The Effect of Investments and Divestments
on the Stock Price
- An Event Study in the Football Industry -
Yu-Cheng Yang
10339892
University of Amsterdam
MSc Finance, Corporate Finance
Supervisor: Dr. R. Perez Ribas
1
stof July 2018
2
Statement of Originality
This document is written by Student, Yu-Cheng (Tony) Yang, who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
3
Acknowledgements
I would like to thank my supervisor, dr. R. Perez Ribas, for his constructive comments and for his advice on the data collection to make this study possible. Furthermore, I would like to thank mr. Y.S. Alaoui of KPMG Sports Advisory and dr. J.C.M. van Ophem of University of Amsterdam for their helpful comments.
4
Abstract
This study sheds light on the relationship between investments and stock prices by investigating whether an investment is considered to be value-enhancing for shareholders. In addition, this study tries to explain whether types of investments and strategies of firms help explain stock movements by using an event study and panel data OLS regressions. A sample of fourteen European football clubs containing over 3,000 transfers has been constructed to test this. No evidence is found that an investment is value-enhancing for investors. The results, however, do point towards an investment being made on behalf of the shareholders. No statistical evidence is found on transfer strategies and type of investments helping explain the stock movements of football clubs. The fact that no significant price reaction was identified may indicate that investors anticipated such moves and stock prices have already reflected this information.
5
Table of Content
Introduction 6
1. Literature Review 8
1.1 The Effect of Investments and Divestments 8
1.2 Types of Investments 10
1.3 Strategies of Football Clubs 11
1.4 The Football Industry 12
1.4.1 The Transfer Market 13
1.4.2 The Bosman Verdict 14
2. Data 15
2.1 Data Sources and Sample Formation 15
2.2 Descriptive Statistics 17 3. Methodology 21 3.1 Event Study 21 3.2 Cluster Analyses 22 3.3 OLS Regressions 23 4. Results 26
4.1 The Effect of an Acquisition on the Stock Prices 26
4.2 Stock Market Reaction on Transfer- and Player Characteristics 27
4.2.1 Transfer Characteristics 30
4.2.2 Player Characteristics 32
4.3 Cluster Analyses 33
4.3.1 Strategy Based on Net Balance 33
4.3.2 Strategy Based on Transfer Attitude Indicators 34
5. Discussion 37
6. Robustness Checks 38
7. Conclusion and Limitations 39
7.1 Conclusion 39
7.2 Limitations and Future Research Suggestions 40
8. References 42
6
Introduction
In an efficient capital market, management constantly seeks to maximize its firm value. A manner how they can achieve this, is by making the right investment and by selling its assets that are no longer resourceful. Selling its unprofitable assets will lead to more focus on the core business by the management. Therefore, they are able to stay competitive in their respective industry and maintain an advantage over their competitors (John and Ofek, 1995; Fotaki, Markellos and Mania, 2009). Previous literatures, however, describe the agency theory problem that occurs when a separation exists between shareholders and management. These problems arise due to unaligned objectives or different aversion of levels to risk and therefore an asymmetric effect of acquisitions and divestments exists on the shareholders’ wealth (Mulherin and Boone, 2000).
The aim of this study is to test how the stock market reacts when an announcement of an investment or divestment is made in the football industry. The football industry offers a unique setting to test the wealth effect of an investment or divestment as transfers of football players are widely followed and publicly announced by football clubs. Furthermore, additional information about the type of the investment and strategies that football clubs follow are also accessible to the general public.
Similar to Fotaki et al. (2009), this study makes an analogy between football players and investments. One can consider football players analogous to investments such as patents. A holder of a patent has the right to make use of a specific device or exclude anyone else from the production for a stated number of years. Patents can be sold to or bought from another company (Griliches, 1998). Football clubs that own a football player have the exclusive right to play him in matches and use him as a marketing instrument to generate revenues as long as his contract is valid.
Using a large hand-collected sample of over 3,000 football transfers that took place between 1998 and 2017 and the unique setting of the football industry, this study examine whether an asymmetric effect exists on the announcement of an investment or divestment. More specifically, it is tested whether an investment follows the shareholders’ value maximization theory or follow the non-synergetic theories suggested by Mulherin and Boone (2000) and Fotaki et al. (2009). It is expected that an investment will follow the shareholders’ value maximization as firms make investment decisions with the expectation that the acquired corporate assets are capable of generating future earnings.
7
reaction to the announcement of an acquisition or sale of a football player by public listed football clubs. For this study, the sale and acquisition of a football player are regarded as corporate events. The impact of a corporate announcement on the stock prices of companies can be measured by the change that the stock prices experience around the time the announcement is made public.
Limited literature exists where the effect of an investment is analyzed in the football industry. Therefore, literature with respect to strategic investment decisions are reviewed to make a valid comparison. In general, a positive and significant effect have been found on the stock returns when an announcement is made regarding an investment (Jarell, Lehn and Marr, 1985; McConell and Nantell, 1985; Burton, Lonie and Power, 1999). Concerning divestments, Jain (1985) finds positive abnormal returns around the announcement date. These results suggest that managers make investment decisions with the view of maximizing shareholders’ value. In contrary, Fotaki et al. (2009) find a negative effect of investment on stock prices around the announcement date, supporting the non-synergetic theories. The non-synergetic theory explains the decision that a sale or acquisition of corporate assets stems for reasons other than creation of shareholders’ wealth. It rather stems from managerial motives such as empire building, entrenchment and hubris.
The results of this study show that an announcement of an investment experiences positive, though insignificant, effect on stock prices around the announcement date. The positive effect is an indication that investments made in the football industry follow the shareholders’ value maximization theory. No conclusions can, however, be drawn that acquisitions of players influence the stock movements around the announcement date.
Next, by taking advantage of the unique dataset, additional tests are implemented on how the stock market reacts to specific types of investments and strategies implemented by football clubs. Some football clubs follow specific strategies that makes it possible to make an analysis on how the market react to certain strategy type. With the use of multivariate analyses, it is possible to cluster football clubs into groups that follow similar strategies. The cluster group is then interacted with the main interest variable, arrival, to analyze the effect of an arrival between different strategies.
This study contribute to the existing literature in some way. To the best of my knowledge, there have been no studies investigating the wealth effect of shareholders to a type of investment in the football industry. The wealth effect of shareholders to type of strategies that firms implement
8
is also lacking in the existing literature. The study of Şener and Karapolatgil (2015) does examine
the differences between strategies implemented by football clubs, however, what effect a strategy has on the wealth of shareholders is absent in their analysis. The results show that strategies implemented by football clubs are not significant and does not explain the stock movements around the announcement date.
The remainder of this study is structured as follows. The following section, Section 1, reviews previous literature regarding the announcement of investments and divestments by publicly listed firms and briefly discuss some main concepts. In Section 2, the data sources and sample formation are described. The methodology is discussed in section 3. Section 4 presents the empirical results. Section 5 tests the robustness of the results obtained. And lastly, a discussion of the results is presented and the conclusion for this study is drawn in section 6.
1. Literature Review
This section starts off with the review of previous literature discussing the effect of an investment and divestment on the shareholders’ wealth. Thereafter, literature regarding type of investments and strategy of firms are discussed. In the end, a short description on how the football industry operates, is given to have a full understanding on the unique setting of the football industry.
1.1 The Effect of Investments and Divestments
This section follows on from the comparison made in the previous section in which football players can be considered as corporate investments and thus as an asset of a football club (e.g. patents, capital expenditures) in order to relate to previous studies. There are a number of empirical studies that have analyzed the relationship between strategic investment announcements and stock returns.
Jarell et al. (1985) have studied the stock price reaction to research and development (R&D) announcements. They argue that the stock market only value short-term earnings and not expected future earnings and therefore they expect the stock market to react negatively to R&D announcements and discipline companies that invest in long-term projects. To test their proposition, they use a sample of 62 R&D announcements published in the Wall Street Journal during the period 1973-1983. They conclude that strategic investment decisions such as R&D
9
expenditures announcements are significantly and positively correlated to stock returns.
McConell and Nantell (1985) have examined the reaction of stock prices to joint venture formation. They consider joint ventures as a form of capital expenditure. Their sample is composed of 210 U.S. firms involved in 136 joint ventures during the period 1972-1979. Their results indicate that shareholders of companies involved in joint ventures earn positive abnormal returns around the announcement date of the joint venture. This is consistent with the study of Jarell et al. (1985), where they conclude that strategic investment decisions announcements are significantly and positively correlated to stock returns.
As for divestments, Jain (1985) examines the effect of voluntary sell-off announcements on shareholders’ wealth. In a sell-off, a parent company relinquishes control on part of its assets (e.g. a subsidiary, a segment) in exchange for cash or securities. The parent company continues to operate in the same form as prior to the sell-off. Using a sample of over 1,000 sell-off public announcements reported in the 1976 to 1978 issues of the Mergers and Acquisitions Journal, their results indicate that shareholders of the buying and selling side in the sell-off both earn positive abnormal returns around the announcement date of the sell-off.
The conclusions of previous studies described, show support for the shareholders’ value maximization where an investment or divestment will lead to positive abnormal returns. Firms make certain investment decisions with the expectation that the acquired corporate assets are capable of generating future earnings. If shareholders’ value maximization is prioritized, one would expect an announcement of an investment to lead to positive abnormal returns (Burton et al., 1999).
However, Fotaki et al. (2009) argue that an acquisition or a sale of corporate assets can be placed under the so-called non-synergetic and synergetic theories. Non-synergetic theories explain the decision of a sale or acquisition of corporate assets stems for reasons other than creation of shareholders’ wealth. It rather stems from managerial motives such as empire building, entrenchment and hubris. Therefore an asymmetric effect of acquisitions and divestments exists on the shareholders’ wealth.
They argue that divestures create value to shareholders’ wealth when a firm sells a component of its business that is not profitable. By selling part of its business, the firm will raise cash and therefore reducing agency costs. The non-synergetic theory predicts that a divesture will lead to positive abnormal returns as investors anticipate the reduction of agency costs as good
10
news. On the other hand, non-synergetic theory predicts that acquisitions will lead to negative abnormal returns as the stock market anticipate an investment as an increase of agency costs and less focus by management on the core business (Mulherin and Boone, 2000; Fotaki et al., 2009).
In contrast, synergetic theories predict a symmetric effect on the shareholders’ wealth. The synergetic theory claims that restructuring occurs, among others, in shifts in technology and when profit opportunities arise as the economy evolves. Synergetic theories create value for shareholders by putting assets more efficiently to use (Fotaki et al., 2009).
It is expected that an investment will follow the shareholders’ value maximization as firms make investment decisions with the expectation that the acquired corporate assets are capable of generating future earnings.
1.2 Types of Investments
While there is a great quantity of empirical studies that have analyzed the relationship between strategic investment announcements and stock returns, most studies focus on a specific type of strategic investment announcement (R&D, Jarell, Lehn and Marr (1985); joint venture formation, McConell and Nantell (1985) and voluntary sell-off, Jain (1985)). The study of Woolridge and Snow (1990), however, takes into account all types of strategic investment announcements that appeared in the Wall Street Journal during the period 1972-1987. Their final sample consists of 767 announcements made by 248 firms in 102 industries. They look at whether certain types of strategic investments announcements experience higher stock returns.
When football clubs make an investment, they assess the anticipated payoffs and the risk that comes with the transfer. Football clubs are more willing to invest in a young talented player rather than an experienced player at the end of his career. Kanyinda et al. (2012) show the relation between age and the margin of improvement of football players. They argue that the younger the player is, the higher his margin of improvement is. A younger football player is thus perceived to be a safer investment than an experienced football players at the end of his career.
Moreover, players being developed in an own youth system can be seen as a safe investment. The costs are relatively low and if it does not work out, the youth player can be sold for a transfer fee to another club that can improve the player more efficiently. Another attribute that can be seen a safe investment is the honor of being able to represent one’s country. Only the
11
most talented players are able to represent their country in a tournament such as the FIFA World Cup. Therefore, it is expected that if an international player is brought in, the market will react positively to these announcements.
The ability to attract crowds and sponsors is another factor that comes into play. Kanyinda et al. (2012) argue that players possessing these kind of abilities will be an additional asset for football clubs signing them as they can rely on additional income being generated. The image of players is therefore important and if they can attract other sources of income, this can then be perceived as a safe investment. One example are players in a forward position, they can determine matches and make headlines in media. Therefore, it is expected that an announcement of an acquisition of a forward position will generate more abnormal returns than other types of players. Playing in the same league as the buying club can also make the decision at ease for football clubs as the player already knows the league and environment. Players with different nationality and players that are transferred from other countries, will experience some mobility barriers, such as language, the need to adapt to another style of play and other cultural factors. It is therefore predicted that players brought in from different leagues will generate less abnormal return than players from the same league as they have to adapt to the new environment.
1.3 Strategies of Football Clubs
Football clubs are considered heterogeneous firms as each of them implement their own unique strategy to achieve their objectives. Some strategies of football clubs are quite similar in some aspects. Large clubs such as Real Madrid, Manchester United and FC Barcelona are globally well known, measured by their fan base and success on the pitch, and generate their revenues through advertising, ticket sales, television deals, branding etc. While smaller clubs tend to use advertising and ticket sales to keep afloat but use major transfer deals to grow and progress (Deloitte Football Money League, 2018). The firms taking similar approach can be grouped in a strategic group and can be considered homogenous as they face similar opportunities and threats according to Şener
and Karapolatgil (2015).
Rosetti and Caproni (2016) study the market strategies of football clubs to evaluate their sportive performances. Using a dataset consisting of 1,756 football clubs over a time span of 25 years, they identify several market profiles and cluster the football clubs that show similar market pattern through time. To avoid bias in their results, they neglect financial information capturing
12
economic power such as revenues, transfer fees paid and focus on trading strategies that a football club has followed such as buying players from the same league, buying young players and turnover of human resources.
Şener and Karapolatgil (2015) have analyzed the strategy of fifty global football club based on revenues, transfers and brand value. Their aim is to identify the main strategic groups and examine the differences between strategies. They found three strategic groups: Industry-leaders, runner-ups and weak clubs which follows an offensive-, distinctive image- and defense strategy respectively. Offensive strategy focuses on maintaining league positions with strong financial resources they possess. Distinctive image strategy is associated with the core values a football club follows that has led them to success in the past. Defense strategy has the aim to defend their league positions and to try to stay in the highest division of their respective league.
1.4 The Football Industry
According to the UEFA Benchmarking Reports (2016), football clubs’ main business revenues are derived from matchday- (e.g. tickets and season-tickets), broadcast- (e.g. broadcasting rights), commercial revenues (e.g. sponsorships, branding and merchandises) and player transfers. These sources of revenues have a strong impact on each other. Acquiring a top quality player will lead to more media attention which subsequently will lead to more commercial revenues and matchday revenues (Andras and Havran, 2015).
Football clubs face the challenge of reaching two objectives at the same time. There have been ongoing debates whether sport clubs’ main objective should be to maximize profit and act on behalf of the shareholders, as the main literatures suggest, or to maximize utility and keep its stakeholders, such as its fans, satisfied (Késenne and Pauwels, 2006; Garcia del Barrio and Szymanski, 2009). Football clubs are thus tasked with a performance trade-off. Football fans demand sporting performance while at the same time, shareholders demand improvement on the financial performance.
Késenne and Pauwels (2006) define maximizing utility as maximizing the winning percentage of the team. Managers face a trade-off between maximizing profit and helping the team reaching a higher level of performance. One possible way in which managers can meet both objectives is by participating in the transfer market by acquiring an even better player and selling football players that have been underperforming or no longer meet the technical requirements.
13
1.4.1 The Transfer Market
To build a competitive team that is able to compete for titles, increase the number of fans and strengthen sponsors- and media attention, football clubs are able to partake in the transfer market to reach their objectives. The transfer market is characterized as a labor market where a personal right with special value of property is the subject of the agreement. There exists a labor contract between the player and the club about the playing and advertising right for a certain period (Andras and Havran, 2015).
When a player with a valid contract is moved on to another club (i.e. a transfer), the signing football club is required to pay a given fee (i.e. a transfer fee) as a compensation. The transfer fee is regarded as an investment for the football club equivalent to an acquisition of an investment in fixed assets. The transfer fee can be in the form of a swap between players (i.e. player-for-player exchange), an all cash payment or a mixed combination containing both cash and player. Although, swapping of players does not occur often, this study focuses on all cash payments as the stock market might react differently when players are swapped and biased the results.
Football clubs have to comply with specific time windows in which they can acquire or sell players. This is known as transfer windows. Transfer windows were introduced as part of a compromise agreement with the European Commission about how the whole transfer system worked and how it could best preserve contractual stability for both the player and the club while allowing movement at prescribed times during the year. There are two transfer window where football clubs have the opportunity to adjust their squad. There is no exact date on the opening of the transfer windows as this varies by a few days each year. In general, however, the summer transfer window runs from the beginning of July to the end of September for most of the European football clubs. The winter transfer window is open in the month January. Football clubs are allowed to negotiate with other clubs outside the transfer window but players are officially transferred once the transfer window opens (Premier League, 2018). As the summer transfer window is open for a longer period of time than the winter transfer window, it is expected that investors will react negatively when a player is brought in during the winter transfer window as most clubs are unwilling to sell during the season and usually will require a premium to persuade the selling club to let go of the player.
Once a football player is acquired, football clubs have certain options on the future of the acquired player. The football club can decide to loan the player out immediately to another club to
14
gain experience, sell the player immediately to another club for a quick profit, incorporate the player in the first team, and send the acquired player to the reserves if deemed not ready to compete at the highest level (Kanyinda, Bouteiller and Karyotis, 2012).
Loans of players between football clubs are a common trading activity that occurs each transfer window. Loan deals are specific types of transfers where no transfer fees are involved, but rather a small financial fee (i.e. loan fee), and are used by football clubs to send out players temporarily to gain experience which they cannot obtain if they stayed in the current team due to competition in the team (Fotaki et al., 2009). It is predicted that the stock market will react positively when a player is loaned out, as the investment or asset will be put more efficiently to use.
1.4.2 The Bosman Verdict
One option that a football player has, is that he can move freely to another football organization after the expiration of his contract. This implies that the current football club will not receive any transfer fee (i.e. transfer fee is zero) if the contract is run down by the player. Players are allowed to enter in negotiation with another football entity once they have six months or less left on their current contract. This rule is known as the Bosman Verdict, introduced in December 1995 and named after football player Jean-Marc Bosman. Bosman took his case to the European Court of Justice after his former club, RFC Liège, refused to release him after his contract had expired and demanded a fee for his transfer. The European Court of Justice stated that the FIFA rules were in conflict with the freedom of movement as described in Article 48 of the Treaty of Rome and ruled the decision in favor of Bosman. Since then, the Bosman Verdict has had an impact on the transfers of players. The Bosman Verdict banned restrictions on the quota of foreign players and allowed players in the EU to move to another club at the end of a contract without a transfer fee being paid. As the player’s former club foregoes revenues, football clubs have changed their policy regarding transfers and the intention is to sell their valuable players while the player is still under contract (Késenne, 2011). It is expected that the nationality of the player will play no significant role in generating abnormal returns since the implementation of the Bosman Verdict.
15
2. Data
To examine the hypothesized relationship between the announcements of an acquisition and sale of football players and the stock prices, a sample of football transfers have been gathered. The sample consists of 3,359 football transfers of fourteen European football clubs that have their shares floating on an exchange as of 29th of June 2018. The sample period spans from the start of the clubs’ IPO date to the end of the 2017 summer transfer window (i.e. 31st of August 2017). An overview of the football clubs taken into consideration together with their IPO date are shown in table 1. All continuous financial metrics in this study are denoted in euros.
2.1 Data Sources and Sample Formation
Different sources are used to form the sample. ThomsonOne Reuters Datastream is used to look up the football clubs that have their shares floated on an exchange as of 29th of June 2018. The following filters have been imposed to obtain a list of football clubs: First, the category ‘Equities’ is selected; second, the sector ‘Travel and Leisure’ is selected; and lastly, a list of European countries have been selected for the ‘Market’ section. The final sample consists of fourteen European football clubs: Arsenal (England), Manchester United (England), Olympique Lyonnais (France), Borussia Dortmund (Germany), AS Roma (Italy), Juventus (Italy), SS Lazio (Italy), AFC Ajax (Netherlands), FC Porto (Portugal), SL Benfica (Portugal), Sporting Lisbon (Portugal), Besiktas (Turkey), Fenerbahce (Turkey) and Galatasaray (Turkey). Daily closing stock prices adjusted for dividends of the football clubs and prices of the market index, STOXX Europe Mid 200 Index, are extracted from the same database. Table 1 presents an overview of the football clubs taken into consideration for the sample formation.
16
Table 1. Sample overview
This table presents an overview of the football clubs taken into consideration for the sample. These football clubs are as of, 29th June 2018, listed on an exchange. The country the football club is located at, the IPO date and the market capitalization as of 2017Q4 (in millions of euros) are presented.
Club Country IPO Date Market capitalization
AFC Ajax NLD 11/5/1998 193.42
Arsenal ENG 9/8/2002 2,073.83
AS Roma ITA 22/5/2000 196.00
Besiktas TUR 19/2/2002 172.24
Borussia Dortmund GER 30/10/2000 539.01
Fenerbahce TUR 17/9/2004 195.92
FC Porto PRT 1/6/1998 15.30
Galatasaray TUR 19/2/2002 136.04
Juventus ITA 19/12/2001 727.61
Manchester United ENG 10/8/2012 2,425.24
Olympique Lyonnais FRA 8/2/2007 170.70
SL Benfica PRT 21/5/2007 29.67
Sporting Lisbon PRT 2/6/1998 46.90
SS Lazio ITA 6/5/1998 98.22
Subsequently, club characteristics such as revenues, earnings before interest and taxes, total assets, long-term debt, total debt, market capitalization, capital expenditures and total cash and short-term investment of football clubs are extracted from financial statements and from the database CapitalIQ.
Incoming and outgoing transfers are gathered from the website www.transfermarkt.com. This website records specific transfer details (i.e. transfer characteristics) such as the date of the transfer (i.e. the announcement date), the transfer fee, age of the player at the moment that the transfer took place and the new club the player has transferred to. For transfers without an announcement date or transfer fee, the official media of the football club has been consulted. If the announcement date or transfer fee is yet still unknown, due to reasons such as transfers having taken place a long time ago or the transfer fee being undisclosed, then these transfers are excluded. Player characteristics such as the position of a player, whether a player plays for its country (i.e. international status) and whether a player is developed in the youth system of the club are also gathered from this website. These data have been hand-collected using a web-scraping tool called
17
2.2 Descriptive Statistics
Before starting the analysis, it is intriguing to look at some descriptive statistics. The initial sample consists of 3,359 transfers that took place between May 1998 and August 2017, spread over fourteen European football clubs. Table 3 shows the total transfer activities for each club together with the total revenues generated and total costs incurred for the transfer activities. From the 3,359 transfers, 1,800 transfers (53.59 percent) involve an investment, while 1,559 transfers (46.41 percent) involve a divestment. After merging the dataset of the events and the financials, a final sample of 2,278 transfers is obtained.
Subsequently, the net balance is calculated for each club to determine whether a football club is a net buyer or net seller. The net balance is the difference between transfer revenues and transfer expenditures. A club is considered to be a net seller if the net balance is positive, while a negative net balance implies that the club is a net buyer. From the statistics, it is shown that nine out of the fourteen football clubs are considered net buyers. The full sample has a negative net balance. An overview on which clubs are net buyers or net sellers can be found in table 3. Further analysis on the net balance will be conducted in section 5.
Manchester United has the lowest transfer activity in the sample. This is explained by the fact that their shares started trading publicly as of August 2012. Therefore, they have the shortest time span of the football clubs in this sample having their shares floating on an exchange. Italian clubs AS Roma and Juventus have been the most active in the transfer market in this sample. Manchester United has the largest market capitalization, followed by Arsenal. The clubs with the smallest market capitalization are located in Portugal, whereby FC Porto has the smallest market capitalization, followed by SL Benfica and Sporting Lisbon. The average market capitalization in this sample equals 202 million of euros.
The youngest football player in the sample is midfield player Anderson, aged 16, who has been transferred from Brazilian football club Grêmio to FC Porto in the year 2006. The oldest football player involves goalkeeper Marco Ballotta, aged 41, when SS Lazio resigned him for a second spell in 2005. The average age that football players are being transferred in this sample is 25.35 years. While the average transfer fee in the sample is 3.95 million euros. The lowest transfer fee is zero euro which is in line with the Bosman Verdict. The most expensive transfer is the transfer of Ousmane Dembélé from Borussia Dortmund to FC Barcelona on 25th of August 2017. Financial ratios have been computed to use as control variables in this study. All financial
18
variables (i.e. club characteristics) are winsorized at the 1st and 99th percentile. Leverage and Cash have been taken to proxy the financial situation, while Profit and ROA are taken to proxy the profitability of football clubs. Market capitalization (MarCap) has been taken to proxy the size of the football clubs. Leverage, cash and ROA are normalized by total assets and profit is normalized by total revenues to make a comparison between football clubs. Definition of the variables can be found in table 4.
Table2.Summary statistics
The full sample consists of 3,359 football transfers between 1998 and 2017. Player and transfer characteristics are retrieved from transfermarkt.com and club characteristics are retrieved from CapitalIQ. Panel A presents summary statistics for the type of investments (i.e. player characteristics). Panel B presents summary statistics for transfer- and club characteristics. All continuous variables are denoted in millions of euros and are winsorized at the 1st and 99th percentiles.
Panel A: Type of Investment
Transfers
Type of investment Arrival
(N=1,559) Departure (N=1,800) Summer (Dummy = 1) Winter (Dummy = 0) Loan (Dummy = 1) Non-loan (Dummy = 0) League (Dummy = 1) Non-league (Dummy = 0) Position (Dummy = 1) Non-position (Dummy = 0) Nationality (Dummy = 1) Non-Nationality (Dummy = 0) International (Dummy = 1) Non-International (Dummy = 0) 84.16% 15.84% 10.01% 89.99% 40.28% 59.72% 31.62% 68.38% 50.61% 49.39% 58.05% 41.95% 81.67% 18.33% 14.89% 85.11% 42.56% 57.44% 31.89% 68.11% 57.39% 42.61% 58.17% 41.83%
19 Panel B: Transfer- and club characteristics
Variables N Mean Median Std. dev. Min Max
Age Transfer fee Revenues 3,359 3,359 3,130 25.35 3.95 34.97 25.00 1.00 24.55 4.25 8.09 30.80 16.00 0.00 4.19 41.00 115.00 151.20
Cash and short term investment 2,937 27.31 9.26 43.15 0.09 203.73
Total debt 3,008 123.44 83.92 124.02 0.00 580.35 Total assets 2,877 337.10 250.03 297.92 46.89 1,684.74 EBITDA 3,130 -3.69 -3.90 13.38 -32.00 41.34 Market cap 2,504 201.76 119.17 383.41 5.39 2,425.24 Leverage (%) 2,861 0.39 0.35 0.32 0.00 1.66 Cash (%) 2,803 0.07 0.04 0.09 0.00 0.45 ROA (%) 2,871 -0.01 -0.02 0.06 -0.21 0.15 Profit (%) 3,130 -0.22 -0.16 0.64 -2.35 1.73
20
Table3.Overview transfers
The full sample consists of 3,359 football transfers between 1998 and 2017. An overview of the transfer activity of each football club, together with the costs and revenues generated from transfers, are presented in this table. Furthermore, the net balance is calculated to determine if a club is a net buyer or net seller. The net balance is the difference between transfer revenues and transfer expenditures. A club is considered to be a net seller if the net balance is positive, while a negative net balance implies that the club is a net buyer. This set of information makes it possible to conduct a cluster analysis.
Football Club Period Subsample
Arrivals Total Costs (in millions EUR) Subsample Departures Total Revenues (in millions EUR Full Sample Net Balance (in millions EUR) Net Buyer/Seller (1) (2) (3) (4) (1) + (3) (4) – (2)
AFC Ajax 1998-2017 103 271.15 147 538.88 250 267.73 Net seller
Arsenal 2002-2017 72 683.39 124 450.18 196 -233.21 Net buyer
AS Roma 2000-2017 175 864.50 175 671.62 350 -192.88 Net buyer
Besiktas 2002-2017 136 186.75 129 83.61 265 -103.14 Net buyer
Borussia Dortmund 2000-2017 98 508.70 134 497.70 232 -11.00 Net buyer
Fenerbahce 2004-2017 84 296.71 78 103.81 162 -192.90 Net buyer
FC Porto 1998-2017 138 470.49 171 937.41 309 466.92 Net seller
Galatasaray 2002-2017 144 282.17 139 116.00 283 -166.17 Net buyer
Juventus 2001-2017 159 999.75 183 641.18 342 -358.27 Net buyer
Manchester United 2012-2017 26 830.83 47 218.34 73 -612.49 Net buyer
Olympique Lyonnais 2007-2017 55 358.29 93 418.07 148 59.78 Net seller
SL Benfica 2007-2017 125 366.99 138 767.97 263 400.98 Net seller
Sporting Lisbon 1998-2017 125 219.07 127 353.69 252 134.62 Net seller
SS Lazio 1998-2017 119 621.40 115 510.37 234 -111.03 Net buyer
21
3. Methodology
3.1 Event Study
This study uses an event study and panel data OLS regressions to investigate the market reaction to the announcement of an acquisition or sale of a football player by public listed European football clubs. The impact of a corporate announcement on the stock prices of companies can be measured by the change that the stock prices experience around the time the announcement is made public. For this study, the sale and acquisition of a football player are regarded as corporate events. To isolate the effect of a particular announcement, its stock return must be adjusted for the expected return on the stock. This study uses the market-adjusted return model to calculate the abnormal returns. The actual return (𝑅𝑖𝑡) minus the normal return (𝑁𝑅𝑖𝑡) is called the abnormal return (𝐴𝑅𝑖𝑡) as shown by equation (1):
𝐴𝑅𝑖𝑡 = 𝑅𝑖𝑡 − 𝑁𝑅𝑖𝑡 (1)
where 𝑁𝑅𝑖𝑡 is estimated as the market return of STOXX Europe Mid 200 Index, 𝑅𝑖𝑡 is the actual stock return of each football club and 𝐴𝑅𝑖𝑡 is the abnormal return for stock 𝑖 on day 𝑡.
The estimation for the market return, STOXX Europe Mid 200 Index, is a fixed component number index that represents mid-capitalization companies in Europe. The index covers companies across Austria, Belgium, Denmark, Finland, France, Germany, Czech Republic, Ireland, Italy, Luxembourg, the Netherlands, Norway, Portugal, Spain, Sweden, Switzerland and the United Kingdom (STOXX, 2018).
To assess the impact of the corporate announcement over time, the abnormal returns are aggregated typically over the days before, during and after the event. This is shown in equation (2):
𝐶𝐴𝑅𝑛 = ∑𝑛𝑡=1𝐴𝑅𝑡 (2)
where 𝑛 is the total number of days. In this study, multiple announcements are made by each of the football club. To make the analysis feasible, each event is treated as if they concern separate firms as pointed out by De Jong and De Goeij (2011). The announcement date is set at time is
22
equal to zero. For this study, different event windows are taken to test the reaction of the market on the types of the investment and strategies of football clubs. In the event window, the announcement date is centered. CAR3 is defined as the cumulative abnormal returns in a three day window surrounding the transfer announcement, thus one day before and after the announcement date are taken in the event window.
3.2 Cluster Analyses
With the aim of examining whether football clubs that follow similar strategies in their trading activities experience the same abnormal returns in their stock returns, a multivariate analysis is conducted. Similar to the study of Rosetti and Caproni (2016), football clubs are clustered using trading activities information rather than information capturing economic power of football clubs. One drawback, however, for this section is that the sample in this study only contains fourteen football clubs. To make this analysis feasible, a cluster of maximum two has been applied.
Football clubs’ annual report and data gathered from www.transfermarkt.com are used to conduct a cluster analysis. Some football clubs report in their annual statement the trading strategy they execute as this strategy has been successful over the years and form part of their core values. The full sample is split up in two subsamples ‘Arrivals’ and ‘Departures’ to get a better understanding of the strategy implemented.
The trading activities are calculated using ratios expressed in percentages as shown in equation (3):
𝜋𝑖 = |𝜋𝑖|
𝑇𝑜𝑡𝑎𝑙 𝑡𝑟𝑎𝑛𝑠𝑓𝑒𝑟𝑖
(3)
where 𝜋𝑖 is the ratio of a transfer attitude indicator variable to the total transfers of club i and |𝜋𝑖| is the transfer attitude indicator of club i. For the transfer attitude indicators, the variables youth,
international, league and nationality are taken. On the basis of these variables, a cluster will be
created to categorize the football clubs into groups. For this study, two cluster analyses are conducted. One on the basis of whether clubs are net spenders or net buyers and the other on the basis of a combination of the transfer attitude indicators.
23
3.3 OLS Regressions
To examine the effect of an acquisition on the stock prices, the cumulative abnormal return of three-, five and seven days are taken as dependent variable. The cumulative abnormal return is a proxy of value creation (or destruction) and regressed on the dummy variable arrival. The variable
arrival is an indicator variable that takes the value of one if the transfer concerns an acquisition
and takes the value of zero if the transfer concerns a sale of a football player. Financial ratios are computed and taken as control variables. The definition of the variables is found in table 4. To examine whether an acquisition is value creating for shareholders, regression (4) is applied:
𝐶𝐴𝑅𝑛𝑖 = 𝛽1∗ 𝐴𝑟𝑟𝑖𝑣𝑎𝑙𝑖𝑡+ 𝛽2∗ 𝐿𝑒𝑣𝑒𝑟𝑎𝑔𝑒𝑖𝑡+ 𝛽3∗ 𝑀𝑎𝑟𝐶𝑎𝑝𝑖𝑡 (4) + 𝛽4∗ 𝐶𝑎𝑠ℎ𝑖𝑡+ 𝛽5∗ 𝑅𝑂𝐴𝑖𝑡+ 𝛽6∗ 𝑃𝑟𝑜𝑓𝑖𝑡𝑖𝑡+ 𝛾𝑡+ 𝛾𝑐 + 𝜀𝑖𝑡
where the dependent variable is the cumulative abnormal return of event 𝑖 for 𝑛 days. Moreover, the regression contains time and country fixed effects to capture the market down- and upswing in stocks, among other time and country related effects, to reduce any potential omitted variable bias. Furthermore, an error term is included which is expected to have a value of zero.
Regression (4) is extended with a cluster group variable, 𝛿, interacting with the main interest variable, arrival, to determine whether strategies of football clubs matter in creating value to shareholders. This is shown in regression (5):
𝐶𝐴𝑅𝑛𝑖 = 𝛽1∗ 𝐴𝑟𝑟𝑖𝑣𝑎𝑙𝑖𝑡+ 𝛽2∗ 𝛿 + 𝛽3∗ 𝐴𝑟𝑟𝑖𝑣𝑎𝑙𝑖𝑡∗ 𝛿
+ 𝛽4 ∗ 𝐿𝑒𝑣𝑒𝑟𝑎𝑔𝑒𝑖𝑡+ 𝛽5∗ 𝑀𝑎𝑟𝐶𝑎𝑝𝑖𝑡+ 𝛽6∗ 𝐶𝑎𝑠ℎ𝑖𝑡+ 𝛽7∗ 𝑅𝑂𝐴𝑖𝑡 (5) + 𝛽8∗ 𝑃𝑟𝑜𝑓𝑖𝑡𝑖𝑡+ 𝛾𝑡+ 𝛾𝑐 + 𝜀𝑖𝑡
To test how investors react to types of investments, the cumulative abnormal return of three-, five and seven days are taken as dependent variable and regressed on transfer characteristics and player characteristics to examine whether investors react differently to certain types of investments. A definition of the independent variables (i.e player- and transfer characteristics) are described in table 4.
24
Regression (6) is used to test the statement above:
𝐶𝐴𝑅𝑛𝑖 = 𝛽1∗ 𝑆𝑢𝑚𝑚𝑒𝑟𝑖𝑡+ 𝛽2∗ 𝑇𝑟𝑎𝑛𝑠𝑓𝑒𝑟 𝑓𝑒𝑒𝑖𝑡+ 𝛽3∗ 𝐿𝑜𝑎𝑛𝑖𝑡+ 𝛽4∗ 𝐿𝑒𝑎𝑔𝑢𝑒𝑖𝑡 + 𝛽5∗ 𝐴𝑔𝑒𝑖𝑡+ 𝛽6∗ 𝑃𝑜𝑠𝑖𝑡𝑖𝑜𝑛𝑖𝑡 + 𝛽7∗ 𝑁𝑎𝑡𝑖𝑜𝑛𝑎𝑙𝑖𝑡𝑦𝑖𝑡+ 𝛽8∗ 𝐼𝑛𝑡𝑒𝑟𝑛𝑎𝑡𝑖𝑜𝑛𝑎𝑙𝑖𝑡 + 𝛽9∗ 𝐿𝑒𝑣𝑒𝑟𝑎𝑔𝑒𝑖𝑡+ 𝛽10∗ 𝑀𝑎𝑟𝐶𝑎𝑝𝑖𝑡+ 𝛽11∗ 𝐶𝑎𝑠ℎ𝑖𝑡+ 𝛽12∗ 𝑅𝑂𝐴𝑖𝑡 (6) + 𝛽13∗ 𝑃𝑟𝑜𝑓𝑖𝑡𝑖𝑡+ 𝛾𝑡+ 𝛾𝑐 + 𝜀𝑖𝑡
where the dependent variable is the cumulative abnormal return of event 𝑖 for 𝑛 days. Moreover, the regression contains time and country fixed effects to capture the market down- and upswing in stocks, among other time and country related effects, to reduce any potential omitted variable bias. Furthermore, an error term is included which is expected to have a value of zero.
Table 4. Variables definition
This table presents the definition of the variables used in this study. TC and PC indicate transfer characteristics and player characteristics respectively.
Variables Definition
Dependent variable(s)
CAR3 Cumulative abnormal returns in a three day window surrounding the transfer
announcement using the market-adjusted return approach (in percentage points).
CAR5 Cumulative abnormal returns in a five day window surrounding the transfer announcement using the market-adjusted return approach (in percentage points).
CAR7 Cumulative abnormal returns in a seven day window surrounding the transfer
announcement using the market-adjusted return approach (in percentage points).
Independent variable(s)
Arrival
Summer Transfer fee
An indicator variable that takes the value of one if the transfer concerns an acquisition of a football player and the value of zero if the deal concerns a sale of a football player.
An indicator variable that takes the value of one if the transfer is completed in the summer transfer window, and takes the value of zero otherwise (TC). The sum football clubs receive (pay) when a football player is sold (bought) to
25 Loan League Position Nationality International Age
(from) another club (TC).
An indicator variable that takes the value of one if the transfer is a loan, and takes the value of zero otherwise (TC).
An indicator variable that takes the value of one if the football player is transferred to (from) another club that plays in the same league as the selling (buying) club, and takes the value of zero otherwise (TC).
An indicator variable that takes the value of one if the football player is a forward and takes the value of zero otherwise (PC).
An indicator variable that takes the value of one if the football player holds a nationality from a country from the European Union, and takes the value of zero otherwise (PC).
An indicator variable that takes the value of one if the football player has played at least once for his country, and takes the value of zero otherwise (PC). The age of the football player at the moment of the transfer (PC).
Control variable(s) Leverage MarCap Cash ROA Profit
Book value of debt divided by book value of assets (in percentage points). The stock price times the number of shares outstanding.
Football clubs’ cash holdings and short-term investments divided by book value of assets (in percentage points).
EBITDA divided by book value of assets (in percentage points). EBITDA divided by total revenues (in percentage points).
Transfer attitude indicator(s)
Net trader Youth
International League
Nationality
An indicator variable that takes the value of one if the football club is a net buyer and takes the value of zero otherwise.
An indicator variable that takes the value of one if the football player is developed in the youth system of the football club and takes the value of zero otherwise.
An indicator variable that takes the value of one if the football player has played at least once for his country, and takes the value of zero otherwise.
An indicator variable that takes the value of one if the football player is transferred to (from) another club that plays in the same league as the selling (buying) club, and takes the value of zero otherwise.
An indicator variable that takes the value of one if the football player holds a nationality from a country from the European Union, and takes the value of zero otherwise.
26
4. Results
4.1 The Effect of an Acquisition on the Stock Prices
In the following section, the effect of an acquisition of a football player on the stock price is tested. To examine the value implication of an acquisition of a football player, an OLS regression is run in which the dependent variable is the cumulative abnormal returns of three-, five- and seven-day, respectively CAR3, CAR5 and CAR7, and the key independent variable is the indicator variable,
arrival. This variable takes the value of one if the deal concerns an acquisition of a football player
and the value of zero if the deal concerns a sale of a football player.
Table 5 presents the results. Regressions (1), (3) and (5) present univariate analyses and regressions (2), (4) and (6) control for additional measures in the form of financials of football clubs (i.e. club characteristics). For each regression, a positive relationship is found between an acquisition and their respective CAR. The variable arrival, however, is found to be statistically insignificant for each regression. The results obtained from the analysis are in line with the study of Burton et al. (1999), McConell and Nantell (1985) and Jarell et al. (1985). They have found a positive relation between an announcement of an investment and stock returns, which supports the view that shareholders’ value are being maximized.
The results show that the longer the period around the announcement date, the larger the coefficient of the main variable becomes. When an investment is announced, most investors have already anticipated that the acquisition is being made, thus less abnormal returns are generated in a smaller event window.
Although the results are statistically insignificant, the size of the coefficients implies an economically significant effect for the shareholders of football clubs. Take for example regression (4): An acquisition will lead to an increase of 0.8 percent in stock returns around the announcement date. Given that the average football club in the sample has a market capitalization of 202 million euros, a 0.8 percent rise in stock returns around the announcement date corresponds to a value increase of 1.62 million euros.
27
Table5.The effect of an acquisition
This table presents estimates from OLS regressions in which the dependent variable is the cumulative abnormal return of three-, five- and seven-day (CAR3, CAR5 and CAR7 respectively). The main interest variable, arrival, is an indicator variable that takes the value of one if the deal concerns an acquisition of a football player and the value of zero if the deal concerns a sale of a football player. Standard errors, which are adjusted for clustering at club level, are reported in parentheses. All variables are defined in table 4. ***, **, * correspond to statistical significance at the one, five and ten percent levels, respectively.
(1) (2) (3) (4) (5) (6)
Dependent variable CAR3
[-1,+1] CAR3 [-1,+1] CAR5 [-2,+2] CAR5 [-2,+2] CAR7 [-3,+3] CAR7 [-3,+3] Arrival 0.002 0.000 0.007 0.008 0.006 0.009 (0.003) (0.003) (0.007) (0.009) (0.008) (0.011) Leverage 0.005 -0.008 -0.007 (0.006) (0.025) (0.033) MarCap 0.000 0.000 0.000 (0.000) (0.000) (0.000) Cash 0.008 0.016 0.021 (0.008) (0.026) (0.037) ROA 0.050 0.048 0.104 (0.061) (0.072) (0.095) Profit -0.005 -0.005 -0.013 (0.010) (0.011) (0.016) Observations 3,359 2,278 3,359 2,278 3,359 2,278 R2 0.014 0.033 0.027 0.050 0.031 0.057
Country FE Yes Yes Yes Yes Yes Yes
Year FE Yes Yes Yes Yes Yes Yes
4.2 Stock Market Reaction on Transfer- and Player Characteristics
So far, a positive relationship is found between an acquisition and its respective CAR. In this section, an analysis is made on the type of investments (i.e. player characteristics) and the decisions taken by the football clubs (i.e. transfer characteristics). The full sample is split up in two subsamples, Arrivals and Departures, to interpret the results. Different event time windows are used to calculate the cumulative abnormal return and regressed on transfer and player characteristics to determine whether certain characteristics generate more abnormal returns. Table 6, panel Apresents the results for the subsample Arrivals. The Arrivals subsample consists of 1,559 acquisitions. After controlling for club characteristics, a total of 1,034
28
acquisitions is left. Table 6, panel B presents the Departure subsample which consists of 1,800 sales. After controlling for club characteristics a sample of 1,242 sales is analyzed. In the first section, the transfer characteristics are analyzed. Afterwards, the player characteristics are examined. An overview of the definitions of the variables can be found in table 4.
29
Table6.The effect of player characteristics and transfer characteristics
This table presents estimates from OLS regressions in which the dependent variable is the cumulative abnormal return of three-, five- and seven-day (CAR3, CAR5 and CAR7 respectively). Panel A presents the results for the subsample Arrivals. Panel B presents the results for subsample Departures. Standard errors, which are adjusted for clustering at club level, are reported in parentheses. All variables are defined in table 4. ***, **, * correspond to statistical significance at the one, five and ten percent levels, respectively.
Panel A: Subsample Arrivals
(1) (2) (3) (4) (5) (6)
Dependent variable CAR3
[-1,+1] CAR3 [-1,+1] CAR5 [-2,+2] CAR5 [-2,+2] CAR7 [-3,+3] CAR7 [-3,+3] Summer 0.002 -0.001 0.012 0.014 0.010 0.019 (0.003) (0.005) (0.010) (0.013) (0.010) (0.014) Transfer fee -0.000 -0.000 -0.000 -0.001 -0.000 -0.001 (0.000) (0.000) (0.000) (0.001) (0.000) (0.001) Loan 0.003 -0.002 0.005 -0.004 0.005 -0.007 (0.005) (0.005) (0.007) (0.010) (0.007) (0.013) League 0.002 0.007 0.011 0.018 0.011 0.020 (0.004) (0.006) (0.009) (0.016) (0.012) (0.022) Age -0.000 -0.000 0.001 0.001 0.000 -0.000 (0.000) (0.000) (0.001) (0.001) (0.001) (0.001) Position 0.001 -0.004 0.008 0.002 0.006 0.002 (0.004) (0.003) (0.005) (0.006) (0.006) (0.006) Nationality 0.001 0.001 -0.004 -0.010 -0.008 -0.016 (0.003) (0.003) (0.008) (0.013) (0.009) (0.016) International 0.002 0.004 0.001 -0.001 0.004 0.004 (0.003) (0.004) (0.006) (0.008) (0.005) (0.008) Leverage -0.000 -0.026 -0.026 (0.010) (0.048) (0.061) MarCap -0.000 0.000 0.000 (0.000) (0.000) (0.000) Cash 0.006 0.008 0.029 (0.024) (0.055) (0.070) ROA 0.046 0.097 0.053 (0.051) (0.117) (0.185) Profit -0.003 -0.008 -0.003 (0.006) (0.015) (0.020) Observations 1,559 1,034 1,559 1,034 1,559 1,034 R2 0.030 0.059 0.063 0.112 0.072 0.130
Country FE Yes Yes Yes Yes Yes Yes
30 Panel B: Subsample Departures
(1) (2) (3) (4) (5) (6)
Dependent variable CAR3
[-1,+1] CAR3 [-1,+1] CAR5 [-2,+2] CAR5 [-2,+2] CAR7 [-3,+3] CAR7 [-3,+3] Summer -0.000 -0.003 0.003 0.001 -0.000 -0.005 (0.004) (0.007) (0.006) (0.010) (0.008) (0.013) Transfer fee 0.000 0.000 0.000 0.000 0.000 0.000 (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) Loan 0.003 0.008** 0.004 0.010** 0.006 0.016*** (0.002) (0.003) (0.004) (0.005) (0.004) (0.005) League -0.005** -0.009** -0.004 -0.008 -0.003 -0.006 (0.002) (0.004) (0.003) (0.005) (0.003) (0.006) Age 0.000 -0.000 -0.000 -0.001 0.000 0.000 (0.000) (0.000) (0.001) (0.000) (0.001) (0.001) Position -0.000 -0.003 -0.001 -0.006 -0.001 -0.005 (0.003) (0.004) (0.004) (0.004) (0.005) (0.006) Nationality 0.004 0.006 0.004 0.006 0.003 -0.000 (0.003) (0.004) (0.005) (0.005) (0.004) (0.007) International -0.001 -0.002 -0.003 -0.000 -0.005 -0.005 (0.003) (0.004) (0.003) (0.005) (0.004) (0.005) Leverage 0.009 0.012* 0.016** (0.006) (0.007) (0.006) MarCap 0.000 0.000 0.000 (0.000) (0.000) (0.000) Cash 0.017 0.023 0.021 (0.019) (0.015) (0.029) ROA 0.048 -0.003 0.161 (0.089) (0.102) (0.168) Profit -0.008 -0.004 -0.023 (0.014) (0.014) (0.026) Observations 1,800 1,242 1,800 1,242 1,800 1,242 R2 0.028 0.050 0.028 0.044 0.026 0.048
Country FE Yes Yes Yes Yes Yes Yes
Year FE Yes Yes Yes Yes Yes Yes
4.2.1 Transfer Characteristics
The indicator variable loan takes the value of one if the deal concerns a temporary loan deal of a football player and otherwise the value zero. For the subsample Arrivals, the results show a negative and insignificant effect for the variable loan. Loaning in a football player (i.e. arrival) from another football club generates negative abnormal returns.
31
For the subsample Departures, loaning out a football player (i.e. departure) generates positive abnormal returns and significant results. For CAR3 and CAR5, the loan variable is significant at a five percent level. For the CAR7 as dependent variable, this variable is significant at a one percent level. Loaning out a player will generate 0.8 percent abnormal returns, which has economically significance for the shareholders of football clubs. The stock market anticipate loaning out of a football player as positive news. The results obtained are consistent with the results of Fotaki et al. (2009) and can be related to synergetic theories where loaning out a player can be considered an analogy where assets are being operated more efficiently.
The variable league is a dummy variable that takes the value of one if the player is
transferred from or to the same league and otherwise the value zero. The results show positive (negative) association with CAR if the deal concerns an acquisition (sale). Selling a football player to another football club in the same league will result in a drop of 0.9 percent in stock returns around the announcement date when the CAR3 is taken as dependent variable. This result is statistically significant at a five percent level for the subsample Departures when CAR3 is taken as dependent variable. This has economic consequences for shareholders of football clubs. A drop in the stock returns for a departure can be explained by investors reacting to the news negatively. Selling a football player to another club in the same league implies the selling club weakening itself and strengthening another competitor in the league.
The variable transfer fee is defined as the transfer sum that the club has paid (received) for an acquisition (sale). Although statistically insignificant for each of the regressions in the results for both subsamples, the sign of the results is expected. In case of an acquisition (sale), each additional one million euros paid (received), will generate negative (positive) abnormal returns. Investors react negatively if more transfer fees are needed to make an investment. Investors would want to pay as little as possible for an investment and receive as much as possible from a sale.
Players can be transferred in either the summer or winter transfer window. The dummy variable summer takes the value of one if the transfer is completed in the summer transfer window and zero if the transfer is completed in the winter transfer window. No statistically significant results are found for this variable for both subsamples. The size of the coefficients is close to zero and the signs of these coefficients are as expected. For arrivals (departures), positive (negative) abnormal returns are experienced during the summer transfer window. Investors anticipate transfers in the summer transfer window as positive news. This can be explained by the fact that
32
the summer transfer window is open for a longer period compared to the winter transfer window, giving football clubs more time to look for the right investment or find the right buyer. Moreover, football clubs are not willing to sell their players during the season (i.e. winter transfer window), as this can have an impact on the rest of the season, such as shortages in employment, and will require a premium to persuade other clubs to sell (Market Mogul, 2016).
4.2.2 Player Characteristics
Age is a player characteristic that is considered to be relevant when determining the transfer fee of
a football player (Ruijg and Van Ophem, 2014). The results, however, show no statistically significance in the stock returns for both subsamples. Age has a negative association with the stock returns for both subsamples. Kanyinda et al. (2012) argues that football players’ potential decreases as a football player grows older. Football clubs are willing to invest in a young talented player rather than an experienced player at the end of his career. Investors thus react negatively when an old player is brought in. For the departure subsample, a negative association between the age of the player and its CAR is somewhat surprising, but can be an indication on players being long servants for the football club and investors reacting negatively to these announcements.
The variable position is a dummy variable that takes the value of one if the football player is a forward and otherwise takes the value of zero. Players that play in a forward position get more media coverage and have the ability to attract crowds and sponsors. The stock market react positively to the news of an acquisition of a forward. The opposite is observed in the case of a departing striker. The results show no evidence that the position of a football player helps explain the stock price movements.
The dummy variable nationality takes the value of one if the football player holds a nationality from a country from the European Union, and takes the value of zero otherwise. The results show that the nationality of a football player does not play a significant role in generating abnormal returns. A negative (positive) association is found for the subsample Arrivals (Departures).
International is an indicator variable that takes the value of one if the football player has
played at least once for his country, and takes the value of zero otherwise. Players that play for their country are considered the most talented of their country and have enough experience to play a high level. Bringing in an international player is seen as positive news by the stock market, while
33
selling an international player will generate negative abnormal stock returns. Although, the results show no statistically significance.
4.3 Cluster Analyses
In this section, two separate analyses are conducted to determine whether certain club strategies will generate higher abnormal returns. This will be done on the basis of whether a club is a net buyer or net seller and the other on the basis of a combination of the transfer attitude indicators. Cluster analysis allows to categorize football clubs into groups that follow similar trend and compare the CAR of these groups with each other. The cluster analysis is composed with the help of multivariate analysis and the annual reports of the football clubs.
4.3.1 Strategy Based on Net Balance
Clubs are considered net buyers (net sellers) if they have a negative (positive) net balance. The net balance is the difference between transfer revenues and transfer expenditures. An overview of the clubs is shown in table 3. Football clubs from Portugal (i.e. FC Porto, SL Benfica and Sporting Lisbon) have the most positive net balance, followed by AFC Ajax and Olympique Lyonnais. The net balance of Borussia Dortmund is slightly negative. The bigger clubs, measured by market capitalization shown in table 1, such as Manchester United, Juventus and Arsenal have the most negative net balance. This suggests that the bigger clubs capitalize on their financial strength and use the transfer market to strengthen their team to compete for trophies. What is inquiring to observe is that the clubs with the smallest market capitalization have the most positive net balance while clubs with the largest market capitalization have the most negative net balance. This can be explained in the transfer strategy the clubs are implementing.
Table 8, panel A presents the results using the transfer attitude indicator net buyer and
net seller to interact with the main interest variable arrival. Similar to previous tests in this study,
different event windows are taken as the dependent variable. Due to multicollinearity, country fixed effects are excluded. Columns (1), (3) and (5) present the results for clubs that are considered net buyers, while columns (2), (4) and (6) present the results for net sellers. Similar results as previous tests are found on the effect of an acquisition on the stock prices. The variable arrival shows a positive association with its respective CAR. The results are statistically insignificant,
34
however, the results show some indication that shareholders’ value are being maximized when an acquisition is announced. When observing the size of the coefficient, it is observed that net sellers generate higher abnormal returns than net buyers when an investment is announced, indicating that investors of net sellers are more positive with the investment than investors of net buyers.
Furthermore, it is observed that negative abnormal returns are generated in CAR3 and CAR5 if the club is a net buyer. The opposite holds for clubs that are net sellers. This point towards investors being more positive with the actions taken by clubs that have a positive net balance. This, however, is proven to be statistically insignificant and thus no conclusion can be drawn that the stock movements are influenced by the strategy implemented by football clubs.
Next, the transfer attitude indicator variable is interacted with the main interest variable
arrival to interpret whether investors take the transfer strategy of football clubs into consideration.
An investment that is made by a club that is considered to be a net buyer generates negative abnormal returns in the event window of one day prior and after the announcement of the investment. For longer event windows, the sign of the coefficients changes and the size increases. The opposite results hold for net sellers. The results are statistically insignificant and no conclusion can be drawn that the transfer strategy influences the stock movements
4.3.2 Strategy Based on Transfer Attitude Indicators
For the next cluster analysis, the ratios of the transfer attitude indicator variables and strategies described in annual reports are used. A maximum of two clusters are created to make the analysis feasible as the sample size of the study only consists out of fourteen football clubs and this limits the analysis. Table 7 presents the two cluster groups which are called ‘grooming strategy’ and ‘ready-made strategy’.
The two cluster groups make a distinction on the strategy of the football clubs. The first cluster, grooming strategy, consists of football clubs focusing on youth development, buying inexperienced players and transferring players when they are experienced enough. This group sells the majority of their football players to another league. For the other group, ready-made strategy, the contrary can be argued. It consists of clubs buying ready-made players hoping to make an immediate impact. Their strategy thus rely on bringing in average-aged or older players that have enough experience.
