46 31 Dns llyjatf - Konkurrensverket · factor, the distribution channel, also can prohibit new...

46 . 31Dns llyjatf

fc/.Jt,

Sales Channels - A Barrier to Entry in the Airline Industry

byMaria Dembrower and Daniel Grenblad,

Stockholm School of Economics, Stockholm1

February 2003 (Work in Progress)

Accepted after Review tothe 7th Air Transport Research Society World Conference (ATRS),

Toulouse, France. July 10-12, 2003.

Keywords: Airline Distribution, Competition, CRS, e-distribution, Entry Barriers,Industry Growth, Low Cost Airlines, Scenarios for the Future, Travel Agents.

1 Maria Dembrower. M Se. Phone: +46-8-736 97 37, e-mail: maria.dembro\verfrj),hhs.se.Daniel Grenblad, Ph D Candidate. Phone: +46-8-736 95 48, e-mail: daniel.grenbladCg.hhs.se.Both al Stockholm School of Economics, Center for Marketing, Distribution and Industry Dynamics,PO Box 6501, SE-113 83 Stockholm, Sweden.Authors are listed in alphabetical order to signify equal contribution.

Dembrower, Maria. & Grenblad, Daniel (2003).

Abstract

Growth in the Airline industry has taken place in the low cost segment. With their lowprices the low cost Airlines have attracted new kinds of customers and travels, creatednew märket segments and thus expanding the total potential märket. This industry growthis hindered by several factors; flight rights, slöt rights, and environmental factors (likewar threats, slow economy, and established alliances). This paper examines if anotherfactor, the distribution channel, also can prohibit new entries into the märket. The mainresearch question in this paper will be to study the distribution strategies chosen by newAirlines, given the two options; the Travel Agent and Internet. Secondly, we will discussif the distribution of ticket sales could constitute an entrance problem.

The CRS (Computer Reservation System) provides centralized information and bookingfacilities for air travel and other products, thus acting as aggregators. The CRS has longhad an important position in the distribution channel serving the travel agents. However,since the CRS required both infrastructure investments and speciali/ed knowledge, thetravel agent has retained an important position in the distribution channel. As the Internetbecame commercialized in the second half of the 90's, the end-customers gained directaccess to various ways of booking and increased their power. For the low cost Airlinesthat emerged, this meant that they had an option to either distribute through thetraditional channel or through the Internet channel. The Traditional Travel Agent stillaccounts for a large part of total sales through established relationships with customers.The Internet, on the other hand, has been a challenge to traditional ways of distribution.

Empirical data is gathered from Airlines operating in Sweden. Supplemental data comesfrom the largest travel agents in Sweden. The interviews were conducted in 2003.Analysis will be made using Network Theory, where actors in the network contributewith resources of different relative strengths and create dependencies through linkages(Håkansson & Snehota, 1995).

Our preliminary findings show that there are ät least five different ways Airlines canposition themselves in the distribution channel network; through Traditional Agent,Agent Web, CRS Web, Proprietary Web, or Collaborative Web (these will be furtherexplored in the paper). Which alternative(s) the new entrant should choose is not selfevident. Once chosen, there seems to be an entrance problem when the new Airline triesto get a position, since it does not yet have a given place in the distribution network.Thus, there may be an entry barrier with the selling of tickets and to attain a position inthe industry, i.e. to reach the customers and create revenue.


Table of Contents

1. Introduction l2. Method l3. Network Actors 34. Theoretical framework 5

4.1. Which distribution strategy does each actor have? 64.1.1. Airline - Empirical Findings and Analysis on Strategy 64. l .2. CRS - Empirical Findings and Analysis on Strategy 74.1.3. Traditional Travel Agents- Empirical Findings and Analysis on Strategy 74. l .4. Web- Empirical Findings and Analysis on Strategy 8

5. Sales Network Link Analysis 96. Implications of Links 12

6.1. Network Changes 136.2. Can ticket sales itself constitute an entrance problem (new Airline)? 16

6.2.1. Airline -Empirical Findings and Analysis on Barriers 166.2.2. CRS - Empirical Findings and Analysis on Barriers 176.2.3. Traditional Travel Agents - Empirical Findings and Analysis on Barriers

186.2.4. Web - Empirical Findings and Analysis on Barriers 19

7. Discussion 208. Conclusion 239. Future Research 2410. References 25


Table of Figures

Figure l : Interviews made with the case companies during April 2003 (Owndevelopment) 2

Figure 2: The Sales Channel graph, showing alternatives that Airlines can use. Airline(A) to Airline (D) depicts typical way of using the Traditional Travel Agents andWeb alternatives. (Own development) 4

Figure 3: The Sales Network Link graph, showing relationships between actors. Strongerlinks are displayed with thicker lines. (Own development) 11

Figure 4: Actor changes in the network relationship perspective. (Own development)... 15


1. Introduction

Deregulation of the Airline industry in Sweden starting July lsl in 1992 has eased theestablishment of new Airlines such as low cost carriers. The increased use of Internet hasmade it easier for new actors to fmd distribution channels and to sell their tickets överInternet. But stil l the Traditional Travel Agents play an important role on the märketbecause of their long term relationships with companies which can offer them loyalbusiness customers. The fäet that the new carriers don't have a well known name anddon't have any established relationships with the Traditional Travel Agents could make ithärd for the new actors to grow.

The separate areas of changes described above makes it interesting to examine theintersection of those, namely the combination of new Airlines and new ways ofdistribution in this industry. The purpose of this study is to examine if ticket sales itselfcould constitute an entrance problem. To do this, we first analyze which distributionstrategy each type of actors has. The intention is that this paper will serve as anexplorative study and render future possible research.

2. Method

We used judgmental sampling and selected two actors from each actor type in the Airlineindustry sales network in Sweden. This sales network consists of Airlines, CRS:s, WebSellers and Traditional Travel Agents. We have chosen to exclude the end-customersfrom the interviews even though they are a part of the sales network. The interviews weremade över the telephone with marketing managers, sales directors and/or managersresponsible for relationships with other actors in the network, see Figure 1.

In this study we have not interviewed all the different types of Airlines. We wanted tointerview a second low cost carrier (in addition to Sterling) that had implemented aProprietary Web, but experienced non-response from the Airline. We also experiencednon-response of one of the CRS:s and from one of the selected actors from the WebSeller group, the Collaborative Web. During the study we became aware about that itwould have been preferable to interview one of the Airline partners in the CollaborativeWebs too. Unfortunately we couldn't incorporate that into our research due to timeconstraints. Since we had several other actors contributing with complete answers, thenon-response and narrow sampling have a limited impact of this explorative study.


Company Actor Tvpe(s) Respondcnts" RolesSAS

Skyways

Sterling

Amadeus

Nyman & Schultz(American Express)

Ti eket

Resfeber (Travelocity)

Airline (TraditionalInternational Carrier, WebSeller (Proprietary Web)

Airline (Traditional), WebSeller (Proprietary Web)Airline (Low Cost), WebSeller (Proprietary Web)CRS

Traditional Travel Agent

Traditional Travel Agent,Web Seller (Agent Web)

Web Seller (CRS Web)

CRS Relationship Mgr (15April 2003), CRSInformation Mgr (15 April2003), Marketing Director(22 April 2003)Marketing Director (25April 2003)Country Marketing Mgr (23April 2003)Marketing Director, (22April 2003) TechnicalSpecialist (24 April 2003)Marketing Director (15April 2003), AirlineRelationship Mgr (21 April2003)Marketing Director (14April 2003), AirlineRelationship Mgr (25 April2003)Vice President/MarketingDirector (11 April 2003)

Figurs 1: Interviews made with the case companies during April 2003 (Own development).

The three interviewed Airlines are of different kinds. Their age varies from established inrecent years to 50 years of operation, the size also varies greatly. The route networkoperated is also different in scope and also type. Some of the airline companies fly frompoint-to-point (like Sterling) while others use a spoke-and-hub network (like SAS).

The current world events like Iraq war, SÄRS, and slow economy may have influencedthe respondents' attitudes and made them more prudent in their statements on e.g.competitive impact.


3. Network Actors

Airlines in this paper are the scheduled airlines, thus charter, fright and taxi airlines etcare excluded.

CRS, Computer Reservation Systems, are the aggregators that gather the Airlines' andother travel producers' offerings in one place and make them available to travel sellersequipped with a suitable computer. In essence, the CRS sell information distribution tothe Airlines and information access to the travel agents etc. Traditional Travel Agentsrefers to the brick-and-mortar travel agents that have a broad range of products incategories such as Airlines, Hotels, and Car rental etc.

Web Sellers mainly use Internet as their customer interface. Four sub types in the WebSellers group have been identified: Agent Web, a travel agent with Internet as the maincustomer interface and sells many Airlines tickets, makes hotel reservations etc. SomeTraditional Travel Agents have a web interface in conjunction with their physical stores.CRS Web, an online travel agent owned by a CRS. Proprietary Web, a web interfaceowned and operated within an Airline's organization. This Web Seller often only offersits' own tickets. Sometimes other Airlines' offerings are also available; tickets onAirlines in the same alliance are more common to be offered in this case. CollaborativeWeb, a joint-venture between Airlines with several Airlines' offerings; the owningAirlines' and others.

Other actors that are selling tickets, such as Airline's ticket offices, call centers, andconsolidators, and tour operators are outside the scope of this paper. Actors that arediscussed here are presented in Figure 2.


Figure 2: The Sales Channel graph, showing alternatives that Airlines can use. Airline (A) to Airline (D)depicts typical way ofusing the Traditional Travel Agents and Web alternatives. (Own development).

Airlines started to develop CRS jointly in groups of Airlines. It resulted in four majorCRS companies; Amadeus, Galileo, Sabre, and Worldspan. For a long time thedominating way of selling Airline tickets was through the CRS to Traditional TravelAgents, like Airline of type "A" in Figure 2. In the mid 90-ties, Airlines started to useweb sites operated by them to sell tickets in addition to the CRS channel, depicted byAirline "B". One of the first was British Midland that started to sell online in January1995. Some chose to only sell through their own Proprietary Web, like Airline "C". Itwas many times low-cost carriers who did it to cut transaction costs associated with ticketsales. The lowest cost would be when the web site is directly linked to the Airline'sinventory. Many Airlines still use CRS, even for their Proprietary Web. (Grenblad &Rosén, 1999) The Internet opportunities also gave birth to a new form of collaborationbetween Airlines, where they created a joint-venture web site, Collaborative Web. In thiscase there is a mix of directly connected Airlines and the ones connected through a CRS.Several Airlines tickets are aggregated to one web site and then sold on the web sitedirectly to the end-customer. Airlines involved as an owner in a Collaborative Web isshown as Airline "D". Today, the traditional Airlines use several sales channelssimultaneously and Buyers alternate their choice of channels.


4. Theoretical framework

This paper takes on the markets-as-networks approach (Mattsson, 1982), which is adifferent research stream compared to the marketing-mix approach (Borden, 1964).Markets-as-networks is well suited in this case because it acknowledge multiple actorsand interdependence among them, rather than a dyadic transaction based relationship. Allactors contribute with different activities and resources, so that the Airline ticket can besold to the end-customer. A definition of a network is:

"A network is groups oftwo or more connected relations ".(p. 725, Cook & Emerson, 1978).

The boarder of the network is limited here by using a more narrow definition of thenetwork where only those actors that are acting towards a shared goal are included (Vanden Ven & Ferry, 1980). Here in this paper the shared goal is to sell Airline tickets.

What characterize a network are the actors and dependencies between them. Examples oflinks could be dependence of each others resources, such as technical solutions, or socialrelationships. Dyadic relationships are also of importance, because of the indirect linksthrough the dyadic actor to the rest of the network. The strength of the links variesdepending on the situation. The links between actors can be of different sorts. The linkcan be direct between actors or indirect through other actors.

Relationships between actors can be look upon in closer detail and three different basisfor links are possible (Håkansson & Snehota, 1995):

> Activity-links; technical, administrative and commercial activities that can beconnected to other actors and in this way develops a relationship.

> Resource-ties; connects various resources (technical, material, knowledge etc) tothe company.

> Actor-bonds; connects actors and affects how the actors are perceived. This linkalso influences their identities and organizational structure.

With a market-as-networks analysis, the actors within the network can better understandtheir business context: Who are the dominating actors, not only through size, but also interms of centrality in the network. Are groupings within the network are unevenlydistributed and less accessible. If there are power hierarchies were actors are able toinfluence how others should act. It also helps the interpretation of other actors' strategiesand actions.


4.1. Which distribution strategy does each actor have?

First we describe the empirical fmdings of our research which will be continued byanalysis of the empirical fmdings.

4.1.1. Airline - Empirical Findings and Analysis on Strategy

Actors the Airline is depending on are CRS, Travel Agents, booking systemsuppliers, IATA and credit card companies. The main sales channel for traditionalAirlines involves the travel agent. Some Airlines see the travel agent as a saleschannel partner and others see them as customers. Difficulties related to theserelationships include high costs of the CRS service and travel agents' trust in theAirline. New cost Airlines are more often trying to use Internet as the mainchannel. Established Airlines often have a diversified sales channel strategy. NewAirline have a more focused strategy, probably because of a lack of resourcesneeded to use multiple channels. It has been suggested that the Internet is moresuitable for younger customers and travel agents for older. Internet sales are seenas a less costly sales channel. Other Airlines can constitute partners as well ascompetitors. When the Airline wants to offer interlining (a combined ticket fortravel on different Airlines) an agreement with the Airline might be needed inaddition to the available industry rules. A small but important part of the Airlines'revenue comes from direct agreements with corporate customers for a certainamount of travel for a predeflned price. The interviewed Airlines also said that theyneeded booking systems and other IT resources such as CRS.

The Airlines seem to be well aware of their dependence on other actors and that theytogether act in a kind of system (here denoted as the sales network). Some of thedependencies are indirect through other actors; the best example is the Travel Agentdependency. They also recognize that they need certain resources that they get from otheractors, e.g. the system provided by CRS companies. These circumstances can better beincorporated in the market-as-network theory compared to a transaction based theory,which would suggest that the theoretical frame work used in this paper is well suited.Other Airlines see the Traditional Travel Agents as customers, a link more based onactivity-link. The Airline/travel agent link is stronger in the partner situation. Difficultiesrelated to the relationship are the CRS costs to involve the travel agents and that is whymany new low cost Airlines use Internet as their main channel.


4.1.2. CRS - Empirical Findings and Analysis on Strategy

The overall activity performed by the CRS is connecting the Airlines with the travelagents. Travel agents are seen as customers. Both web and physical travel agentsare served. A well defined distribution strategy is needed, so that marketing isperformed towards the right target group. CRS systems offers different levels ofsupport aimed ät business travel. The use of CRS Web travel agents varies betweenCRS:s and geographical märkets. When making the choice if a web channeldirectly to the end Buyers should be used, the relationships with travel agents areconsidered. There are a relatively few large actors in the Swedish air travel märketsuch as CRS, business travel agents, and corporate customers. The CRS werecreated to cater to the needs to the traditional Airlines, which means that the needsof Airlines with new business mode Is might not be appropriately supported by theCRS

In those cases new Airlines näve to create new solutions. The CRS companies seethemselves as a service unit towards the Airlines, a resource-tie. Both on andoffline Travel Agents are thought of as customers, thus an activity-link.

4.1.3. Traditional Travel Agents- Empirical Findings and Analysis onStrategy

The view of the travel agent's business is not completely shared by all people, noteven ät the same travel agent. One respondent was certain that their role was toprovide various travel management services to the customers and act as theirpurchasing representative. Another respondent ät the same travel agent saw theirrole as sellers of the Airlines' offerings.

Actors that they depend on areflrst ofall CRS:s, travel producers and ticket wholesale companies.

To contact the customers the travel agents need resources such as telephone, e-mailand web. On the supplier side the CRS is crucial. Some travel agents use the CRSnot only for booking tickets, but alsofor travel management e.g. customer profiles.

There has been a change in the travel agent industry were the agents role has shifted frombeing an advocate of the Airlines offering to being an advisor to the customer. One of theimportant drivers for this change has been the cannibalization of the travel agent channelmade by Airlines' Proprietary Web. This meant that the travel agents had to define theirposition clearer to justify their existence. A second driver is the related change of


compensation given by Airlines; the commission has been lowered or removedcompletely. This meant that the travel agents had to fmd new sources of revenue,resulting in "consulting fees" charged in parallel to the ticket price. (Grenblad & Rosén,1999) The respondents ät travel agents showed that they are not, as of today, completelycertain what they are. Are they still selling on behalf of the Airlines, or are they helpingcustomers to buy?

Resources needed to link to the customers are telephone, e-mail, web, and physicalstores. A necessity for the travel agents business is the CRS resource because without itthey wouldn't have much to sell. Even if some tickets are bought through ProprietaryWeb, the väst majority comes from the CRS.

4.1.4. Web- Empirical Findings and Analysis on Strategy

Activities needed for a web travel agent to function include direct sales, technicalunit, production unit, and sales unit. In addition a focus on marketing is seen asimportant. An Internet access provider is a relationship that has to w or k for theweb actor. "Meeting the customer" is one rule mentioned for organizing thedistribution strategy, meaning adapting to a customer segment 's habits and needs.The Internet is seen as the most important channel, with support from telephone. Amajor advantage of the channel is the speed: E.g. when there is an opportunity tosell an attractive, but time-limited product, it can quickly be offered and sold to thecustomers. The least important is physical stores. Just as for physical travel agents,some web travel agents are also packagingproducts into an offering, in addition totrading travel products.

A special characteristic of the Web Sellers is the dependency on the resource Internetconnection. To reach as many customers as possible it is of importance to create a strongbrand.


5. Sales Network Link Analysis

All actors chosen to be included in the sales networks share the activity "selling airlinetickets". This common characteristic is not discussed for each one of the relationships,only those aspects that are signifying the relationship in other aspects.

Airlines contribute with resources to the CRS:s in the form of information on sellableproducts, i.e. the Airline tickets available of their inventory. A CRS can also be seen ascontributing with resources to the Airline. This is the technical resource of the actualreservation system as a standardized interface to the travel agents. The link cannot becharacteri/ed as based on actor-bonds; the choice is more about how to get a morecomplete coverage of the märket. Therefore it is more of a resource-tie, a relationshipbased on exchange of resources.

The Traditional Travel Agents and Web Sellers use CRS:s as resources for informationon Airlines offerings and book Airlines' tickets. Traditional Travel Agents that servecorporate customers are sometimes using additional resources, such as storing customerprofiles and travel policy guidelines in the CRS, which means that they are moredependent on the CRS. The Traditional Travel Agent and CRS are linked throughresource-ties. It should be remembered that Traditional Travel Agents can also have aweb site were they sell tickets, an Agent Web.

Traditional Travel Agents and Buyers links to each other differently depending on theBuyer type and travel type. Leisure travel is usually less frequent transaction were thecustomer is indifferent to where the ticket is bought. More important is what kind ofproduct is obtained for the price. This means shopping around for the best offer and theTraditional Travel Agent/Leisure Buyer relationship is limited to a resource-tie.

For Corporate Buyers that prefer long-term relationships implying that the linkage isbased on actor-bonds, which means that linkages are based on social relationships andperceptions about the other actor. It is further emphasized by the fäet that CorporateBuyers typically wants to limit the number of suppliers. However, the relationship withthe Corporate Buyer can also be based on resource-ties. This because the corporatecustomers often have demands for additional services in the travel management area, inwhich case the Traditional Travel Agent even might be embedded in the CorporateBuyers' organization.

When the ticket is for a simple travel, e.g. between one metropolitan city to another inEuropé, the link is mostly a resource tie. There is a different case with complex travel,such as a package of complementary products or difficult itineraries with several flightlegs. It can also be an Actor-bond, a social relationship catering for trust, when the brand

Dembrower, Maria. & Grenblad, Daniel (2003). 10

of the travel agent influence which travel product is chosen to be used by the Buyer, e.g.to get advice on how to decide on alternative packages. Furthermore, complex travel caninduce new activities taken on by the travel agent beside the finding and booking oftickets. This could be finding and providing information about the travel locations. Newactivities might spring out of the interaction, such as needs for additional services. It canbe seen as an activity-tie, because it is a co-production with intense interaction betweeneach actor. E.g. providing information on hotel booking or tailored sightseeingarrangements.

Traditional Travel Agents have relationships with Airlines for other purpose than theticket reservation. It could be about promotional campaigns or special agreements onbehalf of Corporate Buyers. The Airline's interest might be to nurture the connectionwith a large Traditional Travel Agent with many Corporate Customers. This relationshipis an activity-link since resource use is not involved in that activity, rather it prepares forlåter resource use (via CRS). Other actors may also have such relationships, but they areless important.

The Web Sellers' links to the Buyer is more based on resource-ties, than the TraditionalTravel Agent because the Buyers are in these cases comparing offers and prices morethan seeking advice and augmented services. CRS Web, Collaborative Web, andProprietary Web have linkages to the Buyer that are clearly resource-tied.

One of the Web Sellers, the Agent Web, is in one way similar to the Traditional TravelAgent. The actor-bond is present in this link too, with the role that brand has as a"guarantee" for the products offered. It is possible that Buyers who are new to onlinepurchasing prefer to use an Agent Web that they are familiar with, i.e. those that areTraditional Travel Agents too. However, it is weaker in the aspect of social relationship.The Agent Web relies on Internet as the main interface; personal contact is possiblethrough the telephone but still not a primary mean of communication.

CRS Web links are strongly based on actor-bonds since they are tied legally throughownership. It would be unlikely to find the CRS Web to use another source for theproducts sold. There is of course a resource-tie with the technical system involved, just ase.g. an Agent Web actor.


Collaborative Web and Proprietary Web have the option to link to CRS for productsourcing, or to connect directly to Airlines' inventories. One of the Collaborative Webactors, Opodo, is partly owned by a CRS, Amadeus, which wants to influence thisstrategy. Both the Proprietary Web's and Collaborative Web's link to the CRS, is mostlya resource-tie. The Proprietary Web is strongly linked to an Airline through an actor-bond, due to the ownership. The Collaborative Web has actor-bonds to several Airlines,since it is a joint-venture between Airlines. While the Collaborative Web is operated as aseparate company, the Proprietary Web is typically a unit within the Airline'sorganization.

Airline (A) Airline (B)1

Airline (C) 1 Airline (D)

Figure 3: The Sales Network Link graph, showing relationships between actors. Stronger links aredisplayed with thicker lines. (Own development)


6. Implications of Links

From the perspective of the Airline, the links discussed above have the followingimplications. Airlines use the CRS not because of the CRS itself. The reason is to reachthe resellers in turn use CRS. Le. the CRS serves as an indirect link to the TraditionalTravel Agents and to the Web Sellers. The advantage of this is the märket coverage thatis achieved. A väst majority of the Airline ticket sales still goes through CRS. The maindisadvantage is the high cost per sale of ticket, compared to the Proprietary Webalternative.

For this reason, new Airlines have been seen to focus on the Proprietary Web channel.Still , it is not associated with lower cost in every aspect. The Airline is believed to need alarger marketing budget in order to make the Buyers find the web site and to trust theproduct offered. When Airlines use the Traditional Travel Agents this marketing cost isin part carried by them.

The advantage of having direct distribution is also to gain better ownership of the Buyer(Grenblad & Rosén, 1999). This means more information on the customer and betterpossibilities to influence them e.g. through customized campaigns. A downside is thereach of the Web Sellers' channels, only counts for a smaller part of total Airline ticketsales.

Airlines try to shift the customers from the Traditional Travel Agents to Proprietary Webthrough using low prices only available ät the Proprietary Web. The reason for this couldbe that the actual ticket cost less there and/or the sales cost associated with the CRS travelagent channel has been eliminated.

The analysis of the individual links above leads the discussion of what it means ät anaggregated network level. The strong relationships to and from the CRS actor andcentrality in the sales network indicate power in the network. It could be an explanationof the relatively higher net profits (DÖT) of the CRS systems, compared to the otheractors in the network.


6.1.Network Changes

The network approach gives the opportunity to describe types of changes: First, arefmement of existing activities and resources, a Structuring process. Structuring is acontinuous process as it changes both the technical and the social elements över time.Structuring implies that actors try to improve the utilization of the ideas, through refmingtechnical connections and rationalizations. This requires increased standardization ofinput and output and in this way resources are used more efficiently. The network ismade clearer and stronger through this Structuring process. This process involves severalactors and relationships that are essential to change the industrial network. The Criticalelements are the social interactions. Individual companies initiate the changes and theynäve impacts on the whole network.

Second, develop new ways of combining activities and resources, Heterogenizationprocess. This process implies that the actors try to apply new ways of using the resourcesof the same activities. Heterogenization competes with the Structuring process, because itdevelops alternatives to the established structure. The process can be done within theestablished network or it can be done with the help of companies belonging to othernetworks. While the Structuring process increases efficiency, the Heterogenizationprocess increases effectiveness. Both of the change processes seek to limit the resourcesused.

The companies want to control resources, which leads to two other change processes. Therelationships can be seen as ways of creating opportunities to combine resources and tomobilize other actors' abilities to solve the problems. Increasing control of activities andresources held by fewer actors can be seen as a third change process, a Hierarchizationprocess. The Structuring process is a foundation of the Hierarchization process. When anetwork is structured it allows for another process, where fewer and fewer control moreand more. Ät the same time the actors get more limited connections between activitiesand resources. One way of getting more power in a network is through controlling moreresources and activities. By controlling the resources, the actors get more power inrelation to other actors. It can be achieved through ownership or relationships with actorsthat posses the resource. For the individual company it means conflicts (e.g. pricecompetition) and cooperation (e.g. coalitions) simultaneously.

A competing process to the Hierarchization that we label the Dilution process. In thisfourth process, the control över activities and resources get distributed över more actors.The different actors continuously affect these development patterns. Each actor usuallybelongs to several networks, how the development patterns look depends on whichnetwork that is used during the analysis. The network approach highlights theinterdependence among actors that changes typically have repercussions for several


network members. Adjustments of strategies etc are important to survive in the networkand those who adapt are more likely to enjoy success. (Håkansson, 1992)

The major changes mentioned in the introduction, deregulation and e-Business, nas leadto changes in the whole Airline industry.

The lack of competition and the dominance of a few actors on the Swedish märket havemade the policy makers to deregulate the Airline märket in 1992. The aim of thisderegulation was to make it easier for new actors to enter the märket and increase thecompetition between the firms. The effects from the deregulation were not before the endof the 90's shown on the märket through the establishment of new Airline firms. Lowcost Airlines perhaps have received the most attention. They have performed the activitytravel in a new way, by flying point-to-point. Airports were used as resources in a newway as they contributed to the marketing of the Airline's flight route. Thus, it can becategorized as a Dilution and Heterogenization process.

When Internet started to be used commercially in the mid 90's (Grenblad & Rosén,1999), Airline tickets were one of the earliest products to be sold. Incumbent Airlinesdeveloped web sites to sell tickets. It was a new combination of resources triggered bythe enabling technology Internet. In other words it is a Heterogenization process. Astronger över all change for the traditional airlines might be structuring of their businesstriggered by the increased competition from the low cost carriers. The change can be seenas a form of developing the existing activities and resources, a structuring process. Therewould probably follow mergers Hierarchization among traditional airlines where feweractors control activities and resources. On the other hand, the low cost carriers' entrieshave developed the industry towards Dilution and Heterogenization.

Internet also introduced new Web Sellers to establish on the märket. Some of these newactors even came from other industries such as Software (Expedia/Microsoft). In additionnew actors appeared that used Internet as their primary channel. Web Sellers are involvedwith the existing activity of sell ing tickets using the new resource. This is probably themost obvious Dilution and Heterogenization process.

Traditional Travel Agents are simultaneously in a Structuring and Heterogenizaionprocess. Structuring is the development of existing activities and resources, e.g. in thearea of services for travel management. Travel agent might merge to increase theefficiency through a Hierarchization process. Some Heterogenization takes place throughusing the new resource, Internet.

The Traditional Travel Agents are competing with the growing Web Sellers. Theentrance of Web Sellers has induced the balance in the network. This has forced theTraditional Travel Agent to rethink their business model and operations. This has meantselling through the web and not just selling through the physical stores and through the


telephone. They have also tried to create stronger relationships with both the airlines andthe CRS actors that they still give them attention. Since the role between the CRS and thebuyers has been challenged, the Travel Agents have been obliged to clarify their positionin the network and where their loyalty is. Are they on the sellers' side or are they on thebuyers' side? If it is a Traditional Travel Agent that target corporate buyers they havebeen leaning to being an allied of the Corporate Buyers. If it is a Traditional Travel Agentthat target leisure buyers they are still in many ways sellers of the travel producers'offerings. Över all it means that the Traditional Travel Agents and Web Sellers need tofind "available" competitive positions.

CRS actors have developed their existing activities and resources, through a Structuringprocess. Beside of this the CRS Web is using the new resource Internet, aHeterogenization process taking place, adding a new type of actor to the network, aDilution process. The numbers of CRS actors are already a few on the märket and wehave not seen that it has been another Hierarchization process yet. The CRS Webcompanies w i l l probably develop as the other Web Sellers will do.

Heterogenization

New combinations of activitiesand resources

Hierarchization

Fewer actors control activitiesand resources

Dilution

Websales

•V 'N.

LowCost

Airline—-^5~.___*s More actors control activities and

Travel \ resourcesAgent

Structuring

Development of existingactivities and resources

Figure 4: Actor changes in the network relationshipperspective. (Own development)


6.2. Can ticket sales itself constitute an entrance problem (newAirline)?

First we describe the empirical findings which then will be analyzed with help from thetheoretical framework.

6.2.1. Airline - Empirical Findings and Analysis on Barriers

Something that can be an entrance problem is the necessity ofearly revenue due tothe high upfront costs when storting up the business. For an individual flight togive a net contribution to the firm a high load factor is needed. Which for a lowcost carrier typically should be higher (perhaps 80 %) than for a traditionalcarrier (perhaps 60 %). This means that a strong sales strategy is crucial. TheAirline industry is regulated in some ways, both by industry actors, e.g. LATA, andpublic actors, e.g. EU. The regulated environment puts constraints on the kind ofactivities feasible to undertake. Direct Internet distribution results in lowerdistribution costs that will enable lower ticket prices.

A more extensive Airline/'Airline agreementfor interlining might not be possible fora new Airline to get. For instance ifthe potential Airline partner already have suchagreement for the route through an alliance. If the new Airline chooses toparticipate in the traditional sales channel, they are depending on the CRS:sbecause theyfunction as the link to the physical and online travel agents. With thealternative of direct Internet sales, a solution for the payment of the tickets isneeded. Usually this means that credit card companies are involved.

The new Airline needs to build a strong brand early. The brand represents the trustneeded for the travel agents to promote the Airlines offering, especially towardscorporate customers. The brand also injluences the traffic and sales of the Internetchannel. Credibility is a key factor for success; the Airline should not riskdisappearing from the märket. Furthermore, since the customers essentially arebuying time when they choose flying as a travel option, they must have reliability inthe booking andflights. Numerous departures on the same route add to the "shelf-space " in the CRS that the travel agent sees. This can be further accentuated bycode-share agreements. In both cases it is an advantage to a larger establishedAirline.

Especially the newer Airline mentioned the importance offinancial endurance toenable the advertising that would attract customers to their Internet channel.Financial strength is also believed to influence the trust given by the travel agents.


Some Airlines have indicated an interest in lowering their dependence on other actors bydoing more of the activities themselves, such as direct distribution through Internet. Therelationships that the actors already have influence both current and potentialrelationship. (Håkansson & Snehota, 1995) An example of this is the difficulty of gettingan Airline/Airline agreement, especially when you are a new actor on the märket (in otherwords an entry barrier).

Brand in the shape of credibility and reliability is once again mentioned that as a resourcethat would be needed to create a good brand to the customers. Combined with the highupfront costs it calls for good finances and early sales volumes.

6.2.2. CRS - Empirical Findings and Analysis on Barriers

A possible barrier for a new Airline could be to get a strong enough brand which isrecognized by the travel agents. A reason for the need of large financial resourcesis the difficulties of changing customer 's habits. This also means that it could bewiser to adapt the strategy to the customers' present behavior, perhaps by usingtravel agents since they still sell the majority oftickets.

The new Airline could experience the cost oftechnical systems as a barrier, e.g. toconnect to the CRS: s. Connection to BSP and credit card company is needed for thepayments. A way to get some of the distribution channel resources needed is byusing hosting services offered by actors such as EDS and Scandinavian Airlines. Itis important to find strategic places were to do the sales, e. g. which travel agents tofocus on. The travel agents must believe in the survival of the new Airline, since thetravel agent is betting their own brand when they sell Airlines tickets. This isespecially important in business travel, because the travel agent and corporatecustomers are usually seeking long-term relationships. It is important to gainrevenue fast due to the fixed costs and since it is impossible to sell seats once theflight has departed.

The trust and expectations regarding the new Airline is important. Rumors couldhave an impact on the relationships with the travel agents.

When choosing channels the new Airlines might choose direct sales when theyfocus on leisure travel, since it is characterized as single transactions. Businesstravel is characterized as continuous relationships. Travel agents offer larger salesvolumes. Large corporate customers do not only mean benefits in terms of largepotential sales. The downside are demands for a more developed booking systemthat includes more features and service, more demanding professional purchasers,andmorepricepressure. The travel agents often "own the customer".


The resource financial assets appear to be an important factor for the success of the newAirline. One reason is that it is directly linked to the perceived brand credibility and alsoa mean to create a strong brand. The brand can be seen as a necessary Airline resource forthe actor bond between the travel agents and Airlines to emerge on the märket. The travelagent does not want to risk their own brand resource especially those who work withcorporate customers since it is a lasting cycle based of activities. Especially largecorporate customers use the travel agent as a resource that provides travel management.New Airlines have to consider the resources as the BSP, credit card company forpayments.

6.2.3. Traditional Travel Agents - Empirical Findings and Analysis onBarriers

An important activity is to create a strong brand, so that the Airline could be seenas a viable actor by the travel agents. The travel agent is putting their own brand ätstake, if they sell a questionable product to the customer. Marketing is needed tocreate the brand and to drive the sales. The direct sales alternative leads to lowerticket distribution cost. Due to the large initial costs, immediate revenue is ofimportance. Which stress the need for a viable marketing strategy andimplementation. The new Airline might want to cooperate with other Airlines e.g. tocreate a strong brand.

Good märket knowledge is needed. Which is the torget group and how should it bereached? Sales stoff is needed as a complement, even if the web is used. Money isneeded for the marketing. Ifthe Airline chooses to use a web interface for the travelagents it must be adapted to the travel agents needs.

Many traditional travel agents cater to all kinds of customer groups. Nowadays therespondents point out the increased importance of defming the target group andunderstanding it so that the marketing activities are effective and efficient. Thecustomer group of this business must be recognized as being different compared toothers.


6.2.4. Web - Empirical Findings and Analysis on Barriers

According to a web travel agent, a new Airline requires travel agent sales andagreements for cooperation with other Airlines. Costs faced by a new Airline arefor adapting the booking system to travel agents and CRS:s. The new Airline mustget large sales volumes to reach the loadfactor targets. The largest sales volumesare believed to continue to be through the travel agent channel in the next fewyears. The have speed in larger sales volumes, the ticket should be electronic.Resources needed include good finances, a long-term business plan, customerknowledge, technical knowledge, IT-support, access to a booking system, andagreements with CRS. The chosen sales channel solution should be simple. TheAirline 's sales system must be adapted to travel agents.

The suggestion to use travel agents is because of their ubiquitous stores that constitute abig resource, since it gives such broad märket coverage. This could be something thatwould help to achieve the load factors needed due to the economics of Airlines.Electronic tickets can be seen as a resource as it eases the distribution of sold tickets. Thenew Airlines resources in the shape of technical systems should be in accordance with thetravel agents demands to create an actor bond. The obvious way of doing that is to use theCRS system and it is however to a small extent also with web interfaces.

The already mentioned necessity of an Airline brand is confirmed with the travel agentrespondent. It can be interpreted as corresponding the actor-bond type of link as animportant impact for the overall relationship. Another area of consensus is the need forthe Airline to get revenues quickly. From a network perspective the need of fast incomeslead to the importance of having a good financial situation before you start to operate. Ifthe Airline cooperates with an actor outside the sales network it can amass the resourcesneeded in a better way than relying on their own finances. Or, a second opportunity is tocreate an actor-bond with an established actor (Håkansson & Snehota, 1995), e.g. anAirline of good reputation and thus improving their own brand. A third issue confirmed isthe need for adaptation of web interfaces.


7. Discussion

The large volume of Airline ticket sales is still in the Traditional Travel Agent channel,approximately 80 % ofall sales (empirical fmding from interviews). High growth of salesis on the other hand taking place in the Web Sellers' channels. Online travel sales inWestern Europé increased by 53 % in 2002, while total travel sales increased only with1.2 % (CRTR, 2003). This indicates that the Web cannot be neglected as a channel, evenif the volumes still are small relative to the Travel Agent channel. For the Airline itmeans that Web Sellers are important to them. This is especially true for the ProprietaryWeb and Collaborative Web, since these are the Web Sellers' channel they can controlthrough their legal ties.

Traditional Airlines have a legacy in the Traditional Travel Agent channel, why choosingbetween using them or not is already made in a way. For them it is more about how muchemphasis they place on the Proprietary Web. For new Airlines it is a different case.Without having any established relationship to take into account, it allows for morefreedom to choose the channel that seems to fit their business model. The two mainalternatives and their differences are:

> The Traditional Travel Agent channel with 80 % märket share (empirical fmdingfrom interviews). The channel suffers from higher cost per sold ticket due to theCRS and Travel Agent intermediaries, which translates into higher priced tickets.These incremental costs for each sold ticket are distributed över time and can thusbe seen as a marketing expense. Since this is an established channel, buyers knowhow and where to fmd a Traditional Travel Agent. Le. there is no marketing costassociated with creating this awareness. All Traditional Travel Agents more orless have possibility to sell all Airlines, i.e. as soon as the buyer is ät a TraditionalTravel Agent the Airline's offering can be sold. Airlines however loose controlöver the sales situation, they have to rely on advertising and incentives toinfluence the purchasing choice.

> A Proprietary Web channel. The Web Sellers as a group has about 20 % märketshare (empirical fmding from interviews). A first limitation of this channel is thatthe potential märket is smaller since not every one use Internet. For countries inEU and North America Internet usage in September 2002 ranged from about 20-70 % of the population (Nielsen, 2003). Secondly, with 40 Million web sites(Netcraft, 2003) the chance of ending up ät unknown site by luck is small. To geta part of the web sales, large marketing expenses must be made to createawareness even before flight operations start. Thirdly, once ät the site the buyermust trust that payment and personal integrity is treated appropriately. Here thebrand is again playing an important role which is costly for the Airline especially


in the beginning. The Airline's Proprietary Web otherwise cannot expect to getany sales. For the reasons above it sums up to large upfront marketing costs. Oneadvantage is the lower channel cost per sold ticket. Also, since the Airline hasdirect contact with the buyer it can result in better customer ownership (Grenblad& Rosén, 1999).

All respondents were aware of the difficulties of entering the Airline märket and amajority of them believed that there were entry barriers for a new Airline in terms ofselling tickets. Even those who answered that there were no entry barriers, mentioneddifferent kinds of aspects that could be seen as barriers:

> Travel Agents choose to cooperate with established actors to limit there own riskof damaging their own brand.

> Limited financial resources. Not only for the flight operations, but also forbuilding the brand. Another aspect for the new Airline to consider is how toestablish a credible image so that customer dears to choose the alternative. As anext step a strong brand could also help in creating loyalty among customers andmake them return as travelers. A question can be raised about those statements.The jargon in today's populär business jargon includes several words which are infashion; "brand" is often used in all sorts of contexts. Caution is therefore calledfor and it is not certain that the brand is the real solution. It could also just be oneof those "catch all" Solutions perceived among the general public to take care ofmore things than it actually does.

> Creating appropriate technical system for the selling of tickets. Integration withCRS could be an entry barrier if the initial costs to adjust the technical systemsare high in the beginning. When integrating with the CRS system it's not selfevident to get good sales if the company hasn't a well known brand yet.

> An Airline that has an innovative business model could find that it is difficult touse the CRS based channels. The CRS:s were created by and for the traditionalairlines to support their business models ät that time. Legacy IT systems,especially complex ones as the CRS are härd to change. Therefore the newinnovative airline might be forced to create its' own proprietary solution for thesales function, such as a Proprietary Web.

> Having sellable products could call for interlining, which in turn is said to requireagreements with other Airlines. The standard industry regulations are notcovering all aspects.

> Cutting through the noise in the CRS system, i.e. having enough shelf spacepresence. Large Airlines with many departures between destinations, especially


those with code share, get many listings presented in the CRS system when asearch for a ticket is made. One possible entry barrier not mentioned by therespondents is due to the CRS position in the network as discussed above, seechapter 6. Because much of the value resulting from the sales activity ispositioned inside the CRS:s in the shape of higher relative profits, it could meanthat other actors they get unfavorable terms of the business. If these terms weremore diversified among the different actors it could be easier for all the actorsoperating in the network and especially for new actors trying to enter the industry.More even profit distribution could help the new actor to get the financialresources which have been a problem for all the actors in the beginning.

The network changes found have been involving all the other actors but the CRS:s. Whatthe CRS has done is only to structure their business. The entry of Web Seller has notweakened their position since many of them choose to connect to a CRS. While the otheractors in the network are going through large changes since actors in a network dependon each other it is possible that the CRS will be affected eventually and forced to changetheir process and strategy.


8. Conclusion

In this paper we have found that there are barriers to entry with regards of sales of ticketswhich fall into three broad groups. First, access to channels is hindered: The Airline's ITsystem requires adaptation, Travel Agents that distribute the tickets select which Airlinesto prioritize, and Airlines might limit the code sharing agreements to a few actors.Second, resources limit sales performance: Financial resources are often limited for anew actor, which means less possibility to create a strong brand and receive necessaryattention. Third, network structure impact return of investments: The central position ofthe CRS in the sales network means a barrier to entry. Much of the value that the saleschannels add results in relatively large profits for the CRS. It means that it is difficult forAirlines to internalize this value and get returns on their investments in sales channels.This is ät least true for channels using the CRS, which still is common practice even forProprietary Webs.


9. Future Research

> To get generalizable results on the issues of sales barriers in particular, a surveywould be suitable with more actors included and perhaps also more respondents äteach actor.

> The industry changes would be interesting to follow with the perspective ofmarkets-as-networks approach. Will structuring take place where use of activitiesand/or resources is refmed to attain better efficiency? Or, will a Heterogenizationprocess nappen where new ways of combining resources result in increasedeffectiveness?

> An area not covered in this paper is about the distribution of tickets that are sold.Similar research could be made on this aspect to see if it could be an entrancebarrier, e.g. does e-ticketing make it more easy or difficult for new Airlines? Whatare the resources needed and what are the difficulties in getting access to them?

> The dominating position of the CRS could be interesting to study and to see if itcould affect the entrance of new airlines. So far much of the competitive impactof the CRS:s has been analyzed in relation to the incumbent companies, see forinstance the US Department of Trade.

> This paper can be an input and starting point for analysis of competition in theAirline industry and forces limiting competition. Sales and revenues are importantfactors to a company and this could be a future research question to include in thecompetitive analysis.


10. References

Axelsson, B. (1996). Professionell marknadsföring. Lund, Sweden: Studentlitteratur.

Borden, Neil H. The Concept of the Marketing Mix. Journal of Advertising Research,June 1964.

Cook, K. & Emerson (1987), R. M. 1978, Non-Exchange Relationships in Networks.American Sociological Review. Vol 43. October, pp 721-39.

CRTR (2003). Western European online travel sales double. Ät NUA, 2003-05-13http://www.nua.com/survevs/index.cgi?f=VS&art id=905358743&rel=true.

DÖT. US Department of Trade, DOT-P-37-88-2.

Grenblad, D. & Rosén, P. (1999). Internet- A Sales Channel in the Airline Industry.Master Thesis, University of Linköping Electronic Press. ISRN LIU-EKIFEK-D—1999:EP:D/40—SE

Håkansson, H (1992). Evolution Processes in Industrial Networks. In Axelsson, B &Easton, G. (eds) Industrial Networks. A New View ofReality. London, UK: Routledge.

Håkansson, H. & Snehota, I (1995). Developing Relationships in Business Networks.London, UK: Routledge.

Mattsson, LG (1982). Om utvecklingen av marknadsföring och distributionsekonomi iSverige. In Brunsson, N (ed) Företagsekonomi - sanning eller moral? Om det normativai företagsekonomisk idéutveckling. Lund, Sweden: Studentlitteratur.

Netcraft (2003). The Netcraft Web Server Survey, 2003-06-13.http://news.netcraft.com/archives/2003/05/05/mav_2003_web_server_survey.html

Nielsen Netrating (2003). How Many Online? Ät NUA, 2003-06-13.http://www.nua.com/surveys/how manv_on]ine/index.html

Van den Ven & Ferry (1980). Measuring and Assessing Organizations. New York, USA:Wiley.

6f.3i

Work in Progress paper to be presented ät the IMP conference in Copenhagen, 2004

Preliminary, do not quote

Maria Dembrower'

Susanne Hertz

Lärs-Gunnar Mattsson

Restructuring the Swedish märket for regular passenger flights after

deregulation- small firms and the dynamics of network interdependence

Abstract

This paper concerns the restructuring of the Swedish domestic märket for regular passenger

traffic after the deregulation in 1992. We especially focus on how smaller firms entered,

developed and sometimes exited this märket. Special emphasis is put on their network

connections. After the deregulation in 1992 SAS monopoly on some destinations was

challenged by entrants, also on some major routes. SAS tried to preserve its domination on the

domestic märket e.g. by alliances and acquisitions and also by refusing inter-line agreement

with a competitor. After rulings by the Märket Court concerning interlining and the use of

S AS's frequent flier program, competition again increased and S AS's dominance somewhat

declined. Still in 2003 the combined märket share of SAS and its partner Skyways was around

75 %. During the period 1992-2003 more than 30 firms have been active in regular passenger

flights of which half the number have left or being acquired in 2003. A combination of entry

and exit of firms characterized the quite turbulent development. Small firms entered from

nearby air businesses and became members of alliances, mostly operational ones. The

domestic development has to an important extent been influenced by SAS's restructuring and

alliance with Skyways and generally by the international development.

Introduction

In 1992 the airline märket in Sweden was deregulated. The political intention was to create a

more competitive and effective märket for domestic regular passenger flights. The public

policy purpose to open up for competition on previously monopolized märkets, is not only to

1 All ät Stockholm School of Economics, Box 6501,11383 Stockholm.Corresponding author Susanne Hertz, e-mail: [email protected]

get price competition but also to stimulate entrepreneurial and innovative activities related to

production and to services offered.

Before 1992 SAS and Linjeflyg had split up the routes between them. In early 1992 Linjeflyg

was acquired by SAS. There existed a few minor routes in remote areas left to small

operators. The deregulation made it possible for established and new firms, with the necessary

qualifications, to enter the märket and to freely set their prices.

It is important to first comment on the specific entry and development problems on the märket

that refers to the characteristics of the passenger airline business in S weden.

1. The customer base is small for most routes which makes it economically difficult to

have more than one operator on more than 6-8 routes in Sweden. Thus it is difficult for

an entrant to compete with an incumbent on more than a few existing routes

2. Customers on one route, to reach their final destination, often need connections to

other routes, international or domestic. An operator with few routes need some

cooperation agreement with other operators, e.g so called inter-lining agreements or

common pricing for journeys comprising more than one route. The "hub and spöke"

organization of route networks favours large incumbents and operators who cooperate

in alliances, including eg code sharing.

3. Customers' scheduling preferences makes it important for an operator to offer several

departures and arrivals per day on a specific route. This makes it difficult for a small

operator with initially few customers and limited resources

4. Customers generally prefer to fly with big rather than small aircraft. On major routes,

where incumbent operators fly big planes, entering firms are ät disadvantage because

of their initially smaller märket share.

5. Customers might be reluctant to use other airlines than the airline they usually fly, eg.

due to frequent flier programs.

6. Ät major airports, especially Arlanda, the number of arrival and departure "slöts" that

can be allocated to new entrants are quite restricted.

7. For some routes in northern Sweden that need to be subsidized by the government as

part of a regional development program, traffic for a period is awarded an operator

after a bidding process.

Since 1992 a number of small and medium-sized firms entered the märket for domestic

flights. The dominating firm, SAS, gradually changed its märket behavior. Competition and

cooperation developed. The märket was restructured. In the early years after deregulation

competition increased but a few years låter SAS was again sole operator on most major

routes. During the last few years, competition, especially price competition has increased

again .

Since the deregulation, growth and survival of the small firms have been in many cases

jeopardized and many have gone bankrupt or have exited. Various reasons for this have been

suggested. Most of them are related to the ability of SAS to meet competition due its scale

and scope advantages, its frequent flier program, its acquisitions and alliances. Important

reasons for the smaller firms ability to stay and develop on the märket are that they fmd

"niches" in the märket, develop new network relations, fmd new ways to cut costs and are

flexible in response to changing conditions. The change processes on the märket interact and

are of different importance över time.

Purpose and disposition of the paper

The purpose of the paper is to primarily to describe how the Swedish märket for regular

passenger traffic was restructured during the period 1992-2003. We will then specifically

focus on the role of inter-firm cooperation and network dynamics. We also offer some

analytical comments. First we will briefly present our analytical framework, then present

information from the empirical study and finally more generally analyse restructuring that has

taken place.

Analytical framework

We have a markets-as-networks perspective and for this paper we will specifically emphasise

reconfiguration of a märket in terms of actors and connections between actors. There is a

focus on the suppliers of a specific service (regular domestic passenger air transportation in

Sweden). The institutionell setting is changed ("deregulation") and the phenomenon we want

to understand is how opening up for competition initiates strategic actions by incumbents and

new entrants. Given specific attributes of the production system for these services that we

indicated above we focus on the resource dimension in the production system and acquisition

of such resources through cooperation, including alliances, mergers and acquisitions.

Cooperation and competition interacts in such a way that competition takes the form of

collective competition.

The network of actors involved in regular domestic passenger air transportation overlaps with

networks for passenger air transportation internationally and in other domestic märkets. This

network also overlaps with other air businesses like taxi flights, charterflights and airfreight.

The reconfiguration process as interdependent with such overlaps. Changes in such overlaps,

i.e. overlapping is a driving force in reconfiguration. Literature references for this framework

are Johanson and Mattsson, 1992; Hertz and Mattsson, 2004 and Gomes-Casseres, 1996.

Reconfiguration when two firms agree to cooperate is influenced by complementarity and

overlap between their services/products and geographical areas covered (Hertz, 1996)

The empirical study

Empirical data has been gathered from secondary sources about all the airlines operating on

the Swedish märket between 1992 and 2003. Some of these went bankrupt or were acquired

during the period. Personal interviews were made in 15 firms, with experts of the airline

märket and with employees from the Air Traffic Department of Sweden (Luftfartsstyrelsen).

The purpose of the study concerns small firm entrepreneurship and effects of deregulation.

However, in this paper we are not concerned with the entrpreneurship as such. It is also

important to keep in mind that our description is preliminary. We have not been able to fully

check the events and the dates.

We have divided the empirical material into a/ entry on the märket, b/ growth and

development and c/ alliances.

Märket entry

The firms that entered the märket directly after deregulation had a history in the air business

and were coming from nearby niches such as taxi flights, charter for businessmen and also

airfreight. The first four that entered routes to compete with SAS were: Malmö Aviation,

Transwede, Skyways and Nordic European.

-Malmö Aviation started with school and taxi flights and some airfreight in 1981. After an

early bankruptcy and change in ownership Malmö Aviation in 1992 began regular passenger

flights between Malmö and Gothenburgh, Visby and Stockholm-Bromma. While the first two

destinations were soon dropped the Malmö-Bromma route continued. A further change in

ownership occurred when the Norwegian firm Braathens bought the firm, thus bringing

Malmö Aviation into an established operatorns resource constellation.

-Transwede started in 1985 as a charterflight operator. It began regular passenger flights in

competion with SAS from Arlanda to Umeå and to Gothenburgh. Braathens became part-

owner in 1995 (and fully owned in 1997).

- Skyways' was established in 1992 through a merger between two small firms (Salair and

Avia). Its history dates back to 1940 when one of the founding firms was employed by the

Swedish Air Defense to fly "target aircrafts". Skyways employed from the beginning

personnell from SAS and Linjeflyg. It soon began to cooperate rather than compete with SAS,

taking över small routes that SAS wanted to discontinue. In 1995 Skyways and SAS signed a

general interlining agreement.

-Nordic European began as a charter operator and began regular fights on the Östersund-

Arlanda route in 1995. There was a price war with SAS and SAS refused to agree on

interlining.

In the early years after the deregulation a few of the firms established in other air

transportation began regular passenger flights on minor routes. Airborne of Sweden (founded

1985) operated Mora-Arlanda and Sveg-Söderhamn-Arlanda. Air Express (1986) operated a

couple of routes from Norrköping, Falcon Air (1960) operated from the beginning in

cooperation with SAS and SKYWAYS. Flying Enterprise (1970) started to operate from the

Skövde Airport, that it managed on routes Skövde-Arlanda and Skövde-Trollhättan-

Copenhagen, Nordkalottflyg (1974) operated a few routes in the North, Holmstroem (1966)

also operated a few minor routes in middle Sweden, West Air Sweden (1962) operated

Gothenburgh-Sundsvall.

The deregulation stimulated a number of entries by firms founded after 1992. Entrants during

the first five years were Air Nordic, Braathens, Euroflight Polaris, European Executive

Express , Golden Air, Highland Air,Reguljär and Värmlandsflyg.

Air Nordic (1993) was owned by an international freight forwarder firm and flew two

domestic routes out of Borlänge. The Norwegian Braathens 'enterred the Swedish domestic

märket by investment in Malmö Aviation and Transwede, both becoming fully owned

subsidiaries. Euroflight Polaris (1996) had to disrupt its operations the same year because it

failed to fulfil government regulations. European Executive Express (1995) specializing in

taxi and charter flights also began some regular traffic. Golden Air (1993) began to operate

some routes from Skåne to Bromma. Highland Air (1995), with roots in Holmstroem Air

operated some routes originating in Småland and Västmanland. Reguljair (1996) emanated

from the failure of Euroflight Polaris and operated Skellefteå-Umeå-Sundsvall. Värmlandsflyg

(1997) operated Torsby-Hagfors-Arlanda.

After the first five years a number of new firms enterred the domestic märket. City Airline,

Danish Air Transport, Direktflyg, Goodjet, Gotlandsflyg, FlyMe, IBA, Kullaflyg, Nordic

Airlink, Swedline Express Swedeways, Tryggflyg, Waltair, Wideroe.

City Airline (2000) acquired what was left (planes, hangar, terminal, licences and permits) of

the bankrupt airline JET 2000 and began to fly domestically Gothenburgh-Linköping. Danish

Air Transport, a Danish specialist in air freight participated in bidding for routes in Norrland.

Direktflyg (2002) flew Gothenburgh-Kristianstad. Goodjet (2002) was an effort to compte

with low prices on major routes in competition with SAS and Malmö Aviation. It soon went

bankrupt but FlyMe (2004) co-founded by one of the owners of Goodjet enterred the märket

with a similar concept. Gotlandsflyg (2001) challenged SAS and Malmö Aviation with low

price flights from Bromma and Skavsta to Visby. IBA (1998) started operations Gällivare-

Bromma. Kullaflyg (2001) sister company to Gotlandsflyg, using Golden Air as operator flies

Bromma-Ängelholm. Nordic Airlink (1999) operates in Norrland. It was acquired by Finnair

as an answer to SAS'acquisition of Air Bothnia. Swedline Express (2002) participates in

bidding for routes in Norrland. The founding of Swedeways was based on Holmstroem Air.

Flies mail during the night and passengers during the day:Östersund-Bromma, Hudiksvall-

Arlanda. Tryggflyg (1998) started Nyköping -Visby but withdraw from the regular passenger

märket the same year. Waltair (1998) located in Norrköping started to fly Norrköping-

Stockholm and Linköping-Göteborg. Finally, the Norwegian company Wideroe, owned by

SAS participated in bidding for routes in Norrland but did not win any such competition.

Above we have just indicated some characteristics of märket entry, taking place during

different years 1992-2004. As we have indicated, entry of some firms was dependent on exit

of other firms.

Growth, decline, development

After entry the types of developments included:

-total withdrawal from the märket, including bankruptcy, focus on other air business

-mergers and acquisitions

-growth on regular passenger traffic involving more domestic routes, international routes and

domestic routes in other countries

-decline in the regular passenger traffic but not withdrawal from the märket

We will comment on each of those but have no quantitative measures to report and can not

comment on volume or märket share development on specific routes.. There might also be a

combination of development types över time. Bankruptcy might be linked to a låter

acquisition of some resources by another company, growth might precede or follow an

acquisition.

Firms that withdrew from the märket during the period (without being acquired) were Air

Nordic, Euroflight Polaris, Falcon Air, Goodjet, Holmstroem, Nordic European, Reguljair,

Tryggflyg and West Air Sweden. However , for many of these firms some of the human (like

managers), immaterial (like licenses or routes) and material (like aircraft, hangars) resources

were transferred to other firms.

Several firms were fully or partially acquired by other firms and thus connected to other

resources. Important for the restructuring of the märket were two sets of changes in

ownership, both involving SAS.

The first is the development of Skyways and SAS part ownership of Skyways in 1995.

Skyways'acquisition or ownership investments during the period Airborne of Sweden, Flying

Enterprise and Highland Air increased its coverage of domestic routes. Skyways grew

considerably on the domestic märket through the alliance with SAS (including transfer of

routes from SAS that SAS considered unprofitable), extension of its domestic route network

through the above mentioned ownership investments, through cooperative agreements with

other firms (Golden Air, Waltair and earlier Falcon Air) and its take över of some routes

operated by firms withdrawing from the märket (Holmstroem, Air Express).

The second development involves Braathens, Transwede, Malmö Aviation and SAS. Through

Braathens' acquisition of Transwede (partly in 1995, fully in 1997), Braathens took över most

of Transwede's routes including Malmö-Bromma, Stockholm-Östersund, Stockholm-Umeå.

Then Braathens also acquired Malmö Aviation and extended its own domestic Swedish

operations to e.g. Stockholm and Halmstadjönköping, Sundsvall repectively. When SAS

bought Braathen's due to the latter's economic problems, Braathens had to sell Malmö

Aviation (to investors outside the airline industry) and withdrew from the Swedish domestic

märket.

A third development of importance was Finnair's acquisition in Nordic Airlink which

operated some routes in Norrland. Finnair's acquisition has been interpreted as a respons to

SAS purchase of Air Bothnia which operates in Finland and internationally.

Entries, exits and acquisitions are important aspects of the reconfiguration of the märket after

deregulation. We have also touched upon alliances and other aspects of cooperation.

8

Before turning to cooperation we will also comment on the development of the set of routes

that individual finns operate, bearing in mind that also this development often is related to

entries, exits, acquisitions and alliances. A general observation is that there is a löt of

flexibility. This has been true both for major routes like between Stockholm and

Gothenburgh, Malmö and Umeå respectively and minor ones like between airports in northern

Sweden and between Stockholm and minor airports in southern Sweden.

Several routes earlier operated by SAS where were handed över to smaller firms, especially to

Skywys after SAS became part-owner. Ät the end of 2003 Skyways operated 32 domestic

routes and had a märket share of 10-15 %. But also Skyways left some routes that were then

taken över by smaller firms like European Executive Express.

International routes and domestic routes in other countries has been added to the operations of

several firms especially during the latter years. An early casels that Flying Enterprise operated

a Skövde-Trollhättan-Copenhagen route, dropped already in 1994. A late entrant, City

Airline, for which the international routes are more important than the domestic, operates

from Gothenburgh to Helsingfors and Manchester, earlier also to London and has an

interlining agreement with an American airline. Golden Air also operates domestic routes in

Norway and Finland. Direktflyg and Swedline Express flies to Lithuania, Waltair to the

Baltic countries and places in Russia. European Executive Express connects e.g. Mariehamn,

Arlanda and Oslo, and has opened traffic between Bromma and Helsingfors. Malmö Aviation

has developed the low-price concept "snålskjutsen" which includes destinations like Nice,

Brussels, Venedig but also the domestic Visby. Nordic Airlink, after its acquisition by Finnair

has expanded to some internordic routes. Skyways flies to several foreign destinations as a

partner to SAS.

A typical feature in the early years was that the activities of the firms enterring regular

passenger traffic covererd many air business activities such as taxi, freight, charter. A few

firms concentrated on passanger traffic when that business grew. Malmö Aviation abandoned

the freight business already in 1994. On the other hand, Falcon Air, being acquired by the

Swedish Post and growing in airfreight discontinued their rather small passenger transport

activities in 2003. Only some very small firms seemed to be able to handle a combination of

many different businesses with small volumes like taxi flight, charter, freight (like taking mail

to different regions) and some minor passenger routes.

Alliances and cooperation

During the period the International passenger airline märket has been characterized by a

somewhat unstable development towards 4-5 global strategic alliances of major airlines.

Related to this development these airlines have developed strategic alliances with smaller

feeder airlines in their own domestic märkets. SAS links to Skyways in Sweden, Widerö in

Norway and Maersk in Denmark are examples of this. Malmö Aviation through its owner

Braathens was associated to the global Wings alliance but change in ownerships (not only for

Malmö Aviation and Braathens) disrupted mat alliance. The large airlines now also have

begun to invest in smaller airlines in their competitors' domestic märkets. This is the case

with SAS for Air Bothnia and Finnair for Nordic Airlink. Also non-equity based cooperation

between large and small airlines have been developed during the period. Golden Air's

cooperation with Finnair, City airlines cooperation with an American airline and Malmö

Aviation's with KLM represent such developments.

This way the national, regional and even global passenger flight networks become

increasingly connected through cooperating groups of firms competing with other groups.

However, there are also less visible, but in individual cases important types of cooperation

involving regular domestic passenger flights. We mention some of those observed in the study

below.

-Flying Enterprise cooperated with Braathens about schedules and ticketing and with SAS

about total price for destinations involving flighjts with both firms.

-City Airline was in the beginning helped in the economic planning by Malmö Aviation but

not yet in synchronising the scheduling of flights from Malmö.

-Golden Air cooperates with Skyways in terms of ticketing and bonus program, charters

aircraft to Gotlandsflyg for Bromma-Visby and operates Ängelholm-Bromma for Kullaflyg.

-Falcon Air code shared its flights withn SAS and sometimes operated scheduled flights for

others

Wang & Evans ( 2002) have classified the different forms of passenger air line alliances into

route specific services, code sharing agreements, joint operations, marketing alliances and

finally the "open sky" alliances. The bilateral route specific services are seen as the simplest

form while the "open sky" is the most extensive. Code sharing is defined as "one partner

assigns its airline designator a code to the flight and its partner. Joint activities involves not

only code share but also baggage checks, honoring tickets between the partners but keeping

the identity of each. The the next stage marketing alliances includes global groups. This is

10

where Står alliance fits. An "open sky" alliance is broader commercial alliance that includes

the other types of agreements and also joint purchasing, maintenance, etc.

Outside of the regular passenger flights small firms reported cooperation in taxi flights,

charter flights and air freight with other firms in the region. E.g. when they were fully booked

or did not have the right size of plane they contacted or recommended a group of other small

firms offering similar services.

Few domestic passenger airlines have alliances with customers. Frequent flier programs,

agreements with local or international firms and other organisations obviously exist. The

recent agreement between the Swedish State and FlyMe is an example of how a big employer

tries to influence the choice of airline for its travelling employees. The airlines with the

majority of their business in taxi or charters flights seemed to have agreements with

customers more frequently. Alliances with customers are also common for airfreight. Many of

the small airfreight firms had alliances with larger freight forwarder firms like TNT, UPS,

DHL.

Discussion

Compared to the network structure before deregulation the reconfiguration has been

substantial. A large number of firms have entered, many have exited. There are now more

routes served and more routes with more than one operator. The dominance for SAS and its

partners is still considerable. In 2003 the combined märket share of SAS and its partner

Skyways was around 75 %. The märket share for the three largest operators (SAS,Malmö

Aviation and Skyways) was 90 %.

The restructuring process can be seen as dependent on

overlapping with entrants coming from other air businesses in Sweden and from other

passenger airlines abroad

- dynamics of cooperation between airlines leading to more emphasis on collective

competition and coordination between domestic routes and between domestic and

international route

enforcement of norms in the institutional setting (court ruling about interlining and

about SAS Eurobonus, development of a bidding procedure for routes in little

populated areas

11

- transferability of resources (including access to routes) from exiting firms or declining

firms to entrants or incumbents that aspired to grow. Entry and exits arerelated to each

other.

- cooperation between small regional firms in overlapped networks (for airfreight etc.)

increased their resources for regular passenger traffic.

small firms were quite flexible in entry and exit from individual routes

Cooperation in various forms between the smaller firms were driven by lack of resources and

by the dominant firm by the need to preserve its dominant position and to become more

effective. In general most of the airline alliances could be classified as a combination of

complementary and overlapping (Hertz, 1996), since they were with other airlines but

covering different regions or being complementary helping each other in the same region.

The small operators developed within domestic regions but also to some extent within regions

in neighboring countries. The overlapping was in this sense also international for the small

firms. The small firms start and leave traffics rather quickly between or within

countries.depending on the situation. The firms seem to be very flexible in starting new direct

flights. These flight might sometimes compete or cooperate with the large monopolist firm or

some of the other large international airline. Therefore internationalisation is very flexible

during the period.

By the end of the period the low cost airlines have attracted new kinds of customers as well

taken customers from traditional operations. In this paper however we have not been able to

focused on the pricing issue or the influence of the international low-cost company Ryan Air

on the Swedish domestic märket.

References

Gomes-Casseres, B.(1996). The Alliance Revolution - The New Shape of Business Rivalry.

Cambridge. Harvard University Press

Hertz, S. (1996) "The Dynamics of International Strategic Alliances- A study of freight

transport companies." International Studies ofMangement and Organisation. Summer 1996

Vol 6 No 2, pp. 104-130

Hertz, S. & L.-G. Mattsson. (2004) "Collective Competition and dynamics of Märket

Reconfiguration" Scandinavian Journal of Management( in press)

12

Johanson, J & L.-G. Mattsson, (1992). "Network Positions and Strategic Action - An

Analytical Framework". In : Axelsson B. & Baston G (Eds.) Industrial Networks - A New

View ofReality . London, Routledge

Wang,Z.H. and M. Evans (2002). "Strategic Classification and Examination of the

Development of the Current Airline Alliance Activities." In Journal of Air Transportation Vol

7,No3

Chapter 9 The growth of small and medium sizedfirms in airline passenger networksSusanne Hert^, Maria Dembrotver and }-jna Norden/öu>

Introduction

:*#4-

The lattet part of the 1990s saw an increasing numbet of airline alliances around theworld. The most important tnotive behmd the alliance frenzy was of course the gradualliberalization of regulations in large märkets such as the North American and the European.

The deregulation was expected to initiate the growth of smaller firms and increasecompetition and decrease the power of the dominating airlines. The airline alliances have,however, rather caused an industry concentration on a larger scale leaving many of the smallerforms outside.

Doganis (2001) mentions the main factors to be behind the push towards theinternational aviation industry concentration: 1) A desire to reduce costs 2) A search for themarketing benefits of large size and scope 3) the need to reduce competition and 4) the"nationality-ruk" making the cross-border acquisitions and mergers virtually impossible.34

How will these factors influence the development of smaller airlines?

The first of the factors mentioned above is the most apparent driving force behind amajor airline wanting to align with a smaller one. This is the case as cost economies of scale onlyexist ät the lower end of the size range. Alliances therefore enable the major partner to benefitfrom the smaller partner's lower operating costs. As airlines increase beyond fifteen or twentyakcraft there are no further significant cost economies, instead other factors become importantcost drivers such as size of aircraft used, level of wages etc.

Increased märket power is an important factor behind alliance formation in general.Alliances can also have a beneficial impact on costs in a number of different ways: Higher trafficvolumes that can lead to economies of traffic density, e.g. through higher utilization of fixedassets; Cost economies which may arise from possible synergies between the alliance partnerssuch as joint procurement of goods and services; Joint purchasing in many areas.33 In the aviationindustry alliance formation has been a way to bypass regulatory barriers due to the "nationality-rule". The fourth driving force, mentioned above, is not so often stated publicly, but a wellknown driving force all the same.

These are all important factors for alliances in airline industry more in general and doesnot necessarily apply to an alliance between a small and large airline. Just in the case of the furstfactor can interpreted as beneficial for the small and large ftrm alliances.

The general concentration of the industry into large multinational groups of alliancesforming world networks is not specifically advantageous to the SMEs in the industry. Since directcompetition with either of the groups would be too costly, the best alternative would be either tostay neutral or to form alliance with either of the groups. However, in order to be competitivethe SMEs need to be able to tie into a larger network, which is difficult without forming any typeof alliance. Therefore it is common for SMEs is to form alliances with a larger airline. Whatimpact would this actually have on the development of the SMEs?

'Doganis, R. (2001) p.71' ibid, p.76-79

117

•stt* r̂*S¥ <3w xv*""*-i • r&A^^%*"&**ffS£S&g!£iig&

We have not seen any research about how this will influence the possibilities andproblems of rhe smaller firms and to \vhat extent the smaller ftrms can take advantage of thissituation and actually grow.

The main purpose of this paper is to study how SMEs in an alliance with a large airline candevelop and grow, with focus on the problems, risks and opportunities attached to such arelationship. There are particularly three questions that will be discussed in order to highlight thepurpose. What different forms of alliances do SME aklines establish? How do these allianceschange över timc? Could the smaller firms be a threat to the larger firms?

The Swedish märket for airline passenger transportation36

The aviation industry is a network in itself, with hubs and spökes. When studying aparticular märket the players' interconnectedness has to be remembered. The firms areconnected to each other in many different dimensions: the infrastructure is used by all airlines,the regional airlines often feed traffic into larger hubs for further transport to other destinations,many airlines own shares in other airlines, there are alliances between different airlines etc.37

Before the deregulation of the Swedish märket in 1992, SAS and Linjeflyg were thedominating participants. SAS operated the heavily used routes between Stockholm andGothenburg, Malmö, Luleå and Kiruna while Linjeflyg rån those routes that SAS were notinterested in. Other than that there was a very limited regional flight service, with routes thatneither SAS nor Linjeflyg were interested in. The monopolistic situation that prevailed was notestablished by law but a well established practice. To be able to start domestic airline permissionhad to be granted by the government. Furthermore, it was forbidden for foreign operators to rundomestic routes. Prices had to be accepted by the the Swedish Aviation Authority(Luftfartsverket). The enterprises were forced by the government to run a regular service,covering the whole country, profitable as well as unprofitable routes. In cases of economic lossesthe firms were entitled to governmental support.

The Swedish deregulation process actually started ät the end of the 80's. The competitioncommittee, that had been appointed in 1989 to investigate what regulatory changes that wouldlead to increased competition, recommended that the eight or ten largest routes would be openedup for competition. Due to capacity constraints ät Arlanda airport the government decided thatcompetition only was allowcd between SAS and Linjeflyg. A month låter SAS bought Linjeflyg.This as well as the recession which had hit härd on the airline business forced the government toderegulate the märket even further. The real deregulation began July l, 1992. All Swedish firmswishing to start a domestic route were then able to, provided permission had been granted by thegovernment. The airlines were able to decide on their own prices. There were ät least sevenexpectations linked to the deregulation and one of the most important was the hope for newairlines to arise. In 1997 it was also decided that foreign operators were allowed to start a route\\ithin the country.

The domestic airlines in the US have, since the deregulation, become competitors on thelarger routes. In Sweden, on the other hand, SAS main competitors have been those with a

j6 This text draws heavily on Hultén S, Alexandersson G. and Nordenlöw L. (1999) De avreglerade marknadernalör långväga kollektiva persontransporter i Sverige, En rapport med utvecklingsperspektivet år 2010, SIKANordenlöw L (1999) Entreprenörskap i industriella nätverk -En studie av småföretagande i avregleringenskölvatten, avhandlingsplan

" Button K.., Haynes K. & Stough R. (1998) Flying into the future - Air Transport Policy in the European Union,Edward Elgar Publishing Ltd., Cheltenham, UK, p.9

[lavc L\w Li~ajbackground in charter as well as freight operations, l rus COUIQ uathose that are involved with passenger transportation are very dependent on the larger äirjines;they often feed traffic to the big hubs where the larger airlines take över. Secondly freight andcharter traffic often use jet-planes while the small regional airlines often use turbo-prop planes.But competition on the Swedish märket can also be seen in another light, a battle between the bigworldwide alliances. (For more information about the alliances, see table l below). From l April2003 also Spanair will be a member of Står Alliance and låter Swiss will become a member ofOne World. Another interesting change is that Ak France actually has acquired the majority ofKLM. This would mean that even some of the smaller state airlines within EU might be in risk of

being acquked by the larger ones.

Changes within these groups take place continuously and this would imply that therewould also be both relational and organizational changes of the regional aklines as well.

Theoretical framework

Alliances between firms have often been an important subject in business theory. Thefocus in this study is on the development and growth within an alliance between a small and alarger partner and its problems, risks, and opportunities. We will both use theory of alliances and

of entrepreneurship to understand this industry.

We start by discussing some basic concepts of the network perspective. Then after a briefgeneral discussion of the motives for alliance formation and development we bring in theposition in the network, development stages, power dependence, positive and negativeconnectedness and uncertainty and risk. We will end with a short discussion about theentrepreneur and critical characteristics that might influence the development of the firm.

An industrial network can be analyzed from four different perspectives. These are thesingle organization, the relationship between certain actors, the net (i.e. a smaller group ofinterdependent actors) and the total network. (Hertz, 1996) The aviation industry can thereforebe studied from the perspective of the single airlines, the relationships between two airlines, alarge alliance of several interdependent airlines i.e. an alliance between nets. Finally the wholeindustry and their connected partners can be seen as the total network in this case.

Establishment of alliances initiates a process of change for the organizations involved inthe network. The company's embeddedness in a context and the opportunities it creates throughboth direct and indirect relationships might even be more important for growth than the singlefirm's abilities. In the aviation business with its necessity of creating interconnectedness thisseems to be even more relevant. Therefore this might be of interest to understand how thealliance partner is connected in order to explain growth of the SMEs.

The motivations for forming or establishing an alliance has been subject to muchresearch. According to Gulati (1998) three main motivations broadly applicable are thetransaction costs of small numbers bargaining, enhancing the competitive position or märket

power and gärning organizational knowledge or Icarning from each other. Since the SME has aconstant lack ot resources all three morives would seem to be easy applicablc.

The rirnVs position in a network can be defined in terms of how it is related to otherfirms of the network or as consisting of its portfolio of relationships and the activity links,resource nes and actor bonds that anse from them. (Johanson & Mattsson, 1992; Ford, et al,1998) Enhancing its network position m the net and in the total network is an important motiveas well as base of strateg)' for firms both in the net and total network. From the networkperspective it might mean to change from existing net to another net. (Hertz, 1996) The reasontor startmg an alliance is not the focus in the network approach but rather how the alliance isformed and how it develops över time. This has to do with the essence of the network view, thatrelationships are normally long term. There is a combination of stability and change in theexisting relationships. Both are necessary for a relationship development. (Johanson & Mattsson,1992)

Wc have also seen that relationships go through different stages in their development.(Dwyer, Schurr, &Oh, 1986; Liljegren, 1987) Development of a relationship is normally a gradualchange that deepens and expands över time as trust develops and new possibilities to increaseetfectiveness and efficiency are revealed. The more the two parties interact the more they get toknow each other and the easier it becomes to transfer knowledge between them. Ät the sametime, the more closely related the businesses are the easier it becomes to get further involved,creating tighter relationships. This increased integration as well as the unwillingness to change hasbeen seen by many researchers (Ford, et al, 1998). However, it often ends up in a stagnation andinstitutionalization, which then might turn into a decrease in commitment and a break of therelationship. The decrease in commitment is often a result of change in other relationships of thenetwork (Hertz & Mattsson, 2001).

Thercfore not only the small atrline but also the large firm would have to expect be anincreased depcndencc över time but also be many conflicts.

Wang & Evans (2002) have made studies of the large passenger airline alliances and havecalssified the difterent forms of alliances which shows different degrees of integration. These areroute specific services, code sharing agreements, joint operations, marketing and finally the "opensky". The bilateral route specific services are seen as the simplest form while the "open sky" is themost extensive. Code sharing is defined as"onepartner assigns its airline designator a code to theflight andits partner1'(ibid, p 79). Code sharing often includes one airline buying block of seats from anotherairline and reselling them. Joint activities involves not only code shares but also baggage checks,honorouing tickets between them but keeping their identky. The the next stage marketingalliances include global groupings. This is where Står alliance, Oneworld, etc fits in. The "opensky" would be a much broader commercial allainces that include most of the other types ofagreements but also joint purchasing maintenance, etc.

However, these types of alliances are used between the large airlines. We do not know towhat extent any of these types be used in an alliance between a small and a large airlines.

As for the choice of partner, the firms' embeddedness in the network will hinder as well asfacilitate the formation of certain relationships and the organizations are therefore seldom free tochoose their partner. Most companies are already tied up in existing relationships and nets andthese will consequently have to be broken or changed in order to establish a new one. Throughthe network the organization will also be able to gain information on another firm, through director indirect relations. This advantage might be turned into a disadvantage in certain situationswhen the network might realize what will happen in advance in based on the what information

they have from the network connections. (Johanson öc iviaussxju, i^~, *...- .the relationship with a certain firm and maybe trigger reaction from other organizations to what.is going to happen and maybe causing chain of reactions (Hertz, 1998).

Earlier studies of alliances have often distinguished among alliance structures in terms ofthe degree of hierarchical elements and issues of control and coordination attached.

The formation of contracts and ownership are seen as vital for managing the behavioraluncertainty. (Pisano, Russo and Teece, 1988) According to Pfeffer & Salancik (1987) theuncertainly does not have to lead to ownership, but rather to make use of other ways to decreasethe uncertainly such as commitment through Board Membership. Being in a deregulated businessfull ownership does not seem to be acceptable for the dominating firm so alternative ways wouldbe more likely in coping with the poiver dependence.

In the network approach, dependence is expected to be mutual and discussed in terms ofasymmetry or symmetry of an alliance. A very high asymmetry would imply certain problems inthe relationship. While the firm having little power or a less important position would have largerdifficulties in realizing the benefits the relationship would also be more difficult to manage.(Easton, 1992) This would have implications for small airlines and their realization of the benefitsof the alliance.

In the network approach trust between the organizations is a vital concept that leads todevelopment of the relationships (Young, 1992). Since there can never be a total control and thecoordination the relationship organizations must trust their partner to voluntarily live up to thedemands set for the alliance. Coping with power dependence and building trust would be two keyissues for a small and a large firm forming an alliance.

In most cases, a joint venture is not recommended when there is a potential conflict ofinterest between partners. (Gomes-Casseres, 1989) This seems to be important for the caseswhere competitors are forming alliances. Furthermore studies show between competing firmsalliances have a high frequency of failure. (Bleeke & Ernst, 1995) Related to the question of analliance being complementary or competitive are the concepts of positive and negative mnmctedness,which are important dimensions of networks. Relationships are positively connected to the extentthat exchange in one relationship increases the likelihood of for exchange in another relationshipwhile negative connectedness involves a decrease instead. (Cook, 1982)

The same reasoning is possible to apply to a net consisting of a large number ofinterdependent relationships.On net level it is rather a question of complementarity or overlap in the

nets.

Partners are often chosen to cover a complementary geographical area or to havedifferent services in the same geographical area. (See figure 1) In areas of overlap there will beconflicts of interest between the partners which could create problems for the alliance ending upin its breaking if not solved. (Hertz, 1996) A typical example could be in the formation of analliance between two large International airlines. The partner's existing nets can be eithercomplementary or overlapping in services and/or coverage.

Different

Geographicalcoverage

Same

Different Same

Fully complementary

Complementary/overlapping

Overlapping/complementary

Fullyoverlapping

Fig l Overlap and complementarity in networks (Source: Hertz, 1993).

Thereby we can expect overlapping organizations, which actually are serious competitors,and less successful when creating an alliance than the others. The most common and dynamicsituation for alliances are a combination of complementarity and overlap. (Hertz, 1996) But youcan find examples of fully overlapping alliances in the märket like Står Alliance or OneWorld.The effects and consequences of change processes in networks are totally different depending onthe situation of the partners and the their expected future changes. Is it moving towards a higherdegree of overlap or is it moving towards a more complementary type of overlap? The smallerfirm will probably have to adapt in order be more complementary and less overlapping in orderto have few potential conflicts.

Entrepreneurship

According to Tsang (1998) studying alliances of smaller firms, uncertainty and risk are largerfor small firms. Therefore they are more likely to use alternatives mat mean a lower resourcecommitment. He also mendons that small firms also choose other types of alliance alternativesthan larger firms and that the possibility to absorb the knowledge from the partner is important.

A young organisation is often by definition small and lacks the financial means to takerisks. The entrepreneur is by definition an actor that seeks new and often unconventional ways todevelop the organization. Even though the entrepreneurship mostly is connected to technicaladvancements (Schumpeter, 1989) much of what is done is also tied to the entrepreneur as aperson. The most common characteristics for a successful entrepreneur are to have goodknowledge and self-confidence, to have technical competence, capital, individualistic, a strongability to imtiate projects and so on. One question here is if these abilities are sufficient to reachsuccess in the airline industry? Maybe the most important thing here is to have the ability tocooperate and to maintain the relations in the network, instead of just acting as an individualisticsntrepreneur. Therefore when forming an alliance with a small firm by necessity has to includetrust and a good social contact with the manager of the firm.

To sum up, we have found particularly three elements that are of importance whendiscussing different types of alliances using a network perspective: complementarity andcompetition in the alliance, power dependence in the relationship and degree of integration. Ontop of this there is a fourth factor that in a way combines all of the others when studying small,newly established organizations as well as alliances. This factor can be referred to asentrepreneurship and has to do with being able to see and use chances that are given to the

iiicrciuic öu»-ii in.organization in relations with others in tne auiance.background in the network and influential relationships över the years that is of importance.

Empirical part

We have selected five early starters among 16 studied firms on the Swedish märket in thederegulated märket and we have totally made 10 interviews with these companies between 1994and 2003. In this study we have focused on their alliances with other airlines and the problems,risks and opportunities met on their ways of growth. These firms have been studied in theirrelations to the bigger firm SAS to describe the firm's situation on the märket. One of the firmshas now been bought by a larger company which could give us an interesting description of themärket development. Many studies have been done on SMEs as well as on that of alliances butvery few have, combined the two. Many studies have also been done on the aviation märket,particularly concentrating on the North American märket, but very few have focused on thesmall firms operating in this field. Therefore our study has been based on a few case studies ofsmall airlines operating partly on the Swedish märket. All except one have a relationship withSAS, the dominating airline on the Swedish märket. (These firms are presented below). The studyis focused on Sweden and airlines operating in this country, but also countries nearby and airlinesoperating in the vicinity play a large role.

Hr

Air

Air

Ire

SASk)

m Tijrnover

Baltic 1, 86 Msek

Botnia 974 Msek

andair 400 Msek

S 51400 Mse.ways 1500 Ksek

300

2000k 3 100(1

700

11

185029

_

250 000

1, 5 milj17 milj

990 000

2003

Table 2 A snört presentation of the studied airlines in numbers. Website information 2003-02-02

Air BalticAk Baltic is a fairly small airline that operates in the Latvian märket. The home märket is

too small to harbor a large airline and to be able to enlarge the märket a partner is essential forgrowth. When a privatization of the publicly owned Latvian airline was discussed in thebeginning of the 90's, SAS was presented as a possible partner. SAS' ambition was to find atransitstation for traffic to Eastern Europé and Ak Baltic was interested in finding a partner. AkBaltic has therefore been a regional partner to SAS from the start in 1995 and is also owned bySAS to 47, 2 %. The company's largest shareholder is the Latvian State with 52, 6 % and alsoTransaero owns O, 2 %. Air Baltic represents SAS in Latvia and it feeds traffic in to SAS' mainhubs and is furthermore responsible for all transit traffic in the country.38

When the publicly owned Latvian akline went bankrupt it was decided that a swiftsolution had to be found, an interim company was therefore set up. During 1993/94 discussionsstarted about privatizing the company and in 1995 a few investors (one of them SAS) had beenfound that were willing to finance the project. Today the akline is still owned by those thatfinanced the privatization, with the Latvian Government and SAS as major shareholders. Theakline started out with four different destinations with Riga in the center, flying to Copenhagenand Stockholm. Today these lines have been accompanied with several others, Warsaw, Kiev,Helsinki, Tallinn, Vilnius, Berlin, Vienna and Prague (from June 1999).

What could SAS give them to put them into growth and enlargement of the märket? SAShad a good financial situation ät this time, an organization with a good structure and a qualified

www.sas.se

administration. All these tactors helped them to tmd growth opportunities and to enlarge theirmärket.

Air Baltic also had other partners apart from thegrowth opportunities. Accordmg to Air Baltic the mosthave dcvelopcd e.g. code sharmg agreemcnts3'' with, toothers that they code share with, with Lithuaman Air towith Malov to Budapest. Anothcr important codeMoreover they have a few "pool"co-operations, whichtheir passengers in the othcr akline's plane, with Träns/Kiev, Riga and Scandinavia.

alliance with SAS which give them goodimportant one is Estonian Ak, that they

F lamburg and Tallinn. But there are alsoVilnius, with Tjeck Airline to Prague andshare partner is Polish Airline LÖT.*means that a few seats are reserved for

\ero to Moscow and with Aero Sweet to

keland airIceland air was established right after the Second World War. In 1982 the two regional

airlines operating on Iceland, were merged into one larger business. Iceland air , which is amecliumsixed airline, has always been big on trips to the US and the akline has used the country'sposition in the middle of the European and American continents in thek favör and created acompetitive product that has become thek major one. Iceland Air has been a member of IATA(International Air Transport Association of European Aklines) since 1950, and has been amember of AEA (Association Flight Safety Foundation since 1966.)41

Today the company operates 12 destinations, mainly in northern Europé and Scandinaviaas wc 11 as the US thanks to agreements with other akline companies. The company is 100 %privately owned by about 5000 Icelandic shareholders and it's the largest privately ownedcompany, employing över 2000 people in Iceland. Iceland ak is one of those rather rare aklinesthat exist almost without any partner. Iceland Air has one code sharing agreement with SASwhich means that they are able to offer passengers travelling to and from Europé to Iceland.Othcr than that they have some special product agreements with a few aklines on certain routes,e.g. with Braathens, Skyways, TWA, Lufthansa and British Midlands (overlapping networks).

Flying EnterpriseFlying Enterprise started as a subsidiary company within another fkm and somehow

along the way grew ancl became Flying Enterprise. The parent company was called Tipp &Ivrantransporter which worked with transportation of e.g. cranes. The aviation part of thebustness began as a flying taxi-service but was shut down in 1988. In 1992 when the whole aklinemärket was struck by the heavy recession and ät the same time deregulated, Tipp &Krantransporter decided to start its aviation business. Golden Ak, a small Swedish akline, was onthe edge of bankruptcy and the business was for sale. Flying Enterprise, which was the divisionwithin Tipp & Krantransporter that handled aviation, had the opportunity to take över thek threeplanes as well as thek route between Skövde and Arlanda. To be able to manage this, the hangarhad to be built ät Skövde akport. Flying Enterprise was then asked by the township of Skövde torun the akport as a contractor. This development meant that Flying Enterprise was really into theaviation business once again.

In the early 90-ties Fying enterprise found new growth opportunities on the märket. In1993 this which resulted in the growth of two more routes - between Trollhättan and Arlanda as

w Code sharing; A system by which two or more airlines agree to use the same "designator" or flight number fora flight or series of connecting flights in order to attract more business by extending their partner carriers. Thiscooperation ean include reciprocal sales, marketing parts, coordination of fares etc. Source; Collis Roger,International Herald Tribune.40 www.airbaltic.com41 www.icelandair.com

well as between irounan.au auu ^j.^^^..» _Skyways. In 1995 two additional routes were introduced, Jönköping - Bromma and Jönköping •Arlanda. In 1996 the route between Bromma and Visby, that Malmö Aviation used to have, wasopened. It now represented the largest route for the company with around 100 000 passengers.Thek märket share was nearly 45% on that märket. The route between Halmstad and Arlandawas also opened that year in competition with Braathens, but it had to be shut down in 1999. In1997 the company was introduced ät the stock exchange where it succeeded in securing 24million SEK. One of the largest independent operators in Sweden ät that time had started out asan ak taxi service fkm. They found thek niche away from large operators taking on an offergiven to them a few years back to take över the management of a small regional akport from the

local authorities and some of the routes to and from that akport.

There were three factors influencing the development of Flying Enterprise. The processdeveloped as follows; fkstly they started out as an ak taxi fkm and focused on a number ofdestinations which made them to growth very fast. Then they wanted to become more neutralwhich caused them problems as they still competed with other akline companies. There were alsoproblems with cooperation with two partners ät the same time which mainly caused problemswith the double time table. To manage this they had to have more resources. Thkdly they talkedabout cooperation with Finnak, which made them a threat to the other actors. These structuralobstacles in the envkonment made it difficult for them to manage thek business into thedkection that was necessary for them to get the company into a successful position.

In year 2000 Skyways Holding AB acquked 900 000 A-shares in the akline company and71 000 B-shares which corresponded 63 % of the 94 % voting rights. Ät a meeting in year 2000the Board of Dkector of Flying Enterprise recommended to sell the shares to Skyways HoldingAB which now became the main owner of the company. The reason for him to sell was that "itwas the right time to sell after working in with the company for so many years".42 Through thesale of the company, the "old company" could start to operate in a network of Skyways Holding

AB.

Air BotniaThe company was founded in 1988 by a few people that were kritated on Finnak for thekinsufficient amount of departures from a certain region of Finland. After a while the companystarted to feed passengers into Finnak's hubs. In 1989 the company was sold to SAS. SAS haddefined the Scandinavian countries as thek home märket while Finland, the Baltic region,Scotland, Iceland, the northern part of Germany and the northern part of England had beendefined as the vicinity. The Finnish märket was therefore considered very important but it wasfelt that thek grip on the märket left more to be wished for. The Finnish akline, Finnak, thatused to have a close relationship with SAS had been on the offensive in Sweden and SAStherefore reacted through the use of Ak Botnia. In March 1999 Ak Botnia's strateg)' was changedand the new aim therefore was to develop the akline into a regional akline instead, feeding trafficbetween points in southern and western Finland and in SAS's hubs in Stockholm, Copenhagenand Oslo. As a result of these operations they started the cooperations with SAS. The newstrategy was fully implemented in November 1999. One of the problems Ak Botnia met was thatthey became the competitors to Finnak as they ät the same time started to operate together withSAS. In what dkections should thek relationship now develop and how should the Finnishmärket look like? Ät the same time there was a growth opportunity for them that SAS became apotential partner for them and they could start hubbing destinations both in Finland and inSweden. Through these operations they could easily take larger märket shares in Finland. The riskwith starting cooperating with SAS was that they could become more dependent on SAS's

42 www.skvwavs.se. 2000-01-18

125

business and that their actions should have a strong influence on Air Botnia. Today Air Botniaserves the destinations in Finland: Helsinki, Jyväskylä, Oulu, Tampere, Turku. In Scandinavia andnorthern Europé: Copenhagen, Gothenburg, Oslo, Stockholm.43 The companv is owned by theFinnish OY Nordair AB which is wholly owned by SAS. All flights operated by Air Botnia arecode shared with SAS and Euro bonus members earn points on these flights.44

Skyways

Skyways started during the 40's with activiries for the military. The company was thencalled Salair after the founder Sven Sahlén. It was not until 1992 that the present history of thebusiness started. Three routes were then introduced that operated from Bromma Airport. Theyear after the home base was changed to Arlanda, Bromma was certainly closer to Stockholm.Still Arlanda was a traffic hub and had the ability to carry people further using other airlines.

Today Skyways is still mainly a family business, 72% of the shares are yet owned by theSahlén family. Along the way Skyways has overtaken smaller airlines, Highland AB in Hultsfred,Airborne AB in Sveg, Air Express and Flying Enterprise (2000), which all operate in a small nicheof the märket. It now depends on three main lines of business. 1) Transit traffic, wherepassengers are fed into a larger network, e.g. such routes as Linköping to Copenhagen orBorlänge to Copenhagen. 2) Regional traffic, to provide a certain region with travelling service.To fly might not be the only alternative but it is a time saver, e.g. Arvidsjaur to Arlanda orTrollhättan to Arlanda. 3) Hub-by-pass, direct flights that are not provided for by the big hubsusing smaller and faster planes than usual, preferably Embraer jet planes. Such examples areroutes between Helsinki to Gothenburg and Brussels or Gothenburg to Paris.

In 1997 Skyways became a partner with SAS. Before that the airline had cooperated withmany different airlines, such as Sabena, KLM, Ak France and so on all but SAS. Earlier onSkyways had asked if SAS was interested in cooperating, but Skyways had received a negativeanswer. Suddenly this had changed and Skyways were contacted by SAS which now wanted tocooperate. This time Skyways could set the conditions. There was one condition though that SASput up, all of the other cooperative agreements that Skyways had would have to end.4

The cooperation with SAS has involved many different levels in the organization, such asco-branding, lounges, code sharing, ownership (25%), Euro bonus and so on but also e.g.inventory service and computer systems. The partnership has worked very well according toSkyways and together the two companies have tried to expand the märket instead of stealingpassengers from each other. Skyways has been able to take över such routes that SAS has notfound to be profitable enough, such as Wasa to Arlanda, Borlänge to Copenhagen and Kramforsto Arlanda. Just now they are waiting for more routes that SAS doesn't find profitable enough.4'

The dependence to SAS could both give Skyways growth opportuniries and risks in theirbusiness development. To be too dependent of another actor made Skyways both vulnerable andsensible of how SAS is acting and every action could have consequences of Skyways' work. Anexample of this concerns their tight cooperation with SAS:s bonus system, which even lead thetwo companies to accusations of "cartel cooperation". This was put to a stop from Jan 2002 bythe Court in Sweden. Skyways tried to find growth opportunities when operating so close to SAS.They had the possibility to develop their business through taking över routes that are notprofitable enough for SAS. Through this cooperation they can easily get new märket shares in adiscrete wav.

' www.airbotnia.comwww.airbotnia.com

' www.svenskallygbolag.com46 www.svenskaflygbolag.com

SAS and cooperation with SASSAS was founded in 1946 as a consortium of the national airlines of Denmark, Norway

and Sweden. The three parent companies are owned fifty-fifty by private interests and therespective governments. The main hub is Copenhagen Airport with StockholnYs Arlanda andOslo's Fornebu as regional hubs. During 1995-1996 SAS signed comprehensive, bilateralcooperation agreements with Lufthansa, Thai Airways International, United Airlines and AirCanada. And in May 1997 SAS and its four partner airlines established Står Alliance. In October1997 the Brazilian airline Varig also joined the alliance. And in 1999 the alliance partners wereaccompanied by Air New Zealand, Ansett Australia and All Nippon Airways. Together thesecompanies have flights to more than 700 destinations in 110 countries. For a relatively smallinternational operator as SAS, cooperation is vital. The company lacks the means to serve all ofthose destinations that they presently attend, but with the help of a partner it is possible to builda worldwide network. The primary motive for the global alliance network was to be able to offerthe business traveler a worldwide network of destinations, a "seamless"travel. Today the goal hasextended and, among other things, the alliance is said to have the following advantages: 1) Moreand more destinations, offering a more dense net of places to travel 2) Acknowledgment, thedivision into two separate categories of passengers makes it possible to recognize thosepassengers that are in need of special attention 3) The passengers will receive bonus on moretrips 4) More special lounges for those that are members 5) Members have the highest prioritywhen checking-in or when put on a waiting-list.

SAS is also involved in other forms of cooperative agreements, with regional carriers aswell as with traffic system partners. These are not as far-reaching as contracts with the alliancepartners but they involve such matters as code sharing, Euro bonus, mutual check-in andmaintenance. SAS regional partners are; Cimber, Widerae, Skyways, Air Botnia, the Latvianaviation Ak Baltic, Greenland ak and Maersk Ak. All these companies belong to the concept"Well connected with SAS" which will quality-assure and harmonize with thek regional partners.Traffic Systems partners are; British Midland PLC, the Spanish charter company Spanak, Icelandak, South African Airways, Singapore Aklines, Regional Aklines and Estonian Ak. Braathens is asubsidiary in the SAS group.

The small regional aklines all have thek own contract with a står alliance partner.Together these form a loosely coupled network with quite a few smaller regional aklines that feedthe traffic into the different hubs of the står alliance partners, e.g. Bangkok which is the main hubfor Thai Airways International and Sao Paolo which is Varig's main hub. The regional aklines aretherefore only indkectly connected to all other står alliance partners. An exception is Maersk Akthat feed traffic to both Lufthansa's hub and to SAS' hubs.

The regional partners are free to decide on thek own activities as long as these are keptwithin the boundaries of the cooperative agreement. A partner must have the best interest of thecooperation in mind; no one can expand ät the expense of another partner's märket share. Newlines must increase the effect of the cooperation; they may therefore not compete with SAS'lines. To ensure this SAS has been forced to buy a part of some of thek partners, such as theaffiliated aklines (less than 50% ownership): Ak Baltic 47, 2 %, Ak Greenland 37, 5 % andCimber Ak 26 %, Skyways 25 % and BMI 20 %. The subsidiary aklines (more than 50 %ownership) are Spanak 74 %, Braathens 100 %, Wideröe 99, 4 % and Ak Botnia 100 %.

After the entrance of companies with low-cost concepts on the Swedish märket, SAS alsointroduced a low-cost concept in 2003 through the subsidiary called "Snowflake". This subsidiaryhas the same business strategy as the other low-cost companies on the märket (with low prices

and self-paid service onboard) and will offer non-stop flights to destinations in the southernEuropé like Athens, Barcelona and Alicante (with a main airport as a destination).47

Discussion

The discussion about the advantages and the disadvantages of the different forms ofcooperations, their problems, risks and opportunities will be focused on four basic dimensionsderived from the theoretical framework. These are the uncertainty and risk for the SME, the powerdependence between the alliance partners, complementarity and conflictl competition and the degree ofintegration of the alliance. Based on these dimensions we discuss different types and effects of thealliances befween the SME and the larger airline partner.

The combination of the four dimensions will also be utilized for identifying why a specifictypc of alliance is chosen and the problems, risk and opportunities with that particular type ofalliance. To what extent could they be tnfluenced by changes of the large alliances? Studymg thedimensions more closely we find that the first two dimensions are mainly tied to the fäet that thestudy focuses on SMEs in the airline industry while the other two are applicable to alliances morein general. Our aim is therefore to formulate a number of ideas about growth and alliances ofSMEs and the effect in terms of problems, risks and opportunities based on our analysis in thestudy.

Uncertainty and risk

We know from the empirical material that starting a new business in the airline industry ismuch more expensive than imagined and more importantly it often costs more than what manyot the entrepreneurs first expected. The main cost is not to lease the airplanes, to set up amaintenance system, to create a safety system or find the right staff even though this can be veryexpansive. It is rather to find the right slöts and to create a märket or develop a märket share. Toget a loyal customer group the small airline must be persistent in its marketing and accept lossesfor a long period. The airline has to fly according to schedule even though there are very fewpassengers on board. The estimated costs of starting an SME in the airline industry variesbetween 50-100 million SEK according to our sources in the empirical study. In spite of thismany small airlines have tried to start with much less.

The network approach underlines the need to form the right relationships in order to stayon the märket. It is important for an SME to create a local customer base, which can be donethrough the contact with some of the more important local companies. This cannot be done untilthe operational and safety systems are in full order. Creating these systems and getting the rightcertificates take a certain time during which existing competitors can work intensively on themärket. Since the SMEs usually work on smaller and local märkets the risks are even larger sincethe vulnerability is much higher. When the smaller firms not cooperating in alliances decide toexpand abroad the most common for them is to have a simple expansion strategy which meansthat they try to find new routes just between two destinations helping larger firms to feed trafficto main airports. This strategy puts the small firms in a risky and uncertain situation; they getdependent on the larger firms which make them even more vulnerable for changes in the totalsystem. This is increased uncertainty and risk and need to creative in the type of alliances formedare very much in line with reasoning of Tsang (1998).

Power dependence or asymmetry-symmetry and trustThe result of a study on the distribution of power between units is to a large degree

dependent on what focus that has been chosen. In this case we are not only discussing and

www.sas.se

comparing the big and the small company but also the large world wide alliances compared tö the . •small firms. SAS has in most cases acquired or tried to acquire ät least a minority share in theSME alliance partner. This way SAS has been guaranteed a seat on the Board of Directors of theSME airline and has therefore a possibility to influence the development of the company as wellas keeping informed about the development of the SME. Having a seat on the Board is acommon way used to influence an alliance partner (Pfeffer & Salancik, 1987). The dependenceon SAS and the network it represents is strong for the SME.

The most important question however, is perhaps not the power dependence as such buthow this dependence is handled both by the big firm and the SME. For SAS the closerelationships with some of the regional SMEs have guaranteed that their development coincides,keeping the home märket calm. SAS is able to pkn together with their regional partners forfuture marketing campaigns connecting the network even further together. For the SME thealliance provides a märket without competition from "the large firm", on the other hand thealliance put restrictions on operations and märket expansion plans. The weaker position of theSME therefore also means that there will be some difficulties to manage expansion of the firm byrestrictions of the larger firm. (Easton, 1992). These problems get intensified when a small firmwants to collaborate with larger airlines of another world wide alliance group.

Complementarity and competitionTo what extent are the partners in the alliance complementary or competing? How can

you categorize the different types of alliances? Two important dimensions of complementarityand competition concern the product/ services offered and the geographical area covered.(Hertz,1993) In all cases the SMEs studied are offering similar types of services as SAS but in a differentgeographical area. Thereby they are complementing each other geographically but overlapping inservices offered. The overlap in services is a necessity since the customers are supposed tocontinue into the SME:s airline system from the SAS system to meet a similar service,

What will happen when the structure of the alliances changes and could the co operationsin alliances lead to higher competition or not? The main problem in a situation like this is theSME's restricted position. SAS has demanded that What would happen if an SME like Air Balticor Iceland ak would start to compete with SAS geographically as well? What would happen to

their relationship?

One of the reasons given by SAS to cooperate with the SMEs was to increase thegeographical coverage and enlarge the existing network, which is a complementary situation(Hertz, 1993). Often the alliance with SAS includes a package of different exchange of serviceslike handling agreement, booking, etc which in terms of Wang & Evans ( 2002) classificationwould be route specific services. However, being a small dependent firm these agreement can beof a more general type where the larger firm with its coverage can take on these services for morethan one or two routes. A problem connected to this situation is that the SMEs have to live up tothe promises made by SAS concerning different dimensions of quality such as punctuality andservice. They are therefore restrained from working with price as a variable in the cooperation. Ifon the other hand, the quality (frequencies, punctuality, service on board, check in services, etc)of the SME is much higher than the quality offered by SAS, each firm won't get credit for thatwhile they have a restricted part in the larger system. The function of one link will not change theattitude of the network as a whole. The customer evaluates the entire system and not each one

separately.

An interesting phenomenon is that partners in a network play different roles in differentsituations, sometimes appearing as partners and sometimes competing with each other. Iceland

air cooperates with SAS to and from Europé to keland but competes from Iceland to the US.Furthermore Iceland ak is a supplier of the aircraft Fokker to many airlines including SAS.

SAS wants to manage the development of the SMEs, not only to hinder competition, butalso to kicrease competition with other large alliances. In the case of Air Botnia for example,which is a wholly owned company by SAS, the company was bought to challenge of the attack ofFinnair ät SAS' home märket. Through Ak Botnia SAS could offer traffic to a reasonable priceon Finnair's home märket. This way SAS are creatmg regional smaller networks with itself as thenode of the network. These regional network compete and are overlapping with Finnak regionalnetwork.

What are the problems, risks and opportunities? One problem is SAS' overshadowing sizeand dominated position compared to other fkms operating on the Swedish märket and thereforeits position as märket leader. It manages and wants to continue to manage the Scandinavianmärket through regional alliances and part ownerships of the SMEs. The SMEs have limitedpossibilities to develop into new areas and take on business where there is competition eitherfrom other regional partners or from SAS. For more or less independent aklines this makes iteven more difficult to enter the märket. One of the few possibilities open for an SME in Swedenis to expand abroad. This is also something that has been done by quite a few of the fkms. BothAk Baltic and Iceland ak have established International routes, some of which are outside SASsphere of dominance. This way Ak Baltic gradually becomes less dependent on SAS and candevelop indkect connections to other groups. Another way for an SME could be to feed trafficinto two of the large alliance partners like Maersk or SAS.

Degree of integration

There are a number of different types of alliances and the cooperation however mightcover many different areas. In three of the fkms that we have studied, SAS has acquked a shareof the company after a period of time. In the case of Ak Botnia for example SAS acquked a 100percent of the firm, only a year after its establishment. This is different to the alliances presentedby Wang & Evans (2002) since SAS have in most of the cases tried to acquke ät least a smallerpart of the SME. It may not be a scale for increasing commitment but radier selected servicesand code sharing and marketing in common.

Out of the five irterviewed companies, Ak Botnia is one of the smallest but ät the sametime the closest to SAS in terms of how its marketing is handled, thek booking systems, thektraffic systems etc. Ak Botnia is highly integrated with SAS and the ties to SAS are very close, themanagkig dkector is still a SAS employee but has been lent out for a time to develop the newfirm, just as he did earlier with Ak Baltic.

The degree of integration between Ak Baltic and SAS is not realiy as tight as the allianceberween SAS and Ak Botnia. SAS has been a part owner, 47, 2 %, from the start of the firm andhas also helped in the process of starting the fkm. In the beginning Ak Baltic lacked almost allthe necessary knowledge about akline business why managers were sent över to Latvia in orderto help the enterprise become a märket oriented company, to create the necessary safety system,to ensure the quality level and to get the operation work efficiendy. Since the home country is tooimall to offer any domestic traffic the firm has developed internationally. In order for this, AkBaltic has developed relationships to other large fkms operating in the Baltic States such as.ufthansa and Estonia Ak. Ak Baltic is therefore getting less integrated with SAS and morentegrated with other fkms. The dependence between Skyways and SAS is in a way increasingiver time as the two companies work thek parallel märkets, Skyways working as SAS' subsidiary)n the home märket. In 2004 Skyways is among the largest aklines in Sweden next to SAS.

As we can see the degree of integration varies for every alliance. It might include such • •things as code sharing, bonus programs, freight service etc. SAS is however working on a morestandardized program for integration on the marketing side. Lower forms of cooperation areadapting the time schedules to each other. Higher forms also include using the same systems,having managers trained together, exchange of personnel, social contacts on all levels, handingöver information to each other etc. A strategic partner in an alliance group, such as the Ståralliance, is somewhat a different thing, the partners are equals and in that kind of alliance it ismore a question of give and take. Therefore what Wang and Evans ( 2002) called the "Open Sky"will hardly ever be possible in the regional cooperations between the large international akline

and the small regional aklines.

Conclusions

In this study we have focused on the situation (roles) of the small fkms in the allianceregarding problems, risks and growth opportunities attached to the relationships between small

and large firms.

All of the SMEs have different roles depending on the situation and the type of alliance.The role of suppliers to SAS often concerns operations. An example of this is when Iceland akservice SAS' Fokker akcrafts. The SMEs function as customers when they use SAS' servicesystems for check-in, for booking, ground handling etc. This can be part of the partner deal but itcan also be a specific need of the SME that can be satisfied by other aklines equally well. Thepartnership to a large extent deals with marketing issues. Examples of markering activities couldbe advertising where the SMEs name is included offering code sharing in order to be wellknown, making customer deals together, offer more efficient traveling through combining time

schedules, offering the same price packages etc.

Could the SMEs also function as competitors to a dominating actor as SAS? The SMEsact as competitors when they establish new routes in competition with thek partner. SAS can onthe other hand make use of this alternative in thek role as alliance partner and offer thisalternative to thek customers by code sharing. Still we ask if working in alliances has given thefirms lower expenses and if there is higher competition between the operators on the märket.

Growth opportunities of an SME in the akline industry can enfold as follows:

First stage might include the development of a märket niche of its own. Normally theSMEs have to find a suitable niche, a route can only handle two competing fkms if the numberof passengers exceeds 400 000 a year and there are quite few of those in Sweden. Finding asuitable niche can be done via developing or talong över routes from another firm that are lessattractive for larger aklines like Skyways, establishing new direct connections between twoakports, this has been done by Ak Baltic, Iceland ak and Skyways for example. Finding newakports and getting support by the local governments to expand the märket, is another way tofind a niche. The niche has to have a potential to grow so that the akline can expand its activitiesand create a network of relationships with customers, suppliers, operating partners, etc necessary

for continued development.

Second stage might include the growth of the company that could make the new SMEattractive on the märket and of interest for an alliance. The alliance with a larger akline is thenformed based on the existing position and development of the SME niche. However, thealliances vary with the type of SME. Some alliances are rather loosely connected from thebeginning developing with the relationship över the lifecycle, while others are based on a closerrelationship. The risk lies in the power dependence of the large alliance partner, which might

rcstrain thc smaller company's activities. Thercby forcing the company to stay small. Throughcombining the different rolcs of customer, supplier, competitor and partner, the SME might beable to increase both its know-how in marketing as well as operations and to establish a numberni ne\v usctul relationships. Often the partnership means better marketing resources, moremarkering activitics and more passengers, which implies a continued growth.

Third stage includes stronger posinoning in the network. If the SME is able to handle itsdifferent roles the right way it might get a much stronger position in the network. This meansbetter conditions vvhen renegotiating the deal with larger party or a possibility to leave the allianceand hnd a ne\v alliance partner. So far this third stage is just a speculation, even though we cansce that some of thc partners have grown in size and importance över time and becomeprohtable. This is not to say that the SMEs that we have studied are able or willing to break awayfrom SAS. SAS is a minority owner in most of the companies studied, except for Air Botnia andSpanair, and are therefore not in full control of their actions. Through establishing operationsthat are separate from SAS some of them are working on the creation of certain independence.

What we have seen herc is that the firms might develop through three stages, fromJcveloping a niche and getting the company in growth, to putting the firm into a good position inthe alliance. A collaboration is needed between large and small firms in order to for the largetirline to create a regional network and for the smaller airlines to be able to offer the passengers:onnection to the large networks. Since the actors would be unwilling to leave their globalilliances, ways of influencing the larger alliances could be to cooperate and to operate inregionally rather than try to be neutral like Flying Enterprise with minor collaboration to more:han one large airline. This was costiy and turned out to be very risky. In the end they did not;urvive.

(kooperation takes place on several levels and change gradually över time. Some firms:ould change their position between different levels in the alliance, which could have:onsequences to other actors in the alliance. Small firms' changes from one level to another couldx* threats to the larger fkms and their changes could even influence larger operations, which•ould have importance for the whole industry. A practical example of this phenomenon is whenlew low-cost companies has »ntered the märket outside the alliance, taken passengers from theraditional ones, and has influenced the larger firms to create new businesses with a similar>usmess strateg}7 as the smaller ones. (Like the introduction of Snowflake in SAS)

The SME relation to the partner could sometimes complicate their performance.komplex systems and rules emanating from the partner might complicate and deteriorate the>MF, business performance. One way to increase efficiency and effectiveness in large alliances iso have compatible and often standardized IT systems in order to gain econornies of scale andcope as well as a guaranteed quality. The problem is that many of large IT systems used by theirger airlines does not tit the SMEs. Therefore adopting such a sytem could cause seriouslamage to the SME through the need of large investments and lack of flexibility. Another.roblem with the alliance is the lack of coordination between administration and production ofhe SME and the much bigger marketing resources of the larger airline making it almostnpossible to plan future volumes. A successful marketing campaign might give too largeolumes. On the other hand a lack of interest would lead to no or very low growth. In both caseshe planning will be difficult. A risk related to this is that the märket size due to increasedolumes becomes so attractive that the large airline might want to take över or start to compete.C'e see both opportunities and problems of an alliance between the large and small airline. Onne one hand summary they get too dependent and too restricted in their expansion. On thether hand if the alliance work well as in the case of Skyways they can grow when the larger firm; downsizing as a result of deregulation due to a tougher price competition.The smaller firms

have a different and more flexible cost situation of which the large airline couio tase auva..,.^.

This makes it possiblefor the SME to gain märket shares and grow.

In this study we have seen different kinds of agreements. In most cases SAS tried toacquire the smaller firms in other cases when their SME partner grew bigger it acquired othersmall airline.The specific route agreement and the code sharing are very common between largeand small airlines. However, the specific route agreement including handling, bookings, check msetc might be of a more general kind covering not just one route but several, which makes theagreement more extensive. In our cases the SMEs all had a need to get access to the largernetwork. In one case trying to stay neutral and getting access to several networks were tooexpensive. Some type of agreements were tools to facilitate growing and concentrating resourceson creating efficient operations. In other cases the agreement were more long term involving

part ownership and /or joint marketing.

The situation for the SME to exist and grow within the ak transportation industry isdifficult due to the interdependence between the different märkets. The SME that want to growwill most likely have to form some kind of agreements with larger airlines to gain access to thelarge International networks. The alliances give obstacles for growth but managed the right waythey might also give growth. Finally, we have seen that a risk in being close partner to largeInternational airline that the SME will be used as a weapon in the competition between two world

widc networks.

<*,/. d

IMP Conference in Asia, 2005Work in Progress

Maria Dembrower1

Susanne Hertz2

Lärs-Gunnar Mattsson3

Entrepreneurs acting in networks - small passenger airlines in Swedenafter deregulation

1 Royal Institute of Technology and Stockholm School of Economics2 Jönköping International Business School and Stockholm School of Economics, corresponding author,susanne.hertz(a>ihh.hj.se and [email protected] Stockholm School of Economics

IMP Coaference in Asia, 2005. Work in Progress

Entrepreneurs acting in networks - small passenger airlines in Swedenafter deregulation

Background

The Swedish domestic märket for regular passenger trafflc was deregulated in 1992. The

public policy purpose was not only to get price competition but also to stimulate

entrepreneurial and innovative activities related to operations and to services offered.

Before 1992. SAS and Linjeflyg (acquired by SAS in early 1992) had split up

the routes between them. Only.a few minor routes in remote areas had been left to small

operators. During the period 1992-2003 more than 30 firms have been active in regular

passenger flights of which half the number left or were acquired during the period. Still in

2003 the combined märket share of SAS and ils part-owned partner Skyways was around 75

%. In the early years after deregulation competition increased but a few years låter SAS was

again sole operator on most major routes. After a Märket Court decision in 2001 ruling that

SAS could not use its frequent flier program on domestic flights märket entries and

competiton increased. During the last couple of years, especially price competition has

increased.

The domestic development has to an important extent been influenced by SAS's

restructuring and its alliance with a fast growing SME, Skyways. A combination of entry and

exit of firms charactcrized the quite turbulent development. Small firms entered from nearby

air businesses and became members of, mostly operatioal, alliances. Some airlines have

focused on low cost/low price while others mainly focused on developing new ways of

thinking and operating on the märket.'

1 The period covered in this paper does not include the recent development of low price airlines operatingSwedish domestic routes.

Purpose of the paper

W e aim to describe and analyse entrepreneurial actions by SMEs after deregulation of the

märket. To understand such actions is important, given that the effects of deregulation on

märket structure and performance depend on them. We will use a definition of entrepreneurial

action that relätes to our markets-as-network perspective.

Some märket characteristics

Some characteristics of the passenger airline business in Sweden, that influence the

opportunities and restrictions for emergence and growth of SMEs are important to consider:

a. The customer base is small for most routes. It is difficult for an entrant to compete

with an incumbent on more than a few (6-8) existing routes

b. .Customers often need connections to other International or domestic route. The "hub

and spöke" organization of route networks favours large incumbents and operators

who cooperate in alliances, including eg code sharing,

c. Variations in customers' scheduling preferences makes it important for an operator to

offer several departures and arrivals per day on a specific route.

d. Customers generally prefer to fly with big rather than small aircraft.

e. Customers might be reluctant to use other airlines than the airline they usually fly, eg.

due to frequent flier programs.

f. Ät major, hub airports, especially Stockholm-Arlanda, the number of arrival and

departure "slöts" that can be allocated to new entrants are quite restricted.

g. For some routes in northern Sweden that need to be subsidized by the government as

part of a regional development program, traffic for a period is awarded an operator

after a bidding process.

Points a-f imply major problems for new SME entrants to successfully compete with SAS on

major routes with similar services. Entrepreneurial activities by the SMEs need to be directed

towards finding opportunities to circumvcnt these barrirers. Point g. implies a risk that

investment in an awarded route can only be used during a limited period.

Conceptual framework

We apply a markets-as-networks perspective. In this paper we will specifically emphasise

how small firms, entering a deregulated märket, adapt to and/or change the network structure

and network processes.

3

An entrepreneur is an actor (individual or organization) that seeks new and often

unconventional ways to develop the organization. Entrepreneurship is often related to aspects

of technical advancemcnts (Schumpeter 1989) but also to changes in economic and social

organization. Much research has focused on the role of personal attributes and social network

connections (Aldrich and Zimmer 1986) but also how industrial network relationships and

interdependencies influence opportunities for entrepreneurial action. Lundgren and

Nordenlöw (1992) discussed the influence of the industrial network structure in terms

tight/loose network structure. Changes in the network structure can affect the expected

amount of entrepreneurial activities. Deregulation is supposed to loosen the structure, thereby

increasing opportunities for entrepreneurial action by SMEs.

Given our network perspective, we regard entrepreneurial actions to be actions

that affect network structurcs and processes. Such entrepreneurial actions are concerned with

exchange between actors (Snehota 1990) and fits the definition of strategic action in networks

(Johanson and Mattsson 1992).

To enter the märket the SME needs to develop relationships to providers of

resources for operational and safety systems, including airports, to public and financial

institutions and to local customers (e.g. local businesses) that are potential users of the

specific route. It needs to develop relations to provide connections to other routes. For further

growth in the märket the SME needs to further develop old and new network relationships,

thus changing its network position.

Thus, the single firm relätes to individual other actors, dyadic relationships that

are more or less interdependent, all embedded in a wider network. Two airlines may e.g.

cooperate with each other in an alliance. More airlines might join thus forming a multi-firm

alliance. The wider network might contain several such alliances and competition might be

characterized as collective competition (Gomes-Casseres 1996; Hertz and Mattsson 2004).

The empirical study

This paper is based on a research project on entrepreneurship in industrial networks fmanced

by the Swedish Competition Authority, preliminarily reported in Dembrower (2005).

Empirical data has been gathered from secondary sources about all the airlines operating on

the Swedish märket between 1992 and 2003. Personal interviews were made in SAS and in 15

small firms, and with experts of the aviation märket.

Even if several studies have been made on deregulated aviation märkets, very

few have focused on the small firms. Furthermore only few studies have been done on SME

alliances. Below we have systematized from the empirical study, under eight headings, the

major types of entrepreneurial actions we have found, given our definition of this concept.

/. Creating new dtrect routes

To create a new direct route was a common way. Mostly these routes covered rather short

distances, between a middle sized Swedish town and Stockholm or one of the other larger

cities. This way the SME avoided direct competition with SAS. Since SMEs had a different

cost situation (e.g. use of small airplanes, use of employees for a variety of activities) they had

better possibility to profitably operate with a smaller märket base than the dominating airlines.

Major unexpected problems experienced, were not the costs to lease the airplanes, to set up a

maintenance system, to create a safety system or fmd the right staff but to find the right slöts

ät the larger destinations, especially Stockholm-Arlanda, and to develop demand. For new

direct routes involving smaller towns and rather short distances the potential customers

needed to change their habits, from the old alternatives travel ing by train or car.

2, Forming new regions

Forming a new region was another way to create a märket base. Some of the even put the

name of the region in their own name like West air, Värmlands flyg or Nordkalott Flyg.

Usually the regions comprised a number of smaller towns but no major city.

These regions did not have to be limited to Sweden but could involve also destinations in

other Nordic countries (like the northern parts of Sweden, Norway and Finland) or in other

Baltic states. These regions are sparsely populated but distances bridging destinations across

national boundaries in east/ west direction is often shorter than domestic flights in north-

south direction.

3. Combining air businesses

Many of the SMEs originated in other related businesses like taxi or charter flights, military

and educational applications and airfreight. Thereby they had a know how, a set up for the

technical and safety issues and an existing organisation. Sometimes, Skyways is a good

example, they combined several of these businesses to be bigger when they entered the

regular passenger business. Size helped when they recruited personnel from larger airlines. In

some cases combining related businesses, like airfreight and passenger flights were shown to

be problematic, since they had a different "logic" as regards operations and marketing.

However, being bigger in the entry period, seemed to have been positive for the survival of

the SME.

4. Using small airports

Most of the SMEs bad a specific airport as a home base, and also used small airports for many

destinations in their developing route network. These small airports were normally cheaper to

use, the service was better and good slöt times easier available. In a few cases, e.g. Flying

Enterprise, the SME took över the responsibility for operating the airport.

The main problem was Stockholm, due to lack of good slöt times (during peak

hours) and high costs. One of the firms entering directly after deregulation, Malmoe Aviation,

got the possibility to use Stockholm Bromma- (the old Stockholm airport, much smaller and

doser than Arlanda to Stockhom City) as a base. Even if this has been very positive for their

development, it has also caused problems to manage expansion.

In most cases the SMEs have had problems with Stockholm as a destination and have either

preferred to stan routes to the second and third largest cities in Sweden, or like Skyways,

allied with SAS to get access to SAS Arlanda hub.

5. Insourcing and outsourcing

The SMEs were very innovative as outsourcing and insourcing are concemed. The lack of

available resources made them outsource a löt of their activities, to other SMEs, to SAS

(mostly for Stockholm flights) or to specialist firms, depending on specific conditions.

Typical activities outsourced were booking, technical maintanence, check in, and luggage

handling. In one extreme case the SME outsourced both crew and planes. However, some

SMEs refiained from outsourcing because they wanted to be sure about quality delivered.

In cases when the SME had, or wanted to acquire, an attractive resource, it

might make it profitable by letting other airlines get access to it. E.g. an SME that had

invested in a good IT-system or a good technical support system, offered other firms, using

the same airport to get access to it. Some growing SMEs actually started education and

rraining programmes for other SMEs about operations, systems, etc.

Sometimes even SAS, for economic reasons was interested to use the resources of the SMEs.

A case in point is when SAS outsourced the operation of some routes to Skyways,

conditioned by SAS becoming part-owner of this rapidly growing SME.

Outsourcing and insourcing during the period studied developed rather

complicated connections of cooperation and competition in the network.

6. Entering and exiting new märkets

The SMEs have quite restricted financial resources and some of the märkets that they focus on

are new in terms of routes and networks served. (The estimated costs of starting an SME in

the airline industry varies between 50-100 million SEK but many have tried to start with

much less.)

Due to high costs and limited resources it is necessary for the new airline to

leam about the opportunities to develop on the märket as quickly as possible. The problem is

to be able to stay the necessary time to find out whether there is a possibility to be profitable

and to develop a loyal customer base. The SME must fly according to schedule even when

there are very few passengers, it must be persistent in it's marketing activities and be able to

accept losses for a substantial period..

SMEs have been quick and flexible in both entering and leaving specific märkets. Sometimes

they acquired the assets of bankrupt airline and took över the routes, thereby leaming by the

mistakes made by the first airline. Skyways added a number of routes left vacant by exiting

SMEs.

7.Forming alliances

Alliances and cooperative agreements were necessary to be able to offer the customers access

to a larger route network, to coordinate activities and to economize on resources. (Wang and

Evans 2001; Hertz 1996). Most of the firms needed to have an agreement ( a so called

interline agreement) with SAS in order to be able to offer the customers to continue in the

larger networks. After a decision in the Märket Court SAS was obliged to accept such

agreements. Thus this arrangement cannot be seen as an alliance in the same way as the code-

sharing arrangements that SMEs have with major operators but Sometimes also with other

SMEs.

Över time however, needing more resources and access to other märkets, SMEs

formed many new alliances. Some of them created a smaller network of firms e.g. helping

each other with extra capacity when needed or running joint frequent flicr programs. In other

cases the outsdrcing and insourcing activities developed into deeper alliances for combination

of skills. To form alliances was a necessity for growth. Skyways developed alliances with

some smaller SMEs and had in tura, a major alliance with SAS. In some case the SME could

use the fäet that there was competition between the large world wide airline alliances (Står

alliance, One World etc.) who used SME airlines as operators of small routes feeding

passengers into the major international routes. Speciflcally the fäet that Finnair and SAS

belonged to competing alliances made Finnair interested to cover some domestic routes in

Sweden and SAS to cover some such routes in Finland. These alliances soon developed into

acquisitions.We have also observed problems for an SME related to alliances. Complex

systems and rules emanating from and adapted to a larger partner migbt be expensive, too

inflexible and not well fitted to the needs and operations of the SME partner. Other problems

are lack of coordination between administrative systems and SME operations. Furthermore,

the much bigger marketing resources of the larger airline partner makes it almost impossible

for the SME to plan future volumes due to unexpected volume changes for the partner.

All the SMEs in our study had a need to get access to a wider route network.

One SME that tried to "stay neutral" and to get access to several route networks found that to

be too expensive. Also, there were differences between the SMEs' alliance agreements in

terms of resources and activities involved and time perspective.

8. Acquiring and being acquired

Several firms were fully or partially acquired by other firms and thus connected to other

resources. Important for the restructuring of the märket were four sets of changes in

ownership. First, Skyways acquired or became part owncr of a couple of other SMEs, thus

adding routes and other resources and SAS became part-owner of Skyways. Second, SAS'

Norwegian competitor Braathen's entered the Swedish domestic märket by acquiring

Transwede and Malmö Aviation competing with SAS on major domestic routes. Then SAS

acquired Braathen's, saving it from bankruptcy. Malmö Aviation, due to competition law

restrictions for acquisitions by dominant firms, was acquired by investors outside of the

airline industry and continued as independent of SAS.Third, Finnair entered Swedish domestic märket by acquiring SME Nordic Airlink

(following the alliance mentioned above) in respons to SAS purchase of SME Air Bothnia

that operated in Finland and intemationally. Fourth, there have been a number of acquisitions

involving SMEs without links to SAS or other major airlines.

We have focused on eight different types of entrepreneurial actions by SMEs in

their entry and growth processes after the deregulation. These were: creating new direct

routes, developing new regions, combining businesses, using small airports, insourcing and

outsourcing, entering and exiting märkets, formation of alliances, acquiring and being

acquired. These are to a large extent interacting, e.g creating new routes, developing new

8

regions, entering/exiting märkets, acquiring and being acquired, using small airports. Another

example is insourcing/outsourcing, forming alliances, creating new regions of interrelated

entrepreneurial actions.

Such interrelatedness is important to consider because it involves more actors in

the network, implies investments in the network and are important aspects of innovativeness.

Discussion

In the introduction we listed seven specific attributes of the passenger airline industry. We

will now see how SMEs have tried to circumvent these barriers to entry and growth. on the

märket, related to the eight types of entrepreneurial actions.

a.The customer base for most routes is too small for more than one airline. SMEs have

avoided such routes by creating new direct routes and new regions. It bas got access to such

routes by forming an alliance with, or being partly acquired by the incumbent sole operator

that tums this route över to the SME. Another way to handlc uncertain routes is to be flexible

in entering and exiting for the SME new märkets (new routes)

b.Customers oflen need connections to other international or domestic routes. This problem

has been handled by forming alliances (eg code sharing), by acquiring or being acquired and

to a small extent by forming new regions.

c.Variations in customers' scheduling preferences. When the SME has created a new direct

route it is likely that the customer base is too small for more than one or two

departures/arrivals a day. There are no competing schedules with more frequent flights, like

those that exist between major destinations where the number of departures/day might be 6-8.

Cooperation between airlines in scheduling flights may solve some of the problems.

d. Customers generalfy prefer to fly in big rather than small aircraft. Again, for the new direct

routes the customer base is too small for large airplanes regardless of the size or the märket

share of the operator. Also the use of small airports makes small planes the only altemative.

However, potential customers who strongly object to fly in small planes further diminish the

customer base.

e.Customers might be reluctant lo use other airlines than the airline they usuallyfly, New

direct routes created by the S ME have no incumbent airline competing on that route but a

passenger might perhaps have an indirect atternative with the airline they are used to. The role

of SAS frequent flier program for customer preferences radically diminished after the Märket

Court decision and some SMEs have their own program, alone or in cooperation with others.

Also acquisitions might diminish this barrier to entry as well as SME flexibility in märket

entry and exit.

/ "Slöts" that can be allocated to new entrants are restricted. Using small airports helps on

such destinations but not to get slöts ät major congested ones. For those alliance with an

incumbent or being acquired are more important.

g.Biddingprocess for some minor routes. Some of those routes might be related to new direct

routes, forming new regions the experiences of entering/exiting märkets as well as combining

air businesses.

We can thus conclude that a variety of interrelated entrepreneurial actions by the SMEs have

been directed towards finding opportunities to circumvent the barrirers. There are thus a

number of interrelated entrepreneurial actions by SMEs aimed ät getting, developing and

coordinating resources for growth on the deregulated märket, preserving flexibility due to

uncertainty and limited resources. Solutions to growth problems invariably involve

relationships to other actors or developing a network of cooperating firms, outsourcing certain

activities and insourcing others. Creating alliances, not only with other airlines but also with

local authorities and local enterprises is very important to get flexibility and the growth. Small

newly started airlines might be more unconventional than large incumbents with large route

networks. But a problem for the SMEs, being so dependent on other actors, including other

SMEs is that "their" network is more vulnerable (Tsang 1998),

A typical growth path for an SME after deregulation

SMEs that survive and grow during the period can be seen, typically, to develop in three

stages. In tfieftrst stage they find a märket niche, eg new direct routes or routes taken över

from a larger firm as part of the latter's restructuring of its route network. An SME might also

find a niche by using a small airport supported by local government and local business. The

10

niche should offer a potential for the airline to expand its activities and its network of

relationships with customers, suppliers, operating partners, etc. In the second stage the growth

of the SME makes it attractive as partner for other airlines. Alliances vary. Some are loosely

connected in the beginning while others are based on a close connection, implying restrictions

on the SME's activities and growth. Through combining different roles of customer, supplier,

competitor and partner, the SME might in this stage be able to increase its know-how and

resources in marketing and operations and to establish new useful relationships, leading to

further growth. During the third stage the SME can use a stronger stronger position in the

networkfor further growth and better outcome of negotiations with major counterparts such as

SAS.

The three stages are stages in the reconfiguration of the deregulated märket for

domestic flights in Sweden involving development of inter-organizational networks rather

than a set of individual firms. During that reconfiguration the entrepreneurial actions of the

SMEs are both vital for their own fäte and important the performance of the märket after

formål deregulation.

References

Aldrich, H. and Zimmer, C. (1986) Entrepreneurshtp through Social Networks, The Art andScience of Entrepreneurship, ed. D.L. Sexton and R.W. Smilor, 3-23. Cambridge, MA, TheBallinger Publishing Company.

Dembrower, M. (2005) Entreprenörskap i industriella nätverk-en studie av nätverksberoendeför småföretagande efter den formella avregleringen. Arbetsrapport till Konkurrensverket.

Doganis, R., (2001), The airline business in the 21st centuty, Routledge, USA.

Gomes-Casseres, B.(1996). The Alliance Revolution - The New Shape of Business Rivalry.Cambridge. Harvard University Press

Hertz, S. & L.-G. Mattsson. (2004) "Collective Competition and dynamics of MärketReconfiguration" Scandinavian Journal of Management

Hertz, S. (1996) "The Dynamics of International Strategic Alliances- A study of freighttransport companies." International Studies of Mangement and Organisation. Summer 1996Vol 6 No 2, pp. 104-130

Johanson, J & L.-G. Mattsson, (1992). "Network Positions and Strategic Action - AnAnalytical Framework". In : Axelsson B. & Easton G (Eds.) Industrial Networks - A NewView ofReality. London, Routledge

11

Lundgren, A & Nordenlöv, L.( 1994) Mellan teknik och marknad -Tekniskt entreprenörskapiindustriella nätverk In Hulthén.S & Mattsson,L-G ( eds) Företag coh Nätverk, Stockholm,Publ Nerenius och Santérus,

Nordenlöw, L., (1994), Entreprenörskap i Industriella Nätverk, EFI Research Report,Stockholm School of Economics, Stockholm.

Snehota, L, (1990), Notes on a theory qfbusiness enterprise, Doctoral Dissertation ät UppsalaUniversity l, Uppsala University, Sweden.

Schumpeter, Joseph, A., (1992), The Theory of Economic Development, TransactionPublishers, London.

Tsang, E.W.K... (1998) "Motives for Strategic Alliance: A Resource-based Perspective" InScandinmian Journal ofMangement Vol 14 No 3, pp. 207-221

Wang, Z.H. and Evans, M (2002), "Strategic Classification and Examination of theDevelopment of the current Airline Activities", Journal of Air Transportation, Vol. 7 No. 3,p. 73-101.

firm is downsizing as a result of deregulation due to a tougher price competition. The smaller

firms have a different and more flexible cost situation of which the large airline could take

advantage. This makes it possible for the SME to gain märket shares and grow.

In this study we have seen different kinds of agreements. In most cases SAS tried to acquire

the smaller firms in other cases when their SME partner grew bigger it acquired other small

airline. The specific route agreement and the code sharing are very common between large

and small airlines. However, the specific route agreement including handling, bookings, check

ins etc might be of a more general kind covering not just one route but several, which makes

the agreement more extensive. In our cases the SMEs all had a need to get access to the larger

network. In one case trying to stay neutral and getting access to several networks were too

expensive. Some type of agreements was a tool to facilitate growing and concentrating

resources on creating efficient operations. In other cases the agreement was more long term

involving part ownership and /or joint marketing.

Final comments

The situation for the SME to exist and grow within the air transportation industry is difficult

due to the interdependence between the different märkets. The SMEs have been very

inventive in finding new ways and creating flexibility that would not have been invented if the

märket were not deregulated. It has however över time also been also difficult for the

consumers since opportunities been shifting very fast as some firms entered and left the

märket and some firms went bankrupt.

The SME that want to grow will most likely have to form some kind of agreements with

larger airlines to gain access to the large International networks. The alliances give obstacles

for growth but managed the right way they might also give growth. Finally, we have seen that

a risk in being close partner to large International airline that the SME will be used as a

weapon in the competition between two worldwide networks. Further the deregulation is not

something that comes över night it takes a number of year before you can offer the newly

started firms a fair competition on the märket. The constraints take so many different forms

and shapes. This goes to begin with mainly for the small firms but över time also for the ex

monopolist.

References

Aldrich, H. and Zimmer, C. (1986) Entrepreneurship through Social Networks, The Art and

Science of Entrepreneurship, ed. D.L. Sexton and R.W. Smilor, 3-23. Cambridge, MA, The

Ballinger Publishing Company.

Doganis, R., (2001), The airline business in the 21st century, Routledge, USA.

Gomes-Casseres, B.(1996). The Alliance Revolution - The New Shape of Business Rivalry.

Cambridge. Harvard University Press

Ford, D., (2001), Managing & Marketing Technology, Thomson learning, USA.

Hertz, S. & L.-G. Mattsson. (2004) "Collective Competition and dynamics of Märket

Reconfiguration" Scandinavian Journal of Management

Hertz, S. (1996) "The Dynamics of International Strategic Alliances- A study of freight

transport companies." International Studies of Mangement and Organisation. Summer 1996

Vol 6 No 2, pp. 104-130

Johanson, J & L.-G. Mattsson, (1992). "Network Positions and Strategic Action - An

Analytical Framework". In : Axelsson B. & Easton G (Eds.) Industrial Networks - A New

View ofReality. London, Routledge

Lundgren, A & Nordenlöv, L.( 1994) Mellan teknik och marknad -Tekniskt entreprenörskapi

industriella nätverk In Hulthén.S & Mattsson,L-G ( eds) Företag coh Nätverk, Stockholm,

Publ Nerenius och Santérus,

Nordenlöw, L., (1994), Entreprenörskap i Industriella Nätverk, EFI Research Report,

Stockholm School of Economics, Stockholm.

Snehota, L, (1990), Notes on a theory of business enterprise, Doctoral Dissertation ät Uppsala

University l, Uppsala University, Sweden.

Schumpeter, Joseph, A., (1992), The Theory ofEconomic Development, Transaction

Publishers, London.

Webbplats SITES/DOKUMENTSkannat till HTTP://sps01:80/SITES/DOKUMENT/DOKDokumentnamnet = HEN0002Rubriken =Ursprung =

*SPLI~P

w

'SITES/DOKUMENT*

*DOK*

Ulllllill'HEN0002"

II

III

0002

Detta blad skall ligga sist i dokumentet!Om dokumentet är dubelsidigt skall detta vara sista sidan

Date post:	24-Mar-2020
Category:	Documents
Upload:	others
View:	1 times
Download:	0 times

46 31 Dns llyjatf - Konkurrensverket · factor, the distribution channel, also can prohibit new...

Documents