A Spinoffs Study Applied to the Airline Industry Erika Kutscher System Design and Management Program I a Spinoffs Study Applied to the Airline Industry Acknowledgments

LIBRARIES The author hereby grants to MIT permission to reproduce and to distribute publicly paper and electronic copies of this thesis document in whole or in part in any medium now known or hereafter created. Abstract Airlines have been recently debated the management of some of their non-core divisions, such as the Frequent Flyer Program (FFP).A spinoff is a form of corporate contraction that many companies have recently chosen. Through a spinoff, both the parent company and the divested subsidiary can each focus on their own activity, which translates into a better performance of both entities. This thesis studies the circumstances in which a spinoff is a good strategy to pursue, along with some important issues that must be considered when reaching agreements. Spinoffs are basically a "downsizing" of the parent firm; therefore, the smaller firm must be economically more viable by itself than as a part of its parent company. The motivation for analyzing this particular topic comes from a question of current interest: Under what circumstances is it advantageous for an airline to spin off its Frequent Flyer Program, or other divisions that are not related with the airline's operation? In this thesis, an extensive literature review introduces the reader to the different forms of corporate contraction and their performance under different circumstances. Three cases related to the airline industry follow: the spinoffs of TripAdvisor from the web agency Expedia, of Air Canada's FFP Aeroplan, and of American Airline's distribution system Sabre. These three cases illustrate some of the key issues that must be carefully considered when spinning off a subsidiary. The thesis concludes that spinoffs are a smart strategy when the focus of the spun off division is different from that of the parent company. However, to safeguard future business relationships, the two entities must negotiate detailed agreements that are robust enough to perform successfully in all foreseeable circumstances.

Acknowledgments I would like start by thanking Professor Richard de Neufville, my supervisor, for his vital guidance and mentorship throughout my thesis. His experience and knowledge were the key factor for bringing together all my research and analysis in a successful manner.
Next I would like to thanks the System Design and Management (SDM) Program Director, Pat Hale, for always being available to give me guidance. I would also like to thanks all of the SDM program staff for their great work on running such an outstanding program. I specially want to thanks Chris Bates, Bill Foley, Amal Elalam and Melissa Parrillo.
I am very grateful for the patient help of MIT's writing and communication center and of my cousin Lynda. Thanks to them I was able to write this thesis properly and also learned a new skill that I know will be of great use.
I am also grateful for the guidance provided by my former employers. They gave my motivation to develop this topic and provided very interesting ideas.
Last, but not least, I would like to thank my husband Matias for his love and support through this amazing journey. List of Tables   Table 1:  Corporate contractions such as spinoffs occur far less frequently and receive much less attention despite the fact that they unlock significantly more value for the shareholders than M&A do.

Table of Contents
In the airline industry, spinning off a Frequent Flyer Program (FFP), for example, can be very controversial. FPPs improve the profitability of the industry in two very important ways: first, they generate a massive amount of information that boosts the capacity of the airline to forecast passenger demand; and second, they reduce, to some extent, the price competition among airlines by making the passenger less price sensitive. Unlike the airline industry, the FFP business has a more benign competitive environment with low fixed costs and captive customers.
At first glance, it appears that the FFP would be a textbook candidate for a corporate contraction; however, only one airline has spun off that division completely. Can an airline avoid conflict of interests between its main operations and the FFP business? Under what circumstances is it better to separate them to maximize shareholder value? What actions are needed to increase the chances of success in a separation? These are the type of questions that many airlines and investors are asking.

Goal of the Thesis
The

The different forms of Corporate Contraction
This section describes the most popular types of corporate contraction or divestitures and summarizes them in four types: tracking stocks, spinoffs, equity carve-outs and sell-offs.
Tracking stocks, also known as letter or targeted stocks, are described by Anslinger et al. (1999) as a class of partner company stock that tracks the earnings of a division or subsidiary. Typically distributed as a dividend to shareholders in the parent company, these shares can also take the form of an initial public offering (IPO). The control of the subsidiary's operation remains in the hands of the parent company's board as in one big company. However, through this form of restructuring, investors can be aware of the profitability of parent and subsidiary separately, which can be valuable information if the two entities are in different industries. Tracking stocks can be useful as an acquisition currency.
In 1999, the Walt Disney Company issued tracking stocks for its web portal go.com right when the dot-com bubble was emerging. When go.com was able to be valuated separately, and in times when all the dot-coms were being highly valuated, Disney's shareholders value increased significantly. When the bubble ended, Disney decided to retire the tracking stock (Lankford, 2000).
Spinoffs have a wide range of definitions and no common definition has emerged yet. Some authors use the term spinoff for any type of corporate contraction, while others are more specific.
This thesis considers that spinoffs occur when an entire subsidiary of a parent firm becomes an independent business with separate shares, separate managers that run the company independently, and separate assets and liabilities. Similarly to tracking stocks, the ownership is divided among shareholders of the parent firm as a dividend, in a pro-rata basis. Because the new shares are just given and not sold, spinoffs (and tracking stocks) are tax-free transactions in the

US.
Two particular forms of spinoffs are split-offs and split-ups. The difference with respect to a traditional spinoff, as explained by Tubke (2005), is that in a split-off shareholders of the parent firm receive shares in a subsidiary in return of relinquishing their parent company shares.
Split-ups, on the other hand, occur when the parent company spins off all its subsidiaries to its shareholders and ceases to exist. The parent shares are exchanged for shares in one or more of the units that are spun off.
On October 31, 2011, ITT Corporation (a global manufacturer of highly engineered industrial products and high-tech solutions) completed the split-off of its Water business, Xylem and Defense business, Exelis. The two spinoffs were distributed in a "one-for-one" basis and the new ITT underwent a "one-for-two" reverse split. For example, if a person held 10 shares of old ITT, after the spinoff those shares were exchanged for 10 shares of Exelis, 10 of Xylem and 5 of new ITT (Sauer, 2011). Equity carve-outs are defined by Gaughan (2011) as a type of divestiture that involves the partial sale of equity interest in a subsidiary to outsiders. As in the previous cases, a new legal entity is created, but with a stockholder base that may be different for that of the parent selling company, given that part of the new firm stocks are sold to the public. Similarly to spinoffs, the divested company has a different management team and is run as a separate firm. The two main differences with spinoffs are, first, that an equity carve-out comes with a cash inflow from the buyers, and therefore it is not a tax free transaction; and second, as it is a partial sale in which the parent company generally keeps the majority of the shares, the decisions of the new management team are controlled by the parent firm's board.
In 1996, American Airlines performed an equity carve-out for 18% of its distribution system, Sabre. This new structure allowed Sabre to grow significantly, however, in order to gain more independence and provide services to other airlines, Sabre spun off completely in the year 2000 and started trading as an independent company.
Sell-offs (also known as Buy-outs) are defined by Cusatis et al. (2001) as a divesting method in which the subsidiary is sold outright to another corporation or to the sub's managers in a Leveraged Buy-Out (LBO). A sell-off is similar to an equity carve-out in the sense that it generates a cash inflow to the parent firm, but differs in that the new divested company is independent of the parent firm because no portion of it remains in the parent firm's ownership.
In 1972 Gaughan (2011) showed how the number of divestitures in the US follows the leading trend of the mergers and acquisitions curve, after a short delay of about two years. Figure 1 shows an intense peak of Mergers and Acquisitions activity in the late 1960s that was then It can be understood that after absorbing the desired division, the company would divest the rest of the acquired firm.
In the airline industry, the current tendency is for companies to merge. The fierce price war that the many competitors have generated has turned the airlines into a very unprofitable industry. Consequently, in the past decades airlines have started to merge to reduce competition, expand their networks and take advantage of costs synergy. Table 2 shows some of the recent M&As of US based and international airlines.

Results
Several studies in the academic finance literature have documented significant positive results around corporate spinoffs. Some of these studies analyze the stock price reactions at the moment of the spinoff announcement, while others analyze it in the long run. Further studies try to explain those positive reactions in the stock price by focusing on performance improvements measured by indicators such as growth in sales, operating incomes, capital expenditures or return on assets (ROA). These studies found that higher returns occur when the parent firm and its subsidiary operate in different industries, from which they conclude that value creation comes from "corporate focus", meaning that managers focus only on their own business. The following paragraphs describe the mentioned studies in further detail.
In the early 1980s, several researchers documented the positive effects that spinoff announcements have on shareholders' wealth. In their studies they propose three hypotheses that may explain this phenomenon: (1) wealth transfers from bondholders to stockholders, (2) relaxed regulatory or tax constraints, and (3) productivity increases from reducing diversity under one management, allowing both entities to use unique sets of contracts in which each has a comparative advantage. All of the researchers conclude that the third hypothesis is the strongest.
A very interesting study by Hite and Owers (1983) shows positive abnormal returns' to shareholders of 116 firms that engaged in spinoffs during the period of 1963 to 1981. By examining press announcements detailing the spinoff reasons, the authors classified their sample Positive abnormal return is a term used to describe the returns generated by a given security over a period of time that is different from the expected rate of return. The expected rate of return is the estimated return based on an asset pricing model, using a long run historical average or multiple valuation. (Source: investopedia.com) in four groups: specialization in operations, facilitation of a merger, legal/regulatory difficulties and unknown reasons. The stock price reactions in the two-day interval surrounding the first press announcement showed no substantial differences and were significantly positive for all four categories of spinoffs. But when analyzing the period from the announcement to the initial trading day of the spun off subsidiary, the four groups show substantial differences. Firms that were specializing operations showed returns of 14.5%; firms facilitating mergers showed 11.6%.
The firms in the third group chose spinoffs as a response to regulatory or potential anti-trust intervention, and unlike the previous groups, these firms experienced negative returns of -4.7%.
Finally, the group of firms with unknown spinoff reasons showed a return of 6.6%, which is approximately equivalent to the average of the first three groups. The higher returns of the first two groups are consistent with the reduced diversity efficiency hypothesis. Veld and Veld-Merkoulova (2004) performed a similar study with 156 spinoffs from companies in 15 different European countries. The authors found an abnormal cumulative average return of 2.62% over the three-day announcement window, which increase to 2.66% for the subsequently completed spinoffs. These returns are smaller than the ones found by Hite and Owers, but still in line with US performance.
The positive impact on the stock price does not occur only at the moment of the spinoff announcement or first trading day. These abnormal positive results also arise during the subsequent years in which operational improvement actually takes place. Cusatis et al. (1993) measured the stock returns of spinoffs, their parent firms, and spinoff-parent combinations for periods of up to three years following the spinoffs. In order to create a comparable base, they analyzed the spinoff and parent firm performance using raw and matched-firm-adjusted returns.
The authors found significantly positive abnormal returns for spinoffs, their parents, and the spinoff-parent combination. Table 3 provides the mean returns for the spinoffs and the parent companies for sub-periods corresponding to 6, 12, 24 and 36 months after the initial day of trading. These increasing returns suggest that spinoffs provide superior long term returns to investors of both the parent and the subsidiary. The authors also found that many of the firms in the study experienced a subsequent takeover. Both spinoffs and their parent firms experience significantly more takeovers than do control groups of similar firms and the abnormal returns in their stock price are limited to those firms involved in takeover activity. Because of this finding, the authors conclude that spinoffs provide a low-cost method of transferring control of corporate assets to acquiring firms who will create greater value.
To explain the increases in the parent and spinoff stock prices, Daley, Mehrotra, and Sivakumar (1997) examined the relationship between spinoffs and corporate focus by comparing the performance of spinoff firms when the parent company and the spun off entity were in two different Standard Industry Classifications (SIC) codes (cross-industry spinoffs) relative to instances in which both were in the same SIC code (own industry spinoffs). The study starts from the premise that dispositions involving assets outside of the core business of a firm are viewed by the market as increasing in value while dispositions of core assets are not. The shedding of these non-core assets is referred to as "increasing corporate focus." Analyzing the 85 spinoffs in the study, the authors looked for evidence to prove the premise, while also distinguishing what the source of this value creation was.
The results, looking at the three-day announcement period, showed significantly abnormal returns around the time of the announcement of cross-industry spinoffs only, which supports the initial hypothesis. The authors found substantial improvements in ROA and various other measures of performance for cross-industry spinoffs but not for own industry deals. They interpreted this difference as indicating that performance improvements provide an explanation for the value increase surrounding spinoff announcements, and that this is associated with increasing corporate focus. The authors conclude that cross-industry spinoffs create value only when they result in an increase in corporate focus. They attribute the performance improvements to companies removing unrelated businesses, allowing managers to avoid distraction of noncore entities and focus attention on the core operations they are best suited to manage.
To complement the findings of Daley et al. (1997), Desai and Jain (1999) examined the corporate focus and increased returns relationship in the long-run stock market performance.
They claim that the full impact of the managerial action is likely to be captured only through a long-run study. Their results for the three day announcement period are similar to Daley et al. findings. However, they show that the superior performance of the focus-increasing spinoffs persists in the post-spinoff period. Table 4 shows the returns for the focus-increasing spinoffs and the non-focus-increasing spinoffs over periods of three days post announcement, as well as one, two and three years following the spinoffs. The results are similar when the parent firm and the subsidiary are examined separately. Finally, the authors explore potential motivations for undertaking non-focus-increasing spinoffs. They find that the reason to do so is simply because they are spinning off poorly performing subsidiaries. They -prove this premise by examining financial leverage and interest coverage ratio to investigate whether these firms are in financial distress. They found no evidence that spinoffs are motivated by a high level of financial leverage or financial distress.
Moreover, they do not find any evidence of disproportionate transfer of debt from the parent firms to the subsidiaries through a spinoff.
Veld and Veld-Merkoulova (2009) summarize the findings of previous studies on spinoffs. They conclude that the higher abnormal returns are associated with spinoffs in which there is an improvement of industrial focus. The authors observe that larger spinoffs show higher abnormal returns, which they relate with the industrial focus result, as bigger companies are likely to present higher diversity in their subsidiaries focuses. The authors also find a surprising result that reveals that spinoffs that were not completed show higher returns at the moment of announcement than spinoffs that were completed later. They assume this happens because spinoffs that were ex-post not completed were less expected by the market participants.

Sources of value
We have seen that when a division operates in a different industry than the one of its parent firm, divesting it allows managers on each separate entity to focus only on their own industry which results in better performance for both parent and subsidiary. Separating parent firm and subsidiary helps to "unlock hidden value" by correcting the undervaluation of certain parts of a conglomerate firm. This section presents different explanations for the positive abnormal returns that are observed in all different types of corporate contraction.
Anslinger, Klepper and Subramaniam (1999) analyze spinoffs, equity carve-outs and tracking stocks. The authors propose that gains in stock prices flow from four changes. First, there is an increase in coverage by analysts. By exposing the division to the market, its operating performance becomes more transparent, which raises shareholder returns by revealing hidden value. The authors explain that this transparency does not come from a greater quantity of information provided by the company, which could happen without restructuring ownership, but from an improvement in the quality of coverage by analysts. They can dedicate better quality time to the separate companies because now each is categorized more specifically in their own industry. In other words, the increase in return comes from accepting that the conglomerate was being undervalued, also known in the academic literature as "diversification discount", and that through a spinoff this lower valuation can be corrected.
Second, the restructured subsidiaries attract new investors. Empirical evidence demonstrates that there is a small overlap between people who invested in a parent company and those who invest in its subsidiaries after a restructuring. To sustain this observation, the authors took 55 examples of all types of divestiture and compared the top 25 shareholders of the parent and the subsidiary. On'average, ownership overlapped by only 17% to 27%. A fair explanation for this small overlap in investors is that, in theory, the market values a company as the sum of its parts, analyzing the growth prospects of each of the separate businesses and using the market view of predicted cash flows to determine the price of the stock. But in practice, the market consists of many investors with their own individual criteria for making a purchase. The problem is that investors who find a particular division of a company attractive might reject the stock of the parent firm because, for example, it competes in a less attractive line of business or has slower growth prospects.
Third, as shown in section 2.3, restructuring of ownership generally improves the operating performance of both parent and subsidiary. Through issuing new equity, a company can offer managers incentives tied to the market performance of the divisions they run. This way, managers can clearly indicate to investors, executives, and other employees, that performance, ownership, risk and reward are bound together. In contrast, if the two entities operate together, conflicts of interest can lead to a sub-optimal performance of one division, for the benefit of a different area in the company. This occurrence is also known as reverse synergy, which means that the parts are worth more separately than they are within the parent company's corporate structure.
Finally, restructuring can improve corporate governance and increase strategic flexibility, which can be achieved by pushing management accountability deeper into the organization. For a subsidiary that is newly exposed to the market, greater examination by investors and analysts creates a level of pressure to which management must respond. Operating performance generally improves as a result. For management in poorly performing businesses, the new accountability becomes tangible through lower compensation when the stock weakens.
Additionally, spinoffs can increase the strategic flexibility of businesses by allowing a subsidiary to form relationships with companies that do not want competitive information to flow to its parent firm. After being spun off from AT&T, Lucent was better able to do business with international telecommunications companies that perceived its parent firm as a rival.
For the case of sell-offs, the source of value generation lies in the fact that the high bidders of the divested subsidiary are likely to get more benefit from it, while the parent firm will probably get more benefit from the cash than from the subsidiary profits. When the market responds positively to this asset reallocation, it is expressing a belief that the firm will use the cash more efficiently than it was utilizing the asset that was sold. The selling firm has several options when it is contemplating the disposition of the newly acquired cash. It may pay the cash to stockholders in the form of a dividend or it may repurchase its own shares at a premium.
Either option will give the stockholders an immediate payout. Alternatively, the firm could retain the cash and use it to pay debt or to finance internal investment.
Chemmanur and Yan (2003) present a new explanation for the performance and value improvements that follow spinoffs. They develop a theoretical model to demonstrate that spinoffs increase the probability of a takeover by a better management that will improve value.
The model consists of a firm with two divisions under one current management and a rival management team that wants to get control of one of the divisions. Giving up control can benefit shareholders, but involves a loss of control by the current management. A spinoff increases the chances of losing that control, motivating the current management to two possible choices: either work harder on running the firm or relinquish control of one of the two divisions to the rival management. Both choices lead to an increase in the combined equity value. The authors also show that spinoffs can increase value even when there is no better rival management to take over the division, just by serving to discipline firm management.

Concerns
Despite all the mentioned advantages of a corporate contraction, companies sometimes hesitate because they fear the costs and complexity that a restructuration may bring. The following paragraphs describe some of the main concerns that firms face when they analyze a divestiture.
Loss of stability: When performing a spinoff, the parent firm may doubt the ability of the subsidiary to survive independently. In the case of tracking stocks and carve-outs, some companies may fear that the subsidiary might be taken over by their rivals, or that analysts could exert pressure to spin it off completely. However, according to Anslinger et al. (1999) Redundancy and complexity of separate operations: When issuing new equity greater complexity is expected to arise. On the one hand, the board of directors will have to be responsive to more than one set of shareholders. Moreover, the creation of equities to attract different types of investors places a burden on senior management to communicate effectively and consistently with each group. On the other hand, companies have some internal divisions that provide service for the operations of both parent and subsidiary. Examples of such divisions are the website, call centers, sales offices and the R&D department. Afterwards, the parent firm and subsidiary will have to provide these services separately, losing the previous synergy.
Additionally, in the cases in which parent and subsidiary maintain business relationships after the spinoff, the subsidiary will have to maintain interaction with many of the parent's departments.
In a scenario of separate equity, the complexity of these interactions is expected to increase.
Transaction costs: According to Anslinger et al. (1999), the direct transaction costs associated with raising new capital in the market through an equity carve-out can represent 2% to 5% of the transaction's total value; for a spinoff or tracking stock, the figure is around 2%. The higher percentage for carve-outs may reflect the fact that their offerings often come in the form of an IPO and not a stock distribution. On top of these transaction costs, divestitures that will maintain a business relationship must account for the costs of developing a robust agreement.
Furthermore, a subsidiary might want to pursue business with its parent's competitors. The Sabre Group, for instance, provides reservation systems not only for American Airlines, its parent, but also for American's rivals.
This chapter has defined the different types of divestitures, followed by historical trends, performance and the main factors involved. Although there are a large number of papers covering this field of study, there is little research on the contractual relationships that the parent and spun off company need to develop when these two entities pretend to maintain tight commercial interactions. The following section presents real examples that will contribute in finding the important factors that the parent firm and the spun-off subsidiary need to take into consideration before they close the deal.

Introduction
After having defined and understood spinoffs and divestitures in general, this chapter analyses three cases of spinoffs around the airline industry. Airlines are a very difficult industry in terms of profits. Since deregulation in 1978, many airlines have born and died. The industry has faced alternate periods of gain and loss that have been amplified through time. In the last decade the picture has become even more complex, which can be seen in figure 2. Through the analysis of these three cases, a new finding is presented. When a parent firm and its divested subsidiary are going to maintain a business relationship after splitting up, what determines the success of the divestiture is the robustness of the agreements that the two entities sign, prior spinoff. Among the three cases presented below, the first one requires no future relationship and the outcomes predicted by the academic literature indeed occur. The second and third cases consider a very important business relationship after the spinoff. In one of them, the previous agreements were very robust, while in the other, many flaws can be found.
The results, as expected, are very different.

Expedia spins off TripAdvisor
The spinoff of TripAdvisor from its parent Expedia is a very clean example that illustrates many of the concepts presented in Chapter 2. Stephen Kaufer, CEO and co-founder of the subsidiary, explained in a public interview that "the evolution of TripAdvisor into an independent, public company is about unlocking shareholder value." Expedia is one of the largest online travel agency in the world. The company allows business and leisure travelers to book flights, cruisers, hotels, cars and many other travel related services. At the same time, Expedia allows travel suppliers to post their inventory and fares in a much known website that customers visit very frequently. On the one hand, the website benefits a travel supplier by driving in more demand. But on the other hand, the supplier is hurt because it must drop prices for customers to choose it over other competitors. This sort of price war is the biggest benefit for the consumers and the reason why they visit Expedia in the first place.
Consequently, the website provides a vast opportunity for advertising, as it reaches a very valuable audience.
TripAdvisor is also a travel website, but unlike Expedia, it does not sell anything to travelers. TripAdvisor assists customers for free in gathering travel information, posting reviews and opinions of travel-related content and engaging in interactive travel forums. TripAdvisor is one of the earliest adopter of user-generated content. Users are the ones that provide most of the website content, and they also do it for free. In order to generate profit, TripAdvisor is supported by an advertising business model, boosted by the popularity that the website has earned.

History and Business Plan
This section relates a brief history of Expedia and TripAdvisor along with their business strategies, as it is presented in the two companies' websites.

Financial Results
The spinoff of TripAdvisor from Expedia has shown results that are in line with the predictions of the academic literature. Figure 4 shows

Air Canada spins off its FFP, Aeroplan
Air Canada, founded in 1936, is Canada's largest full-service airline. It serves over 170 destinations in five continents and is a founding member of Star Alliance, one of the world's most extensive air transportation networks.
A Frequent Flyer Program (FFP) is a type of loyalty program that rewards passengers for their ticket purchases with miles or points to gain their fidelity. Clients enrolled in the program can exchange their points for more flight tickets, use them to obtain an upgrade from coach to business class, or exchange them for goods or services of other companies registered in the program. The first airline in the world that came up with the idea of a fidelity program was Texas International Airline in 1971. Their idea was to give special fares to frequent customers, but they never managed to implement it. In 1981 American Airlines launched AAdvantage, the first FFP with the format we know today. It was quickly followed that same year by United Airlines with MileagePlus and Delta Airlines with SkyMiles. (Rowell, 2010). Nowadays, most airlines offer a FFP to their clients.
Airlines also benefit by having a FFP. First, clients stay loyal because in order to keep earning points they are more inclined to choose this airline over its competition. And second, the enrollment of the clients facilitates the airline's data analysis when it comes to customer behavior.
An FFP spinoff meets the condition of corporate focus. An airline's main activity is to provide transportation for passengers and freight, in other words, move them from point A to point B as fast, safely and profitably as possible. The FFP's activity, on the other hand, is to capture clients' loyalty while making good use of distressed inventory 3 . The Airline's source of profit comes from the margin of selling air tickets, freight space and other ancillary revenues.
The FFP on the other hand has a different set of three sources of profit. The first one is the difference between the sales price of miles and the cost of the goods and services that are given as awards. A second source of profit comes from the percentage of miles that are never used.
Aeroplan estimates this portion as 17%. Finally, the third source comes from the fact that redemption of miles occurs, on average, 30 months after accrual. The FFP invests its sales money during these 30 months before spending it on rewards, and receives interest.
Because of the different nature of Air Canada's and Aeroplan's value generation, it makes sense to have these two entities work independently and focus on their own purposes.
This second case relates the spinoff of Aeroplan, Air Canada's FFP, as described by Aeroplan itself in its Annual Information Form (AIF), an annex of the company's annual report.
An examination of Aeropan's business model follows, accompanied by the financial performance that the two entities have shown since the spinoff.
3 Distressed inventory is the inventory whose potential to be sold at a normal cost has passed or will soon pass.

Business Model
Aimia is a global leader in loyalty management, whose principal business activities fall into three categories: i) coalition loyalty programs, ii) loyalty marketing services and iii) other related services, such as data analytics. This thesis focuses in the coalition loyalty programs category, because that is the one that runs the FFP Aeroplan.
As a coalition loyalty owner and operator, Aimia is responsible for establishing relationships with Commercial Partners, issuing the applicable points or miles. Aimia also has responsibility for the programs in terms of funding any required reserve, owning the redemption liability and managing and earning breakage. In general terms, the coalition loyalty business is based on two major streams of activity: First, the sale of points and related marketing services to accumulation partners; and second, the delivery of rewards to members through the purchase of rewards or shopping discounts from its redemption partners. Figure 5 presents a functional diagram of these two activities.
Members use services and "accumulate" points  Based upon past experience, management anticipates that a number of points issued will never be redeemed by members. This is known as Breakage. By its nature, Breakage is subject to estimates and judgment, and is recognized as revenue prorated over the estimated average life of a point. For the Aeroplan Program that average life is 30 months and it represents the average period elapsed between the sale of a mile and its redemption for rewards. For the Nectar Program the estimated life of a point is 15 months.
On an ongoing basis, the total estimated future redemption cost for outstanding points is determined by Aimia as the product of first, the total outstanding number of unredeemed points on a specific measurement date net of estimated Breakage, and second, the average unit cost per points redeemed in the period. Given that the future unit cost per point redeemed may fluctuate, the Future Redemption Costs liability is periodically revalued using the actual average unit cost per point redeemed, incurred in the most recent period. Finally, this future redemption cost is paired with a present cash asset that Aimia can invest during the 30 month period before purchasing rewards.

Agreements
As two independent companies, Air Canada and Group Aeroplan engaged on a long-term strategic relationship that includes several agreements. One of these agreements states that Aeroplan is required to purchase annually a minimum number of reward travel seats on Air Canada Flights, representing 85% of the average number of seats utilized in the three preceding calendar years. Based on the three years ending on December 31, 2010, Aeroplan is required to purchase reward travel seats amounting to approximately $417.8 million each year. In a like manner, Air Canada is required to purchase, on an annual basis, a pre-established number of Aeroplan miles at a specified rate. The annual commitment is based on 85% of the average total Aeroplan Miles actually issued in respect of Air Canada flights or affiliate products and services in the three preceding calendar years. The estimated minimum requirement for 2011 is $215.3 million. Through these two agreements and many others that are part of the long-term strategic relationship, the two entities protect themselves from abrupt changes in the structure of the other entity.
The long-term strategic relationship also includes agreements to prevent competition. Air Canada is not permitted to create or participate in any other FFP or customer loyalty recognition program and Aeroplan cannot provide services to any other transportation business that competes with Air Canada. However, these agreements exclude some Star Alliance member airlines and their respective FFPs. In 2009, Aeroplan added TAM Airlines and two new Star Alliance partners, Brussels Airlines and Continental Airlines, to its roster of travel partners. In 2010, Aeroplan further announced the addition of Aegean Airlines to the Star Alliance group, bringing the total number of airline partners to 33. As all of these partners primarily operate in different markets, competition with Air Canada is prevented.

Stakeholders Analysis
The following analysis shows the interactions among the main stakeholders in the loyalty generating system. These main stakeholders are the FFP, the airline, the partners (all represented as one standard partner), and the clients (all represented as one standard client). Figure 6 shows these four stakeholders and their interactions.
In Figure 6, arrows with the same pattern represent interactions that happen at the same time. The award transaction among client, FFP and airline is shown by the arrows in the left of Figure 6. In this transaction the airline awards a client with miles when he makes a ticket purchase, in order to get his loyalty. Depending on the flight length and the ticket type (full or discounted fare) the client receives a different number of miles. The airline has to purchase these miles from the FFP at a previously agreed price. That price must satisfy two conditions: First, the price cannot be higher than the loyalty valorization that the airline perceives; otherwise, the airline has no incentive to purchase the miles. Second, the price must be high enough to cover for the FFP expenses in reward tickets; otherwise the FFP has no incentives to be in the business as their balance would always be negative. Using these two boundaries the two entities will have to agree on a price that prevents them from transferring value from one to the other.

Figure 6: Stakeholders interactions Architecture diagram
A second award transaction, now among the client, the FFP and a partner, is represented by the right side arrows in Figure 6. Here the partner is the one that wants to award miles to one of its clients for a purchase that this client makes to the partner. The price that the FFP charges the partner for these miles must also satisfy the two conditions described previously. However, the price charged to the partner does not need to be the same price charged to the airline. Given that the airline has more leverage in the equation, it will be charged less. Moreover, the price charged to different partners does not need to be the same either. Depending on the negotiations in every contract, each partner could be charged differently.
The reward transaction is represented by the top and center arrows. This transaction consists of rewarding a client with a flight ticket in exchange for some of his miles. The client decides he wants to redeem his miles, so he chooses a flight. The airline then calculates the cost of that flight ticket, which is not necessarily the same as the ticket price. The airline then communicates this cost to the FFP as a transfer price. Based on the FFP strategy and that transfer price, the FFP calculates a price in miles to be charged to the client. If the client agrees with the miles price, all the reward transactions occur. If the client does not like the price or his miles are insufficient, then the transactions do not occur, but the information of the inquiry is saved in a database for future studies. All of these cost calculations must be automatized so that the client receives an instant response on his inquiry.
Finally, there could be a final transaction, not represented in Figure 6, to show a reward from a partner to a client through the FFP. The client gives miles away in exchange for a product or service provided by the partner, not the airline.
To analyze shareholders value before and after these companies separated, this section first recapitulates the ownership events that have occurred in the past decade:  value by 7% in the first day and 12% in the first week and a total of 24% in the first month, with which ACE reached its highest market value in its history. During the subsequent months the stock price dropped back, and by the day of the IPO its value was the same as three month earlier when the divestiture had been announced.
Right after the IPO, the market value of Aeroplan stayed relatively steady around $2,500 million of Canadian dollars for about nine months. However, in the subsequent months, Aeroplan's value starts increasing, reaching a peak of almost twice its initial amount in October of 2007, after two years of separate operations. ACE did not show positive results initially either.
Its share price started increasing in October 2006 when they announced the spinoff of Air Canada.
Once it was divested, Air Canada started behaving very similarly to the world airline industry, which can be observed by comparing its value with the SP500 index in Figure 7. In fact, after dropping severely in 2008, both companies have been following the airline industry trend. This suggests that even though Aeroplan runs a different business, it is highly tied to the airline industry behavior.
Discounting the effect of the general airline industry's oscillating behavior, Air Canada and Aeroplan have shown positive results. The fact that they are able to work well in an independent structure, despite the dependencies they have to each other, is really remarkable. Air Canada is Aeroplan's main customer; without the airline's partnership the program loses most of its value. At the same time, Air Canada needs to have an FFP in order to maintain competitiveness in the market. Almost every airline offers a rewards program nowadays; not offering one could cost the airline many of its customers. The companies have very important leverages on each other and yet they manage to work well. My belief is that these good results lie in the fact that this two companies took the time to write very strong and fair agreements before they split up in two separate companies. This idea will be discussed again after presenting a final case, in which the relationship post-spinoff has not been as smooth as the one between Air Canada and Aeroplan.

American Airlines spins off its distribution system, Sabre
American Aeroplan. This section relates the history and business plan as presented by AA and Sabre on their websites, followed by some recent events collected from the news that reflect some of the problems that the separate entities have faced.

History
In the 1950s, a high growth in AA's passenger demand made it very hard for the airline to handle flight reservations. Flight bookings were managed through a manual system, run by eight operators who would sort cards for every flight in a rotating file. To book a seat an operator would place a mark on the side of a card and visualize whether the flight was full. The system had limited room to scale, as eight operators maximum could fit around the file. To solve the problem, in 1952, AA introduced the "Magnetronic Reservisor", an electro-mechanic computer that replaced the card files. This new system allowed for a large number of operators to look up information simultaneously and tell ticket agents over the phone whether a seat was available.
However, a staff member was still needed at each end of the phone line, so handling tickets still took a considerable amount of effort.
In 1953, the president of AA, Cyrus Rowlett Smith, was traveling on an AA flight to New York when he met R. Blair Smith, a senior sales representative for IBM. They started talking about an idea for a data processing system that could create and manage airline seat reservations and instantly make that data available electronically to any agent at any location. Six years later, the ideas became a reality: AA and IBM jointly announced their plans to develop a Semi-Automated Business Research Environment, better known as SABRE, the first real-time business application for airline seat bookings.
Between 1960 and 1962, AA installed the first Sabre system on two IBM 7090 computers, located in a specially designed computer center in Briarcliff Manor, NY. The stateof-the-art mainframe system processed 84,000 telephone calls per day. The initial research, development and installation investment in this system took 400 man-years of effort at a development cost of almost US$40 million. In 1964 the network was completed, becoming the largest private, real-time data processing system, second only to the U.S. government's system.
The system became an integral part of AA, saving the airline 30% of its investments in distribution staff.
In 1972, Sabre moved to a new consolidated computer center in Tulsa, Oklahoma, that was designed to house all of AA's data processing facilities. In 1976, the system was installed in a travel agency for the first time, triggering the wave of travel automation. By the end of the year, 130 locations had received the system. Technology was improving very fast, boosted in part by the feedback of its main users: the travel agencies. Thanks to these improvements, by 1978, the Sabre system could store over one million fares.
During the 1980's, the technology of the system continued to improve. In 1984, a lowfare search capability called "Bargain Finder SM" was introduced. This tool, completely new in the industry, could automatically advise which class of service was the least expensive for the flights booked. In 1985, a new tool called "easySabre" was introduced. This tool allowed consumers using personal computers to tap into the Sabre system via online services, and access airline, hotel and car rental reservations. In 1986, Sabre released the industry's first revenue management system, helping maximize airline income by optimizing the fare at which each seat was sold. Along with its revenue management system, Sabre invented the virtual and continuous nesting concepts, which improved the efficiency of the system and are still used today. Because of its rapid growth and genius inventions, Sabre became a division of AMR, the parent company of AA, in 1986.
In 1992, Sabre introduced the "AirFlite" flight scheduling system, enabling the development of flight schedules that meet the airline's customer preferences. With such many technological improvements, Sabre's system was able expand its market by extending coverage to the United Kingdom and providing software, consulting and systems management services to other airlines in areas such as revenue accounting, yield management and crew scheduling.
In 1996 AMR performed an equity carve-out of the Sabre Group through an IPO of 18% of its equity. That same year, Sabre launched "Travelocity.com", the first online travel agency that allowed consumers themselves not only to access Sabre's fare and schedule information, but also to reserve, book, and purchase tickets without the help of a travel agent or broker. In 1998, Sabre formed a joint venture with ABACUS International to establish the "SabreSonic" passenger solution as the GDS market leader in Asia. In 1999, Sabre introduced "Best Fare  platforms. Numerous smaller companies have also formed distribution systems in niche markets not served by the four largest networks. These companies include the Low Cost Carrier segment and some small and medium size domestic and regional airlines.

Business Model
Sabre is organized in four business units: These solutions are used by more than 12,000 hotel properties around the world, generating more than US$12 billion in revenue each year for its customers.
" Travelocity, an online travel company, provides consumer-direct travel services for leisure and business travelers. It markets and distributes travel-related products and services directly to individuals through Travelocity and its various brand websites and contact centers, and websites owned by its supplier and distribution partners.
A traditional airline manages the distribution of its inventory in three main stages. I like to call them the Factory, where the inventory is created, the Warehouse, where the inventory is stored, and the Shop window, where the inventory is displayed for clients to see and purchase. The Sabre Airline Solutions business unit is a tool that can help airlines manage their Warehouse and/or Factory. In the Factory, airlines first decide where to fly, daily frequencies, schedules and size of aircraft. With all of these inputs they create an itinerary. Then, airlines use a Revenue Management system to define different fares with specific regulations for each destination and to decide when to make each fare available in order to maximize revenues. Sabre Airline Solutions, as well as other vendors, sells software to optimize the itinerary and the fares.
In the Warehouse, the airline stores the optimized result of the previous stage and accounts for the seats that have been already sold or seats that get canceled. For these purposes, the warehouse must be in constant communication with both the factory and the shop window.
Sabre Airline Solutions also sells software to accomplish this task.
In the Shop window, all seats and their fares stored in the warehouse are presented to the customers. These customers can access them through different channels. The direct channels are through the airline website, call center or ticketing offices. The problem with these channels is that customers go to them only if they know about the airline and remember about it when they are looking to purchase a ticket. The indirect channels are travel agencies and web based intermediaries, also call web agencies. Examples of web agencies are Expedia, Travelocity and Orbitz. These agencies and web agencies access the inventory of many different airlines in parallel through one or several GDSs, one of which is Sabre Travel Network.
Customers generally prefer to search for flight tickets through indirect channels because thanks to the GDS that runs in the back, they can see many airlines at the same time and compare them in terms of price, schedule and service. Clients can also find alternatives that come from connecting two flights from different airlines, which is something that those airlines may have not been able to offer through their direct channels.
Because of this preference of clients for indirect channels over direct channels, GDSs have won a tremendous amount of power over airlines. GDSs charge an airline an average of $

Agreements
When American Airlines and Sabre separated, they were not as careful as Air Canada and Aeroplan in building their agreements. Although the beginning of the relationship was smooth, after a decade their disputes have gone so bad that and AA ended up suing Sabre.
In their initial agreement, AA was committed to pay a fixed fee for each AA booking made through Sabre. Because of the high flight ticket fares charged in the early 1990's, this fixed fee represented a very small percentage of the price of an average ticket. However, later on, when competition increased and fares started dropping, the distribution fee became a more significant percentage of the price. The airline had to reduce all of its costs in order to be able to charge less, but the distribution cost could not be reduced.
A second problem was that after the initial agreement expired, Sabre started to increase the fixed fee, not based on anything else than their negotiation power. Sabre was one among only Unfortunately, the internal efforts were not enough; during 2011 revenues did not cover the costs and AA went bankrupt by the end of 2011 (Bloomberg).
One of AA's biggest costs comes from the fees it has to pay to Sabre and other GDS for the bookings that are sold through them. In the latest years, the GDSs have been increasing its fees. AA tried to fight this increasing cost during 2011; some of the actions that the airline took during that year are related next.

Recent events
On January 5th 2011, American Airlines filed a complaint against Sabre, for biasing AA availability and shopping displays in the Sabre GDS. The airline qualified this action as "wrongful and unjustified retaliation that threatens to seriously disrupt American's business and harm members of the public and the travel industry." They said that this bias was making it very difficult for travel agents to find and book AA's flights in Sabre, and that it is a violation of the American-Sabre full-content agreement. Sabre is AA's largest non-direct booking source; in 2010, $7 billion of its sales were booked through Sabre. The GDS was looking to more than double the booking fees it charges to American, a move that would increase the airline's distribution costs by $157 million annually. When the airline declined, Sabre decided to bias AA's availability. Five days later, AA obtained a temporary restraining order barring Sabre from biasing its flights in Sabre and AA flights and fares were back in the Sabre system in the manner they appeared before Sabre took the action January 5th. 5 A few months later, AA filed another lawsuit against Sabre, Travelport and Orbitz Worldwide alleging that the three GDS companies were engaging in anticompetitive behavior to exclude AA from entering and competing in the GDS market. Later in the year, AA and Sabre agreed to extend their current content agreement until after American's antitrust claim against Sabre settled. The agreement gave travel agencies and corporate customers continued access to AA's content via Sabre until 14 days after the antitrust claim against Sabre in Texas state court had been resolved (which has not happened as of June 2012). A similar extended agreement was signed with Travelport, meaning that AA was also continuing to appear on the Worldspan and Galileo GDSs. 6 In April 2011, US Airways also took action by filing an antitrust action against Sabre.
The airline alleged that the GDS had engaged in anticompetitive behavior in order to secure monopoly power in the GDS market. US Airways was concerned because 35% of its revenue came from tickets booked through Sabre and its affiliated travel agents, giving the GDS "disproportionate" market control, according to US Airways. The airline claimed that Sabre was forcing travel agents to rely on a single GDS to purchase tickets by the imposition of penalties to agents using a different system. US Airways' complaint alleges that Sabre has "engaged in a pattern of exclusionary conduct to shut out competition, protect its monopoly pricing power and h tt p://www.i nooz.coim/2011 1/02/0 I /news/american-airlines-sues-t ravel port-and-placates-travel-agencies-onsurcharge/ maintain its technologically-obsolete business model." Sabre filed a motion to dismiss these accusations in July 2011, arguing that US Airways was merely attempting to renegotiate its contract with the GDS and that the monopolization complaint was based on an incorrect market definition. 7 As of June 2012, the United States Justice Department (DOJ) has initiated an investigation of possible anticompetitive behavior by the GDS companies. No resolutions have come to light yet. 7 h ttp://www.tnooz.com/20 I I/ /1 0/news/american -airl in es-sues-sabre-over-d ispi ay-bi as-and-booki n g-fee-increase/

Chapter 4: Conclusion
The airline industry presents cases of both successful and unsuccessful spinoff. This thesis has presented three different cases. Expedia and TripAdvisor, show a recent spinoff that has done well so far, mainly because the two separate entities don't need to maintain a complicate business relationship. Air Canada and Aeroplan have also been successful, although they need each other greatly, because they negotiated a very robust agreement before performing the spinoff. Finally, the relationship between American Airline and Sabre has been a real challenge, as they established poor agreements prior the spinoff, that did not took into consideration how important and powerful GDS's were going to become. This thesis has reviewed the existing academic literature on value creation through corporate contraction. The four different forms of corporate contraction are tracking stocks, spinoffs, equity crave-out and sell-offs. In all of these forms, new stocks are issued for the divested subsidiary. The four versions differ from each other in the extent of independence that the subsidiary gains after the separation and whether the transaction brings or not a cash inflow for the parent firm.
Studies show that, in general, the four forms of deconglomeration have positive returns for shareholders, especially if the parent firm and the subsidiary operate in different industries.
Investors seem to have a preference for firms that operate in one single industry rather than conglomerate firms. Their lower valuation for conglomerate firms is known as "diversification discount". Gaughan (2011) explains that performing a spinoff is partly fueled by investors' pressure to release internal values that are unrealized in the company's stock price. Splitting the companies increases shareholders value by attracting new investors, increasing coverage by analysts, improving operating performance and increasing strategic flexibility.
To complement these findings, this thesis focuses on what happens after the spinoff of cross-industry companies, when parent firm and subsidiary have to maintain business relationships. When the companies follow different courses and do not need each other much, like Expedia and TripAdvisor, then a cross industry spinoff is very likely to be successful. But, if the companies need each other in the future, like Air Canada and Aeroplan or AA and Sabre, then success will depend on the level of robustness of the previous agreements. Before performing a spinoff, both parent firm and subsidiary must define the agreements that will rule their future relationship. These agreements must consider as many potential future scenarios as possible, in order to both protect and give flexibility to the parent firm and the subsidiary.
Air Canada and Aeroplan made a very strong agreement in which the price of the goods that they sell to each other (flight tickets and miles or points) is calculated automatically based on general performance indicators. The availability of seats that the airline offers the FFP is also part of the agreement and is defined based on a proportion of seats in each route and each month.
Additionally, each firm is required to purchase a minimum number of goods of the other firm per year, representing 85% of the average number of goods purchased in the three preceding calendar years. Through these agreements the two entities protect themselves from abrupt changes in the structure of the other entity. Finally, the companies prevent competition by forbidding Air Canada to create or participate in any other FFP and forbidding Aeroplan to provide services to any other transportation business that competes with Air Canada (excluding some Star Alliance member airlines and their respective FFPs).
American Airlines and Sabre did not come up with a very strong agreement in advance.
In terms of competitiveness, it is understandable that the airline cannot demand exclusivity because the beauty of GDSs is that they are global and show the inventory of many airlines at one time. Consequently, Sabre should not try to force airlines or agencies to publish only through Sabre's GDS and if it does, it should definitely not trammel their ability to switch GDSs over time. When it does, Sabre is being anticompetitive.
In terms of fees, Sabre and AA could have better agreed in a variable fee as a percentage of the ticket price, instead of the fixed fee per booking. Airlines have increased demand by lowering their prices. This action benefits both the airline and the GDS, but only the airline is making the efforts for it to happen. Through the higher volumes of passengers the airline is capable of scaling most of its costs, but not the distribution costs because they remain fixed.
If one takes Sabre's side, it can be seen as unfair that now airlines skip the GDS in their direct sales. Initially, all sales went through the GDS, including the ones made in the airline's office. The reservation system (warehouse) and the GDS (shop window) used to be tied together, but now, airlines have built their own reservation systems in order to skip the GDS in their direct sales. If AA had initially agreed to sell a specific percentage of its bookings through Sabre only, this problem could have been avoided.
Finally, there is an issue around the priority that the GDS uses to display the different options in a flight ticket search. Sabre used to do it alphabetically so that AA would show up first. Now it displays the shorter flights first, and agencies are biased by that. Web agencies have created software to read the entire GDS response and display it in a new order based on price.
These software platforms have had to overcome the problems of the obsolete technology that GDSs still use. The lack of technology improvement is in fact another reason of why airlines opted to skip the GDS in their direct web sales.
Strong agreements are a key factor of a spinoff's success when the parent firm and the subsidiary plan to maintain a business relationship. A good practice is to define variable transaction prices that depend on performance indicators of both companies rather than the negotiation power that each may have at a given time. Those variable transaction prices should be evaluated in various scenarios, using modeling tools such as the Montecarlo simulation, to ensure that they will generate a fair outcome.
Spinoff and other forms of corporate contraction have proven to be an effective way to unlock shareholders value by showing investors separate pictures of the division's performance. This is especially the case when the two divisions operate in different industries. Unfortunately, the initial positive returns can disappear or even become losses if the two divisions plan to maintain business relationships but are not able to come with strong agreements in advance. This thesis recommends that airlines and all types of companies dedicate a fair amount of time and effort to build strong agreements prior to the spinoff that are clear, transparent and fair to both