on prices and welfare

7

companies will realise that increasing prices will negatively affect their alliance partner’s traffic levels, and, in turn, their total revenue in the market; ii) there will be an increase in the interhub market prices since instead of two, virtually only one company will be operating. The total effect is ambiguous. The second model, based on a hub-and-spoke network with one hub and four airlines, has no interhub market, and its prediction for the prices in the interline market is similar. However, the models somewhat differ in the assumptions they make, and in the network structure. Therefore, I will focus on, firstly, describing the models and identifying the differences, secondly, discussing the effects that those differences are expected to have on the results, and finally, analyzing what implications could these results together have on the view of airline alliances as problematic.

III. Models In order to get more familiar with how airline formations affect price level, the following section will analyze two theoretical models. Both of them deal with equilibrium positions with and without alliances and analyze what is expected to happen with prices once the alliances are introduced. In the first paper, Brueckner (2001) develops a hub-and-spoke model in which two airlines operate. The structure of the model can be seen in the following graph:

Network structure in Brueckner (2001)

Airline industry: Effects of alliance formation on prices and welfare

8

Airline 1, marked with the solid line in the graph, operates on two spokes, AH and BH, and on the route between two hub airports, H and K. Similarly, airline 2, marked with the dotted line, operates on spokes KD and KE, and on the same inter-hub route, HK. Such a setting gives three domestic city-pairs for each airline, and additional three international city-pairs at which the passenger does not need to use both airlines. It is assumed that such interline trips do not happen on these routes; one reason for such an assumption could be that the price of interline trip is likely to be higher than the price one airline could provide in any circumstances. Furthermore, the structure gives four city-pairs where an interline trip is necessary: AD, BD, AE and BE. The demand functions are assumed to be equal for every city-pair market. In addition, the cost functions are dependent on traffic density and are also assumed to be the same on every route. It should be noted that both airlines have to identify the arbitrage possibility when setting their fares. For instance, fare on DE route has to be lower than the sum of the fares on DK and KE routes. If not, passengers traveling from point D to point E would buy two round tickets, on routes DK and KE, instead of one “through” ticket on route DE. Under such fares no one would be buying “through” tickets, which would render the company’s pricing policy invalid. However, there might be no need for imposing the constraints since the equilibrium fares may satisfy them automatically. In the paper, Brueckner analyses two equilibria, with and without an alliance. In order to stay within the scope of this paper I will not go into the mathematical analysis of these equilibria. Instead, I shall focus on the ending results and their intuitive explanation. To get to the equilibrium solution, author equalizes marginal revenue to marginal cost in every different market i.e. every city-pair. Having set all demand and cost functions to be the same for every citypair simplifies the analysis. Furthermore, since both companies face the same demand functions in their “domestic” markets, as well as in the interline markets, only company 1’s maximization problem should be analyzed; company 2’s

Airline industry: Effects of alliance formation on prices and welfare

9

quantities and prices will automatically be obtained since the solution is symmetric. With an alliance in place, companies agree to split the revenue in markets in which a passenger needs to travel on both carriers in order to make his trip. Also, in the interhub market both companies will provide services to the half of the total number of passengers. This is reflected in the total revenue function which will consist of full revenue in the markets operated only by that company, and half of the total revenue in the interline markets; the total cost function is derived in a similar fashion. On the other hand, without an alliance, companies will face different maximization problems in the interhub and the interline markets, comparing to the alliance situation. Since being competitors, in the interhub route they will individually determine their traffic quantities while assuming other company’s choice as fixed, whereas in the interline market, since the quantities are by definition set to be the half of the total number of passengers, they will each set the price for its part of the interline trip. Based on the model the author derives a couple of conclusions about the effects that an alliance could have on prices and traffic levels. He argues that with the alliance present the number of passengers in the interhub market should decrease while in the interline markets it should increase (in all other city pairs there will be no change in prices or traffic levels). This, in turn, implies that the prices in these markets will move in the opposite directions; that is, they will increase in the interhub market and decrease in the interline markets. However, it should be noted that the different magnitudes of traffic levels in different markets will affect, and possibly overturn, the abovementioned results. In order to avoid this problem, constant returns to scale are introduced: under that assumption the traffic levels will have no impact on the results. The intuition behind these results is fairly simple. With an alliance, the companies will not be competing, but colluding, which will increase the prices in the interhub market. Furthermore, without an alliance the companies would be setting the prices for their portion of the interline trip ignoring the negative effects on the other company’s profits. However, when in the alliance, they would set the

Airline industry: Effects of alliance formation on prices and welfare

10

prices cooperatively; that is, they would choose a full interline trip price in a way that will maximize the entire interline profits. In other words, both companies will be trying to maximize their joint profits instead of only its own. Finally, in order to illustrate these results, author presents and solves a linear demand case under the assumption of constant returns to scale. In the second article, Brueckner & Whalen (2000) also develop a hub-andspoke network. However, unlike in the model discussed in the previous section where two hubs were present, this model has only one. Furthermore, the model

differs in the number of airlines in the market. Its structure can be seen in the following graph:

Network structure in Brueckner & Whalen (2000)

Airlines 1 and 2, presented with the dotted line, operate on two spokes, AH and BH. Similarly, airlines 3 and 4, presented with the solid line, operate on the other two spokes, DH and EH. This setting gives three domestic city-pairs to both pairs

Airline industry: Effects of alliance formation on prices and welfare

11

of companies, namely, AH, BH and AB to airlines 1 and 2, and DH, EH, DE to airlines 3 and 4. Additionally, there are four city pairs that require interline trips, AE, BE, AD and BD. Unlike in the previous model, there is no inter-hub route. Also, for the sake of simplicity, it is assumed that in the interline market company 1 will always form a pair with company 3, and company 2 will always form a pair with company 4. The demand functions are the same for every city-pair, similar to the model above. Same holds for the cost functions. An important difference comparing to the model above is that now authors assume brand loyalty. In other words, passengers are willing to fly with one of the companies even tough its fares are higher than the other fares in the market. This allows airlines to charge higher prices than its competitors and still keep some of its passengers. With four companies in the market there are three different equilibria possible. Two symmetric ones, where every company either forms an alliance or remains operating alone, and one asymmetric equilibrium, where two companies operate as an alliance and the other two remain on its own. The results are summarized below where PS denotes prices obtained in the symmetric equilibrium and PA in the asymmetric one. The prices in the domestic markets are presented in the first row; international market is on hand in the second: PA alliance < PS < PA nonalliance PA alliance < PA nonalliance < PS alliance < PS nonalliance When comparing the two symmetric equilibria, it is concluded that companies will set lower fares when engaged in an alliance. The intuition is similar as it was in the previous model: without an alliance an increase in, for example, company 2’s sub-fares in the interline trip will decrease the amount of traffic in the market putting a downward pressure on company 4’s revenues. On the other hand, when engaged in an alliance company 2 will try to avoid this negative externality since the total revenue in the market is to be shared between the companies. The most striking result, however, is obtained when asymmetric equilibrium is analyzed. It

Airline industry: Effects of alliance formation on prices and welfare

12

is concluded that companies engaged in an alliance will charge lower domestic fares than their no-alliance counterparts. Furthermore, the domestic prices under symmetric equilibria discussed above will lie somewhere in between; that is, alliance companies will set domestic fares lower than the domestic fares in the symmetric equilibria, whereas companies not in an alliance will set them higher. It is also concluded that in the interline market alliance partners will again charge lower fares than the companies not in an alliance. However, now, both of these prices will lie below the prices set in the symmetric solutions. To summarize, both of the models predict lowering prices in the interline markets. Furthermore, the first model also predicts price increase in the interhub market with the total effect somewhat ambiguous and dependent on various things, such as traffic levels in the particular case. Next to the differences in the network structures and number of airlines in the market, the second model makes an assumption of brand loyalty. However, it appears that these differences have no effect on the final results, something that will be analyzed in the following section. In this section we have seen what effects, according to theoretical models, an airline alliance could have on prices in the market. Next section will present a discussion about the two models presented above, and will highlight potential problems that they could face. Furthermore, it will question whether airline alliances present any danger to welfare in the first place.

IV. Discussion In the previous section we have seen the basic characteristics of the models, and their results. Now I shall focus on analyzing some of the underlying assumptions and the results of the models. Furthermore, the next section will point out few implications that the abovementioned results, combined with a number of other factors, could have on the topic of airline alliances. The two models discussed are somewhat different in the assumptions they make. Their network structure is different, the number of companies is not the same, and in the second model an extra assumption of brand loyalty is made. Still, all these differences appear to have no effect on the final results as both models

Airline industry: Effects of alliance formation on prices and welfare

13

have very similar predictions. However, it seems that both of them lack to properly address the interhub market. In the first model there are only two companies implying that after an alliance is formed there will be a virtual monopolist in the market, whereas the second model, which indeed provides plausible competition even in the alliance case, has no interhub market in the first place. Even though the second article mentions that, with four companies in the market, an analysis of a more complex network structure, which would include interhub market, would be virtually impossible, one could argue that with current settings both models simply do not make a proper assessment of a gateway-togateway market. Providing the four companies model with an interhub market while simplifying the mathematical part to keep the analysis manageable could be a useful addition to the airline alliances research. Both models predict price movements in the same direction; that is, in the interline market alliance formation will push the prices down, even when there are only two companies present in the market. However, it is not completely clear, especially in the two-companies scenario, where will this decrease come from. It is argued that when an alliance is formed companies will decrease prices for their part of the interline trip which will increase the total profits. But this means that, before alliance formation, total interline trip prices were higher than potential monopoly prices; lower prices are deemed to increase profits which is exactly what a monopolist would have done. Even though this issue is not explicitly analyzed in the paper, bearing in mind demand characteristics in the airline industry (rather inelastic etc.) it is unlikely that companies’ non-cooperative pricing would have pushed the prices above the potential monopoly prices. Putting it differently, it is not clear why would the companies decrease the prices after alliance formation and not simply take the benefits from the lower overall costs for themselves; unless the initial prices were above the potentially most profitable ones, which is a very strong assumption to make. Nevertheless, the fact that, regardless of their differences, both models predict virtually the same, generally positive, alliance-influenced price movements speaks in favor of airline alliances. Furthermore, both papers go on to

Airline industry: Effects of alliance formation on prices and welfare

14

empirically show that airline alliance formation should increase the total welfare since the positive impacts should prevail over the negative ones. And when analyzing alliances one must not oversee other important aspects of the industry. It has been argued that the industry, throughout its history, has made a cumulative loss implying that companies would not have been able to survive without state subsidies (the economical reason for the subsidies could be seen in large positive externalities related to the industry). The ongoing liberalization, which has drastically changed the rules in the industry, additionally complicates the situation for airlines. In the last two decades the industry has witnessed sharp price reduction, both in real and nominal terms. Even some of the major companies, such as Swissair and Pan Am, were not able to cope with the changes and were forced bankrupt. Furthermore, introduction of low-cost companies and the constant danger of their entrance into any market mean that full-fare airlines need to closely watch the prices they set if they want to avoid competing with them (it should be noted that only full fare companies are willing to become members of alliances, low-cost airlines have a different business model in which alliance membership is not a feasible option). Airlines had to develop various ways to cope with all the difficulties that industry liberalization has brought up on them. Forming an alliance appears to be one of the most effective ones. It seems that the idea behind joining an alliance is not profit increase at the expense of consumers; instead, companies form alliances in order to better cope with the changing environment in which they found themselves. In other words, it appears that alliance formation simply presents a next step in, or airlines’ response to, industry liberalization.

V. Empirics So far we have seen what effects an alliance formation could have on prices and welfare according to two theoretical models. Even tough based on somewhat different assumptions, both models predicted very similar results. However, before making any conclusions, it is important to test whether the robust theoretical results are similar to what we observe in real life. Therefore, in the following section the results of empirical research on couple of alliance

Airline industry: Effects of alliance formation on prices and welfare

15

formations are presented and put up against aforementioned theoretical predictions. In their paper, Park & Zhang (2000) empirically analyze the impact that four alliance formations had on prices, welfare and total number of passengers. The

analyzed

alliances

are

British

Airways/USAir,

KLM/Northwest,

Lufthansa/United Airlines and Delta/Sabena/Swissair. The authors use panel data in the 1990-1994 period and estimate the effects that each of the four alliances had, as well as what would have happened if alliances were not formed. The analysis is based on estimating two equations, demand equation and price equation. Since alliances could have an impact on both of them, and number of passengers and fares are interdependent, the equations are analyzed “as a system of demand and price equations, so as to examine the effects of alliances on market outcome and consumer surplus” (2000. p. 371). After setting up the model and successfully addressing few econometric problems, most notably the ones related to specification, estimation method and error treatment, the paper proceeds by introducing data and variables. 19 North American and 12 European gateways were identified, yielding a total of 21 routes on which the alliances were formed; 4 of those are excluded due to lack of data or a small sample, leaving 17 routes in the analysis. The coefficients in the demand equation showed that 3 out of 4 alliance formations induced an increase in aggregate demand. More precisely, British Airways/USAir, KLM/Northwest and Lufthansa/United Airlines all experienced an increase in aggregate demand by 13%, 35% and 13%, respectively. On the other hand, Delta/Sabena/Swissair alliance experienced a decrease in the aggregate demand by 25%. The price equation estimates show that both, Delta/Sabena/Swissair and KLM/Northwest have decreased the prices on alliance routes by 19% and 22%, respectively. Further analysis also revealed that even though the mark up ratio increased in both cases implying that companies were able to earn more, the decrease in marginal costs, mostly due to increased efficiency, was dominant. In other words, even tough the companies were able to reap higher profits by increasing the markup ratio, efficiency increase was high

Airline industry: Effects of alliance formation on prices and welfare

16

enough to bring the final price down. The price effects of the remaining two alliances appeared to be statistically insignificant, most likely because airlines did not integrate operations sufficiently, or because the alliance was formed recently and the data set was too small. In order to estimate the effects on equilibrium price and number of passengers, authors compare the alliance equilibrium with the hypothetical equilibrium where no alliance occurred. The results show that in markets where British Airways/USAir and Lufthansa/United Airlines formed an alliance the number of passengers increased by 65,626 and 25,616 on an annual base, respectively. For the other two alliances both changes, in price and quantity are estimated. KLM/Northwest alliance induced an increase in number of passengers by 46,866, whereas the price was reduced by 99$ and Delta/Sabena/Swissair alliance impact was reflected in a decrease in total number of passengers by 28,863 and decrease in fares by 90$. Joint effect of all four alliances is estimated to be an increase in number of passengers by 35,998 and a reduction in price by 41$. Finally, 3 of 4 alliances are estimated to have a positive impact on consumer surplus. British Airways/USAir, KLM/Northwest and Lufthansa/United Airlines increased the consumer surplus in their respective markets by 223 million $, 193 million $ and 122 million $, respectively, whereas Delta/Sabena/Swissair alliance is estimated to negatively impact the consumer surplus and decrease it by 120 million $. The empirical research shows that, overall, alliance formations should have a positive impact on prices and consumer surplus. More precisely, it seems that a complementary alliance, such as KLM/Northwest or Lufthansa/United Airlines, is likely to increase both, aggregate output and consumer surplus, whereas a parallel alliance, such as Delta/Sabina/Swissair, will most likely have a negative effect on surplus. 3 out of 4 alliances in question had a positive effect on passenger volume and prices. It should be noted that analysis does not neglect a potential increase in market power due to alliance formation. On the contrary, it is acknowledged that alliance formation is bound to increase partners’ market share and market power. However, the potential negative effects that market power

Airline industry: Effects of alliance formation on prices and welfare

17

could stimulate are simply dominated by the increased efficiency and cost reduction. Therefore, the final effect of alliance formation is estimated to be a reduction in prices and an increase in consumer surplus, something that above mentioned theoretical models have also predicted.

VI. Conclusion Since the deregulation process in the airline industry started, companies have faced an ever-increasing level of competition. They have been trying to adapt to this new market structure in every way. Furthermore, they have been trying to find ways to soften the competition between them and make them more cooperative. Engaging in an airline alliance appears to be one of the most common ways to achieve this goal. However, since alliance formation is bound to decrease competition and could induce collusive behavior in the market it could present a threat to the welfare and consumer surplus and, therefore, needs to be carefully analyzed. The analysis of the models reveals that, even though they do face certain problems, both models predict the same impact that alliances could have on prices. Namely, interhub market prices should increase, whereas interline markets will experience price reduction. The total effects appear to be somewhat ambiguous; however, they are more likely to be welfare increasing. Furthermore, the empirical research, based on four major alliance formations, shows that alliances usually bring price reduction, increase in number of passengers and in total surplus. Even tough it is noted that such formations will increase airlines’ market power, the efficiency increase is deemed to dominate any negative effects that might rise from higher market concentration. Combining the theoretical and empirical results with the other industry aspects, such as industry’s historical (lack of) profitability and recent bankruptcies of major companies, implies that airline alliances should not present a threat to prices and welfare, but are simply a next step in the airline industry evolution. They present a way in which full fare carriers can increase their competitiveness and make sure the liberalization will not force them out of the market.

Airline industry: Effects of alliance formation on prices and welfare

18

Since the first airline alliance has been created competition authorities have been keeping a close eye on them. And rightfully so one would say, since any form of cooperative action could be a potential threat to competition. However, due to industry specific characteristics and increased market liberalization, airline alliances provide a way of staying in business for many full fare operators. If not been allowed to engage in an alliance some companies may have been forced out of the market and, due to the specific industry position and its connections to other branches of economy, a shut down airline could have severe effects on the entire economy. The authorities indeed should keep an eye on the developments in the airline industry and, in particular, on airline alliances. However, at this point in liberalization process, it seems that damages from prohibiting alliances would be greater than any potential negative effects an alliance could have.

Airline industry: Effects of alliance formation on prices and welfare

19

Bibliography Brueckner, J. K. (2001). The economics of international codesharing: an analysis of airline alliances. International Journal of Industrial Organization, 19, 1457-1498. Brueckner, J. K. & Whalen, W. T. (2000). The price effects of international airline alliances. Journal of Law and Economics, 18, 503-545. Chen, Y. & Gayle, P. G. (2007). Vertical Contracting between Airlines: An Equilibrium Analysis of Codeshare Alliances. International Journal of Industrial Organization, 25, 1046-1060. Goh, M. & Yong, J. (2006). Impacts of Code-Share Alliances on Airline Cost Structure: A Truncated Third-Order Translog Estimation. International Journal of Industrial Organization, 24, 835-866. Lin, M. H. (2005). Alliances and Entry in a Simple Airline Network. Economics Bulletin, 12, 1-11. Nolan, J. F, Ritchie, P. C. & Rowcroft, J. E. (2001). Open Skies and Open Gates. Transportation, 28, 119-135. Park, J. H. & Zhang, A. (2000). An Empirical Analysis of Global Airline Alliances: Cases in North Atlantic Markets. Review of Industrial Organization, 16, 367 – 383.