Olympic sponsorships, stock prices, and trading activity.
Olympic sponsors provide significant financial and in-kind resources that are vital to both the International Olympic Committee (IOC) and the local organizing committees. There were four sponsorship levels for the London Games that resulted in a total of $2.2 billion in sponsorships (Rogers, 2012).
Elite sponsors are known as worldwide Olympic Partners. These firms have established long-term relationships with the IOC through participation in the IOC's The Olympic Partners (TOP) Programme. Each TOP company contributed $100 million to the 2012 Games (Rogers, 2012). Sponsors at the other levels contracted directly with the London Organising Committee of the Olympic and Paralympic Games (LOCOG). These sponsors were known as London 2012 Official Olympic Partners, Official Olympic Supporters, and Official Olympic Providers and Suppliers. Each firm at each of these levels contributed $63 million, $31 million, and $15 million, respectively (Rogers, 2012). In addition, sponsors incur development costs related to their sponsorships.
While managers can easily calculate the costs of Olympic sponsorships, the benefits are less obvious. Corporate sponsors gain exposure and recognition, but placing a value on these measures is problematic. In addition, sponsorships provide perquisites to managers. Corporate sponsors are given blocks of tickets for Olympic events. At little personal cost, managers are invited to receptions and athlete visits, and are provided access to the Games in ways not given to the general public. Some have suggested that the presence of perks colors the decision of managers to commit the firm to activities that do not add to shareholder wealth, allowing managers to engage in what Jensen and Meckling (1976) call "non-pecuniary consumption." Thus, one might argue that managers may enter into costly sponsorship agreements partly because of the personal benefits they receive. This potential conflict of interests is known as the "agency problem" (see Eisenhardt, 1989; Jensen & Meckling, 1976).
This paper tests for the existence of abnormal stock returns and changes in trading volume on the dates companies announced their sponsorships of the London 2012 Olympic Games. Stock prices provide a metric that directly reflects changes in shareholder wealth, a measure that business managers should seek to maximize. The paper also investigates the possibility of agency effects in influencing a firm's decision to become an Olympic sponsor.
Specifically, we address the following research questions: 1) Do stock prices and trading volumes differ from the norm on sponsorship announcement dates for LOCOG Official Olympic Partners? 2) Do stock prices and trading volumes differ from the norm on sponsorship announcement dates for LOCOG Official Olympic Supporters? 3) Does the impact on stock prices and trading volumes differ between British and non-British firms? 4) Is there evidence that suggests non-pecuniary consumption influences managers' decisions to enter into Olympic sponsorships?
Managers and shareholders will find this investigation useful in evaluating if a firm's expenditures for Olympic sponsorships create value for the company's owners.
We use event-study methods to analyze the impact of sponsorship announcements on market values of equities of the sponsoring companies. Event studies were conducted as early as in the 1930s in a paper measuring the price impact of common stock splits (Dolley, 1933). The first rigorous attempt to measure the impact of events on stock prices was done by Fama, Fisher, Jensen, and Roll in 1969. Subsequent articles indicate that the event study technique is appropriate for relatively small sample sizes (Brown & Warner, 1985). Event studies rely on the assumption of semi-strong market efficiency, that stock prices correctly and quickly incorporate all publicly available information. The usefulness of event studies comes from the fact that the abnormal returns attributable to the events around the event announcements provide a direct measure of the impact of the event announcement on the announcing company's market value. This helps in corporate decision making. Many types of event studies appear in the literature, including event studies that examine trading volume (Beaver, 1968; Campbell & Wasley, 1996) and operating performance (Barber & Lyon, 1996). Smith (1986) reviews event studies of corporate financing decisions, while Jensen and Warner (1988) and Jarrell, Brickley, and Netter (1988) provide surveys of corporate control event studies. MacKinlay (1997) and Campbell, Lo, and MacKinlay (1997) provide surveys of event studies in finance and economics, and discuss event-study methodologies.
Numerous papers have analyzed the effect of sponsorship announcements on sponsoring firms' stock price reactions and have found conflicting results. Mishra, Bobinski, and Bhabra (1997) enumerate several factors that may increase or decrease stock prices following sponsorship announcements.
The event-study technique also has been employed by many authors to investigate the relationship between stock prices and sponsorship announcements in other sports. Cornwell, Pruitt, and Van Ness (2001) found that automotive industry sponsors of the winning car in the Indianapolis 500 had positive abnormal returns following the race, but sponsors from other industries did not (Cornwell et al., 2001). Pruitt, Cornwell, and Clark (2004) estimated that NASCAR sponsorship announcements were associated with mean increases in shareholder wealth of more than $300 million. In addition, the NASCAR study found a negative link between cash flow per share and returns, suggesting the presence of an agency problem (Pruitt et al., 2004). Cornwell, Pruitt, and Clark (2005) found that sponsorships related to the "five most prominent sporting associations in the United States" (p. 403) increased shareholder wealth by an average of $257 million, net of development costs (Cornwell et al., 2005). The same paper noted that "a direct product linkage to the sponsored sport is an important facet of the stock market's acceptance of an official sport sponsorship" (p. 410).
Reiser, Breuer, and Wicker (2012) expanded the focus to include naming rights and sponsorships for US baseball, basketball, football, golf, motor sports, soccer, and tennis, as well as the Olympics. The paper also assessed the existence of differential effects related to the sponsoring companies' locations (Reiser et al., 2012). While the authors found a positive association between announcements and returns for North American and European companies, they found a negative association for companies from the Asia/Pacific region for some event windows (Reiser et al., 2012). In addition, the authors found a slightly significant positive relationship for Olympic sponsorship announcements (Reiser et al.).
Examining the relationship between Olympic sponsorships and stock prices seems particularly appropriate because, according to Cornwell et al., 2005, "the meaning and marketing importance ascribed to the term "official sponsor" might be traced back to the 1984 Olympic Games" (p. 402).
Floros (2010) and Yelkur, Tomkovick, and Pennington (2012) measure the effect on sponsor stock prices at times other than announcement dates. However, the current study focuses solely on the capital market reaction to sponsorship announcements and builds on six prior papers related to the Atlanta, Athens, and Beijing Games.
Two papers studied the Atlanta Games. Farrell and Frame (1997) found significant negative abnormal returns for the two days following an announcement, but no significant effects on the announcement day (day 0). The authors noted that the negative effect was mitigated for firms with substantial institutional ownership, indicating a potential agency issue. In contrast, Miyazaki and Morgan (2001), using a different combination of companies, estimation periods, and announcement windows, did not find significant negative abnormal returns and found a significant positive abnormal return for one window.
Spais and Filis (2006) and Samitas, Kenourgios, and Zounis (2008) both studied the effects of Athens sponsorship announcements. Spais and Filis (2006) analyzed the capital market behavior for three Greek companies by comparing performance during a 41-day event window to the performance over a 200-day pre-event and a 200-day post-event window. They found significantly positive abnormal returns for one Grand National Sponsor, but insignificant abnormal returns for the other two (Spais & Filis, 2006). In addition, the authors noted changes in volatility and increased trading volume for two of the companies (Spais & Filis, 2006). Samitas, Kenourgios, and Zounis (2008) used bootstrap techniques to study the impact on stock prices on dates surrounding 21 sponsorship announcements. The authors also examined the relationship between sponsoring firms' stock prices and the opening ceremony (Samitas et al., 2008). The paper found positive abnormal returns for three windows surrounding the announcements, but did not find significant results on day 0 (Samitas et al., 2008). The paper generally did not find significant abnormal returns surrounding the opening ceremony (Samitas et al., 2008).
Tsiotsou (2011) studies the effects of announcements for five sponsors of the Athens Games. The paper found insignificant price reactions over every window and on day 0, but found small significant offsetting effects two days and one day before the official announcement date (Tsiotsou, 2011).
Molchanov, Stork, and Zeng in a 2010 working paper, investigated stock price behavior around announcements and opening ceremony dates for the Beijing Games. In contrast to Samitas et al. (2008), Molchanov et al. (2010) did not find significant abnormal returns around announcement dates, but did find significant positive abnormal returns for international sponsors surrounding the opening ceremony. The authors suggested that international sponsors may have been focused on stock returns while domestic firms may have been driven by national pride instead of shareholder wealth maximization (Molchanov et al., 2010).
Farrell and Frame (1997) and Pruitt et al. (2004) offer another non-shareholder wealth maximizing motivation for managers to enter into sports sponsorship agreements--specifically, to enhance their personal satisfaction. Our study examines three metrics--cash holdings, institutional ownership, and insider holdings--which some have argued are linked to principal-agent conflicts.
Due to the inability of shareholders to monitor managers when there are large cash holdings, Jensen (1986) argued that excess cash holdings present managers an opportunity to acquire investments that yield below-market returns. This reasoning has led many to argue that "the effectiveness of shareholder monitoring of corporate expenses declines as cash flows rise" (Pruitt et al., 2004, p. 290). Because sponsorship agreements provide unique personal gains to managers, they may be the type of investment that Jensen envisioned. As mentioned earlier, Pruitt et al. (2004) concluded that an agency problem related to cash flows might explain the decision to enter into some NASCAR sponsorships.
Institutional ownership as a second test for the presence of an agency problem is also related to the ability of shareholders to oversee managerial decisions. Agrawal and Knoeber (1996) asserted that "concentrated shareholdings by institutions, or by block-holders, can increase managerial monitoring and so improve firm performance" (p. 377). Following this reasoning, Aggarwal, Erel, Ferreira, and Matos (2011) concluded that firms with larger institutional holdings more quickly terminated poorly performing CEOs.
Our third test for the presence of an agency problem originates with Jensen and Meckling's (1976) seminal paper, in which they argue that managers who become owners have interests more aligned with those of outside shareholders. This reasoning suggests that firms with more insider holdings should have fewer agency issues. However, subsequent studies lead to a more ambiguous conclusion. Sanchez-Ballesta and Garcia-Meca (2007) summarize several articles by stating that "when ownership is too concentrated the value of the firm starts to decrease" (p. 880) due to the difficulty in firing owner-managers.
Our first research question relates to stock price and trading volume performance on sponsorship announcement dates for LOCOG Official Olympic Partners; our second question looks at the same performance measures for LOCOG Official Olympic Supporters. We investigate the two levels separately because of the substantial difference in commitment levels. We test the following four null hypotheses:
[H.sub.1A]: There are no abnormal returns for LOCOG Official Olympic Partners on their sponsorship announcement dates.
[H.sub.1B]: There are no significant changes in trading volume for LOCOG Official Olympic Partners on their sponsorship announcement dates.
[H.sub.2A]: There are no abnormal returns for LOCOG Official Olympic Supporters on their sponsorship announcement dates.
[H.sub.2B]: There are no significant changes in trading volume for LOCOG Official Olympic Supporters on their sponsorship announcement dates.
Our third research question relates to the importance of a sponsor's home country. This question arises from Molchanov et al. (2010), which found different capital market effects for host-country and foreign sponsors around the opening ceremony. Due to sample size issues, we do not differentiate between supporters and partners for this test. We test the following two null hypotheses:
[H.sub.3A]: There are no significant differences in stock price behavior between British and non-British sponsors.
[H.sub.3B]: There are no significant differences in trading volume behavior between British and non-British sponsors.
Our final research question deals with the presence of agency effects. Several articles have raised the possibility that the decision to sponsor sporting events may be influenced by the lure of non-pecuniary consumption. Farrell and Frame (1997) and Pruitt et al. (2004) found evidence to support opportunistic managerial behavior. Agency theory literature suggests that high cash holdings and low levels of institutional ownership create opportunities for managers to engage in behavior inconsistent with shareholder interests. Therefore, we propose the following two null hypotheses as one-tailed tests:
[H.sub.4A]: Sponsoring firms' cash holdings are less than or equal to those of industry peers.
[H.sub.4B]: Sponsoring firms' percentages of institutional holdings are greater than or equal to those of industry peers.
Because the evidence on the effect of insider holdings percentage on a manager's likelihood of engaging in opportunistic behavior is mixed, we use a two-tailed test for this hypothesis.
[H.sub.4C]: There is no evidence that percentages of insider holdings of the sponsoring firms' differ from those of their industry peers.
Table 1 lists the seven official Olympic Partners and seven official Olympic Supporters of the 2012 Olympic Games (Rogers, 2012), along with their announcement dates (day 0), which were found using LexisNexis and Factiva. Bloomberg Professional service was used to obtain the rest of the data in Table 1.
We also used LexisNexis and Factiva to identify any major events unrelated to the sponsorship announcement that occurred from three days before to three days after day 0 that might have affected prices or trading volumes of our sample firms. Because of confounding events, we eliminated one partner (British Airways) and one supporter (Arcelor Mittal) from consideration.
Bloomberg Professional and Datastream were used to obtain stock prices and trading volumes for the sponsors and benchmark companies, as well as sponsor and benchmark cash holdings, institutional ownership, and insider holdings levels. Two supporters (Cadbury and Deloitte) were removed from the sample because of the absence of trading data.
Stock Prices and Returns
Underlying all of our analyses is an assumption of market efficiency, that stock prices correctly and quickly incorporate all publicly-available information. To measure the relationship between announcements and sponsors' stock prices, we adopt basic event-study methodology (Brown & Warner, 1985) in calculating abnormal returns. The event in our case is the sponsorship announcement. Let t = 0 represent the announcement day, which, as Table 1 shows, is different for each sponsor.
Abnormal returns are computed as the difference between the realized and expected returns in the event window, around the announcement day. These abnormal returns are then attributed to the sponsorship announcements. Significance is determined using t-tests.
Various models of expected return have been used in event studies, the most popular ones being the market model, the constant expected returns model, and the capital asset pricing model. We used the capital asset pricing model in estimation of expected returns. For robustness purposes we also used the market model and the results were similar to the ones obtained by CAPM.
The expected return for firm i in period t is estimated using the Capital Asset Pricing Model as follows:
E([R.sup.f.sub.t]) = [R.sup.f.sub.t] [[beta].sup.i](E(R.sup.M.sub.t]) - [R.sup.f.sub.t])
where, E([R.sup.f.sub.t]) is the expected return for firm i in period t, [R.sup.f.sub.t] is the risk-free rate in period t, E([R.sup.M.sub.t]) is the expected return on the market portfolio in period t, and [[beta].sup.i] is the beta of firm i. Thus, expected (or normal) returns are estimated, by adjusting them for risk. (1)
As a proxy for the market portfolio, in estimating expected returns for each sponsor, a benchmark portfolio is derived from the value-weighted index of all firms on the stock exchange where the sponsor is listed.
When a sponsor is cross-listed (e.g., Adecco), we use its European exchange for the benchmark.
For firm i we measure actual returns, [R.sup.f.sub.t], and estimate abnormal (or unexpected) returns on day t as follows:
[AR.sup.f.sub.t] = [R.sup.f.sub.t] - E([R.sup.f.sub.t])
In addition, we measure abnormal returns for various time-intervals as:
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]
where, [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] is the actual return and [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] is the expected return for firm i on the time-interval [[t.sub.1], [t.sub.2]] (2) This allows us to aggregate the abnormal returns along time.
Average abnormal returns for all firms on day t and on the time-intervals [[t.sub.1], [t.sub.2]] are defined as:
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]
We use aggregation of abnormal returns through time and across sample firms, by defining the cumulative abnormal return (CAR). The CAR for firm i from ' is defined as:
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]
To aggregate the abnormal returns across sample firms, and through time, we assume that the abnormal returns of sample firms are not correlated. This assumption is easy to justify, given that the announcements are apart from each other and that event windows do not overlap.
The average CAR from [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] and is the average of all individual firms' CARs from day '.
The standard errors of the abnormal returns are calculated as:
[Se.sub.t] = [square root of 1/n - 1 [[summation].sup.n.sub.t=1] [([AR.sup.f.sub.t] - [[bar.AR].sub.t]).sup.2]/[square root of n]
and t-statistics of the abnormal returns are calculated as follows:
t - stat = [[bar.AR].sub.t]/[SE.sub.t]
This is exactly the same way we calculate the standard errors and i-statistics for CARs. The calculations for standard errors and i-statistics for volume and agency tests are computed using the same techniques as were used for abnormal returns. For more details about measuring and analyzing abnormal returns in event studies see Chapter 4 of Campbell, Lo, and MacKinlay, 1997.
Another test for market interest involves examining the number of shares traded around announcement dates. To perform this test, we compare the trading volume for each sponsor on day t to its average trading volume over the 31 days in the [ -15, +15 ] window. (3) Specifically, the ratio for day 0 is:
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]
where [Vol.sub.t] is the daily trading volume for day t. To determine whether there is evidence that information leaked prior to the announcement or whether the market was slow to respond, we take a similar approach for days -1 and +1.
To investigate the possibility of agency effects, we compare each sponsor's percentage of cash holdings, institutional holdings percentage, and insider holdings percentage to those of its peers. The peers for each of our sample firms were identified by Bloomberg Professional as those companies that were headquartered in the same country as the firm, and had the same 4-digit Standard Industrial Classification (SIC) code as the sponsoring firm in our sample. Thus, for each firm, we first tried to get domestic peer companies with the same 4-digit SIC codes. In some cases, there were one or fewer domestic peers, so we used regional peers for comparison purposes. For example, for Thomas Cook Group (TCG), the only domestic peer, having the same 4-digit SIC code, is TUI Travel Plc. But we also looked at regional (Western Europe) peers that had the same 4-digit SIC Codes as TCG. Tru and Kuoni Reisen Hldg. were two Germany-based companies satisfying the criteria. On the other extreme, there were companies in our sample (e.g., CSCO) that had 15 US peer companies with the 4-digit SIC Codes. Thus, we tried to collect data on peer companies by using domestic and regional peers that had the same SIC codes as our sample firms.
Cash Ratio Difference. To examine the sponsors' relative cash holdings, we compute a cash ratio for each sponsor and peer, defined as:
[CR.sup.i] = Cask holdings of firm i/market capitalization of firm i
where [CR.sup.i] is the Cash Ratio for firm i.
The sponsors' cash ratios are compared to the average of their peers' cash ratios using the following equation:
[CR.sup.i] = [CR.sup.i] - [summation].sub.i.sub.(i=1)] [CR.sup.j]/I
where [CR.sup.i] is the Cash Ratio Difference for sponsor i, [CR.sup.j] is the Cash Ratio for peer firm j, and I is the number of peer firms for firm i.
Institutional Holdings Difference. We follow an analogous strategy to compute an institutional holdings percentage for each sponsor and peer, as:
inst[H.sup.i] = number of firm i shores held by institutions/total number of firm i shares outstanding
where Inst[HD.sub.i] is the Institutional Holdings Percentage for firm i.
The sponsors' ratios are compared to the average of their peers' ratios using the following equation:
Inst[H.sup.i] = Inst[HD.sup.i] [H.sup.J] where InstHDi is the Institutional Holdings Percentage Difference for sponsor i, and Inst[H.sup.i] is the Institutional Holdings Percentage for peer firm j.
Insider Holdings Difference. Similar to the procedure used for the prior two variables, we compute an Insider Holdings Percentage for each sponsor and peer, as:
Inst[H.sup.i] = number of firm l shares held by insiders/Total number of firm i shares outstanding
where InsiHi is the Insider Holdings Percentage for firm i.
The sponsors' ratios are compared to the average of their peers' ratios using the following equation:
Inst[H.sup.i] = Inst[H.sup.i] - summation over (j=T)]
where is the Insider Holdings Percentage Difference for sponsor i, and is the Insider Holdings Percentage for peer firm j.
Panel A of Table 2 provides the descriptive statistics of our sample firms: the mean, standard deviation, minimum and maximum daily returns during the estimation period, the (-250, -16) window, and in the (-15, 15) event window. As can be seen from the reported numbers, the returns were more volatile in the event window, relative to estimation window. Not only the standard deviations of returns are higher in the event window compared to the estimation window, but also the min and max of the returns are more extreme in the event window.
We estimate the abnormal firm value changes and report them in Panel B of Table 2. We estimate the abnormal returns and cumulative abnormal returns in the event window and report the results in Table 3.
Panel A of Table 3 shows equally-weighted abnormal returns and cumulative abnormal returns for every day of the event window, aggregated across all sample firms, only British firms, only non-British firms, only Supporter firms, and only Partner firms. Panel B of Table 3 shows value-weighted abnormal returns and cumulative abnormal returns for every day of the event window, aggregated across all sample firms, only British firms, only non-British firms, only Supporter firms, and only Partner firms.
As it can be seen from Panels A and B of Table 3, the announcement day abnormal return was 0.75%, on average, for equally-weighted portfolio of all sample firms, and it is statistically significant. The average return for the value-weighted portfolio of all sample firms was even higher, 0.85%, and still statistically significant. The announcement-day abnormal return was 1.53%, on average, for equally-weighted portfolio of all British firms and it was statistically significant at the 99% level. However, the announcement-day abnormal return was only -0.03% for the equally-weighted portfolio of all non-British firms and it was statistically insignificant. The returns of value-weighted portfolios show the same pattern: the British firms, on average, yielded 1.11% statistically significant abnormal return on the announcement day, but the non-British firms, on average, yielded only 0.46% statistically insignificant abnormal return on the announcement day.
Also, the announcement day abnormal return was 0.92%, on average, for equally-weighted portfolio of Partner firms, and it was statistically significant at the 95% level. The announcement day abnormal return was only 0.49% for Supporter firms and statistically insignificant. The value-weighted average shows the same pattern: the Partner firms yielded 1.01% statistically significant abnormal return on the announcement day, but the Supporter firms on average yielded 0.54% insignificant abnormal return on the announcement day.
The day-to-day abnormal and cumulative abnormal returns of value-weighted portfolios are less variable than the ones of equally-weighted portfolios. A potential explanation for this observation is that the returns of small cap companies are more volatile. Thus, portfolio returns will be more volatile if smaller cap and large cap companies are assigned equal weights. In value-weighted portfolios, large-cap companies dominate, therefore day-to-day returns are more stable. For this reason, we rely more on value-weighted portfolio returns to explain our findings.
As Panel B of Table 3 shows, there are positive and highly significant cumulative abnormal returns (CAR) for the British cohort, if we use the event window. The British firms of our sample not only gain 1.11% abnormal return on the announcement day, but also their CAR is 0.68% over the event window and it is statistically significant at the 99% level. In terms of CAR, no other cohorts, and not the entire sample, exhibit statistically significant CAR in the event window. Partner firms gain about 1% abnormal return on the announcement day, but their CAR is 0.34% on the event window and it is only somewhat statistically significant. In terms of economic significance, we measure the abnormal dollar-value change for different cohort of our sample, and report the findings in Panel B of Table 2. The total abnormal increase in sponsoring firms' market value was $4.17 billion, averaging about $417 million per sponsoring firm. However, British firms got the lion's share of the gain: the total abnormal increase in sponsoring British firms' market value was $3.26 billion, averaging about $653 million per sponsoring British firm. The total abnormal increase in Partner firms' market value was $3.25 billion, averaging about $541 million per Partner firm. Thus, our results indicate that domestic firms and Partner firms benefit statistically and economically from their sponsoring activities.
Table 4 reports our findings regarding the association between sponsorship announcements and trading volume. As before, Panels A and B report the findings for partners and supporters, respectively. The day 0 cohort results for partners are insignificant, but volumes on day -1 and day 1 are significantly below average. Trading activity for supporters, as a whole, is insignificant for all days analyzed. However, the insignificant results for both cohorts on day 0 obscure the fact that every sponsor's day 0 trading volume varied significantly from its 31-day average. Two of the partners had significantly larger trading activity on day 0, while four had significantly lower volume. On the other hand, three of the four supporters had at least marginally higher trading volumes on day 0, while one had lower trading volume.
Panel C provides insight into the results from Panels A and B by reporting host country differences in trading volumes on day 0. Trading volume for British companies was highly significantly positive, while trading volume for non-British companies was highly significantly negative on day 0. Overall, the difference in trading levels between British and non-British sponsors was statistically significant.
Table 5 summarizes the findings related to our three measures associated with the potential for agency effects. The data in the second column show that all sponsors, except BMW, had a smaller cash ratio than the average of their industry peers. The average cash ratio for all sponsoring firms in our sample is less than the average cash ratio for their peers, although the difference is not statistically significant. As can be noticed, BMW and CSCO are two outliers that may significantly affect the findings, but due to the limitations of our data size, we keep them in the sample.
The differences in institutional holdings percentages vary widely among sponsors, and overall the average of the differences is insignificant. Similarly, the average of the differences for insider holdings percentages is also insignificant, although 9 of 10 sponsors had smaller insider holdings percentages than their peers. By looking at the three measures above, and comparing our sample firms with their peers, we see no statistically significant difference in the three measures that would indicate the existence of agency problems.
Hypotheses [H.sub.1A], [H.sub.2A], and [H.sub.3A] address the relationship between sponsorship announcements and stock prices. Based on the results reported in Table 3, we reject H1A, that there are no abnormal returns for partners on the announcement date. We find, instead, that there are significantly positive returns on day 0 for partners overall. (4)
In contrast, based on the results in Table 3, we cannot reject H2A, that there are no abnormal returns on the announcement date for supporters overall. (5)
In general, the results of our stock price analyses suggest that sponsorship announcements are important events, in that eight of 10 sponsors had significantly positive abnormal returns on day 0. Furthermore, market participants appear to view sponsorship announcements as shareholder wealth increasing events for partners, although the results are mixed for supporters. The positive results for partners compared to supporters may reflect the greater exposure afforded to partners.
Our findings contrast with those of previous papers that examined stock price effects on announcement dates. Farrell and Frame (1997), Samitas et al. (2008), Molchanov et al. (2010), and Tsiotsou (2011) all found no significant effects on day 0. Our unique findings may be due to the fact that ours is the only study to analyze separately the stock price effects based upon sponsorship level.
Based on the results in Table 3, we also reject H3A, that there is no difference in the stock price behavior of British and non-British sponsors. British firms had highly significant positive returns on day 0, and their CAR was positive and statistically significant for the event window. In contrast, the non-British cohort had insignificant abnormal returns on day 0. Since stocks for the British firms all are traded on the London Stock Exchange, the uniformly positive results for these firms could stem from British investors' appreciation for home company contributions to the Olympic effort and to the firms' associations with a generally popular event.
All British firms had positive significant returns on the announcement date, and the effect was not transitory (the CAR for the weighted-average portfolio of British firms was 0.68% over the event window and it was both statistically and economically significant) These findings deviate from the findings of Molchanov et al. (2010), who reported insignificant abnormal returns on day 0 for both domestic and international sponsors.
Hypotheses H^, H2B, and H3B address the relationship between sponsorship and trading volume. One of the contributions of the current study is that it is the first to analyze changes in trading volume for Olympic sponsors on day 0. Based on the cohort results in Table 4, we cannot reject H1B or H2B that trading volume for sponsors on day 0 was not different from its long-term average. Both the partner and supporter cohorts had insignificant variations on day 0. However, we do reject H3B, that there is no difference in volume between British and non-British sponsors. A closer look at the individual firm data suggests that the highly significant results for British firms may be due to the fact that Lloyds' day 0 volume was more than double its 31day average.
Hypotheses H4A, H4B, and H4C address the potential for agency issues. We cannot reject any of the agency issue hypotheses. On average, sponsors had insignificantly less cash holdings than their peers (H4A), insignificantly more institutional holdings percentages (H4B), and insignificantly less insider holdings percentages (H4C). Thus, we do not find evidence that the conditions often linked to the potential for agency conflicts (high excess cash, low institutional holdings, or unusually high or low insider holdings) existed for the 2012 Olympic sponsors.
In addition, the relationship between announcements and stock prices was neutral or significantly positive for nine of the 10 sponsors. Thus, investors do not appear to have considered the sponsorships to be low-return endeavors. This finding contrasts with Farrell and Frame (1997), who found significantly negative abnormal returns around Olympic sponsorship announcement dates, but noted that the results were mitigated for companies with higher levels of institutional holdings. Our findings also differ somewhat from that of Pruitt et al. (2004), who found general evidence of agency conflicts in the NASCAR sponsorship study.
This study has some limitations, the most notable and unavoidable of which is sample size. There were only 10 firms with usable data. Given that multiple stock markets in different regions are involved, a question also arises regarding when an investor could first trade on a sponsorship announcement. However, there is no evidence that trading related to the announcement occurred on days other than the ones we identified.
This is the first paper to analyze separately capital market behavior on announcement dates based upon different sponsorship levels. We find that London 2012 Olympic sponsorships are associated with increased share values for major contributors and for host-country companies. More broadly, most sponsors had significant positive abnormal returns on the announcement days, indicating that investors deem the benefits of sponsorships to be at least equal to the costs. In contrast to other research, we do not find evidence that agency issues influenced the decision to become an Olympic sponsor. The British firms in our sample not only gain a 1.11% abnormal return on the announcement day, but also their CAR is 0.68% over the event window and it is statistically significant at the 99% level.
Partner firms gain approximately 1% abnormal return during the announcement day, but their CAR is 0.34% on the event window and is only somewhat statistically significant.
The total abnormal increase in sponsoring firms' market value was $4.17 billion, averaging about $417 million per sponsoring firm. However, British firms got the lion's share of the gain: the total abnormal increase in sponsoring British firms' market value was $3.26 billion, averaging about $653 million per sponsoring British firm. Also the total abnormal increase in Partner firms' market value was $3.25 billion, averaging about $541 million per Partner firm. Thus, our results indicate that domestic firms and Partner firms benefit statistically and economically significantly from their sponsoring activities.
We wish to recognize Marc Vinyard for his assistance with data collection, the participants in the 5th International Sports Business Symposium, two anonymous reviewers, and the editor for their constructive comments and suggestions related to the paper.
Aggarwal, R., Erel, I., Ferreira, M., & Matos, P. (2011). Does governance travel around the world?
Evidence from institutional investors. Journal of Financial Economics, 100, 154-181.
Agrawal, A., & Knoeber, C. R. (1996). Firm performance and mechanisms to control agency problems between managers and shareholders. Journal of Financial and Quantitative Analysis, 31, 377-397.
Barber, B., & Lyon, J. (1996). Detecting abnormal operating performance: The empirical power and specification of test statistics. Journal of Financial Economics, 41, 359-399.
Beaver, W., (1968). The information content of annual earnings announcements. Journal of Accounting Research Supplement 6, 67-92.
Brown, S. J., & Warner, J. B. (1985). Using daily stock returns: The case of event studies. Journal of Financial Economics, 14, 3-31.
Campbell, J. Y., Lo, A.W., & MacKinlay, A. C. (1997). The econometrics of financial markets. Princeton, NJ: Princeton University Press.
Campbell, C., & Wasley, C. (1996). Measuring abnormal trading volume for samples of NYSE/ASE and NASDAQ securities using parametric and nonparametric test statistics. Review of Quantitative Finance and Accounting.
Cornwell, T. B., Pruitt, S. W., & Clark, J. M. (2005). The relationship between major-league Sport official sponsorship announcements and the stock prices of sponsoring firms. Journal of the Academy of Marketing Science, 33, 401-412.
Cornwell, T. B., Pruitt, S. W., & Van Ness, R. (2001). The value of winning in motor sports: Sponsorship-linked marketing. Journal of Advertising Research, 41, 17-31.
Dolley, J. (1933). Characteristics and procedure of common stock split-ups. Harvard Business Review, 11, 316-326.
Eisenhardt, K. M. (1989). Agency theory: An assessment and review. Academy of Management, 14, 57-74.
Fama, E. F., Fisher, L., Jensen, M. C., & Roll, R. (1969). The adjustment of stock prices to new information. International Economic Review, 10, 1-21.
Farrell, K. A., & Frame, W. S. (1997). The value of Olympic sponsorships: Who is capturing the gold?. Journal of Market-Focused Management, 2, 171-182.
Floros, C. (2010). The impact of the Athens Olympic Games on the Athens Stock Exchange. Journal of Economic Studies, 37, 647-657.
Jensen, M. C., & Meckling, W. H. (1976). Theory of the firm: Managerial behavior, agency costs and ownership structure. Journal of Financial Economics, 3, 305-360.
Jensen, M. C. (1986). Agency costs of free cash flow, corporate finance, and takeovers. American Economic Review, 76, 323-329.
MacKinlay, A.C. (1997). Event studies in economics and finance. Journal of Economic Literature, 35, 13-39.
Mishra, D. P., Bobinski, G. S., & Bhabra, H. S. (1997). Assessing the economic worth of corporate event sponsorships: A stock market perspective. Journal of Market Focused Management, 2, 149-169.
Miyazaki, A. D., & Morgan, A. G. (2001). Assessing market value of event sponsoring: Corporate Olympic sponsorships. Journal of Advertising Research, 41, 9-15.
Molchanov, A., Stork, P. A., & Zeng, C. (2010). The 2008 Beijing Olympic sponsorships: Value for money?. Retrieved from http://ssrn.com/abstract=1649132
Pruitt, S. W., Cornwell, T. B., & Clark, J. M. (2004). The NASCAR phenomenon: Auto racing sponsorships and shareholder wealth. Journal of Advertising Research, 44, 281-296.
Reiser, M., Breuer, C., & Wicker, P. (2012). The sponsorship effect: Do sport sponsorship announcements impact the firm value of sponsoring firms? International Journal of Sport Finance, 7, 232-248.
Rogers, S. (2012, July 19). London 2012 Olympic sponsors list: Who are they and what have they paid? Retrieved from http://www.guardian.co.uk/sport/datablog/2012/jul/19/london-2012olympic-sponsors-list#data
Samitas, A., Kenourgios, D., & Zounis, P. (2008). Athens' Olympic Games 2004 impact on sponsors' stock returns. Applied Financial Economics, 18, 1569-1580.
Sanchez-Ballesta, J. P., & Garcia-Meca, E. (2007). A meta-analytic vision of the effect of ownership structure on firm performance. Corporate Governance: An International Review, 15, 779-893.
Spais, G. S., & Filis, G. N. (2006). The Athens 2004 Olympic Games: An event study. Journal of Integrated Marketing Communications, 6, 52-61.
Tsiotsou, R. (2011). A stakeholder approach to international and national sport sponsorship. Journal of Business and Industrial Marketing, 26, 557-565.
Yelkur, R., Tomkovick, C., & Pennington, J. (2012). The alchemy of Olympics advertising and sponsorship: Turning the Games into gold. Journal of Management Policy and Practices, 13, 34-45.
(1) It is well-known that in long-horizon tests, appropriate adjustment for risk is critical in accurately estimating abnormal returns. However, in short-horizon tests, such as ours, this is not as important, as Brown and Warner (1985) claim: "...simple risk-adjustment approaches to conducting short-window event studies are quite effective in detecting abnormal performance"
(2) As an alternative, we also computed abnormal returns as:. The results were qualitatively similar; thus, we do not report them separately.
(3) We also studied a different estimation window for average trading volume: [-250,-2]. Our results were very similar to the ones reported in the paper that is using [-15,15] windows.
(4) Although we do not report the numbers here, the abnormal returns were significantly positive for five of the six partners on day 0.
(5) We observed, however, that three of the supporters had significant abnormal returns on day 0--two positive, and one negative.
Dean V. Baim , Levon Goukasian , and Marilyn B. Misch 
 Pepperdine University
Dean V. Baim, PhD, is a professor of economics and finance and the chair of the Business Administration Division, Seaver College. His research interests include stadium finance and Olympic studies.
Levon Goukasian, PhD, is a professor of finance in the Business Administration Division, Seaver College. His research interests include asset pricing, investment, portfolio management, and risk management.
Marilyn B. Misch, PhD, is a professor of accounting in the Business Administration Division, Seaver College. Her research interests include accounting pedagogy and curriculum development, financial accounting, and Olympic studies.
Table 1. Panel A: London 2012 Partner Firms: Descriptive Data Company Ticker Announce Sector/ Date Industry Adidas ADS 9/17/2007 Cons. Disc./ Pers. Goods BMW BMW 11/18/2009 Cons. Disc./ Auto & Parts British Airways IAG 2/5/2008 Indust./Travel & Leisure British Petroleum BP/ 7/3/2008 Energy/Oil & Gas Producers British Telecom BT/A 3/5/2008 Telecomm Serv./ Fixed Line Tel. EDF EDF 7/11/2007 Utilities/ Electricity Lloyds TSB LLOY 3/14/2007 Financials/ Banks Company Country/ Market Cap Beta Benchmark Adidas GER/DAX $9,131 0.83 BMW GER/DAX $21,609 1.08 British Airways UK/UKX $3,378 1.39 British Petroleum UK/UKX $104,730 1.00 British Telecom UK/UKX $17,410 1.03 EDF UK/UKX $138,795 0.95 Lloyds TSB UK/UKX $30,418 1.69 Table 1. Panel B: London 2012 Supporter Firms: Descriptive Data Company Ticker Announce Sector/ Date Industry Adecco ADEN 1/14/2009 Industrials/ Support Serv. Arcelor Mittal MT 5/25/2010 Materials/Ind. Metals & Mining Cisco CSCO 7/10/2009 IT/Tech Hard. & Equip. Thomas Cook TCG 10/21/2009 Cons. Disc./ Partners Travel & Leisure UPS UPS 9/30/2009 Industrials/ Industrial Transport. Cadbury KFT 10/20/2008 Cons. Staples Deloitte NA 12/04/2007 Private Company Country/ Market Beta Benchmark Cap Adecco SW/SMI $6,461 1.42 Arcelor Mittal USA/AEX $36,135 1.33 Cisco USA/SPX $105,783 1.06 Thomas Cook UK/UKX $1,927 1.57 Partners UPS USA/SPX $56,157 0.97 Cadbury UK/NA NA NA Deloitte USA/NA NA NA Table 2. Panel A: Descriptive Statistics for sample firms in the estimation window, (-250, -16) and in the event window: (-15, +15). The reported numbers are daily returns. Estimation Window ADEN csco TCG UPS Average 0.39% 0.45% -0.33% 0.18% St. Dev. 2.82% 1.65% 2.42% 1.39% Min -4.51% -2.70% -5.05% -2.25% Max 7.65% 5.77% 7.41% 4.44% Event Window ADEN CSCO TCG UPS Average -0.15% -0.04% 0.15% -0.05% St. Dev. 2.60% 3.39% 3.99% 2.83% Min -7.76% -10.63% -23.26% -8.50% Max 9.22% 13.80% 13.48% 9.36% Estimation Window BMW ADS BP LLOY Average -0.25% 0.31% -0.45% -0.26% St. Dev. 2.20% 0.98% 1.69% 1.48% Min -6.26% -1.71% -3.41% -4.35% Max 3.36% 2.52% 3.72% 2.62% Event Window BMW ADS BP LLOY Average 0.22% 0.06% 0.02% 0.05% St. Dev. 3.31% 1.34% 1.63% 1.72% Min -13.19% -7.15% -6.32% -3.96% Max 14.84% 7.30% 5.96% 3.94% Estimation Window BT EDF Average Average -0.19% 0.18% -0.05% St. Dev. 1.99% 1.44% 2.16% Min -3.18% -3.37% -4.15% Max 4.04% 4.84% 5.20% Event Window BT EDF Average Average -0.11% 0.27% 0.03% St. Dev. 1.68% 1.49% 2.45% Min -9.80% -3.10% -9.05% Max 5.21% 5.90% 8.80% Table 2. Panel B: Abnormal firms value change on the announcement- days (all numbers in $ millions). Total Change per Average Change per group group ALL Firms $4,170.34 $ 417.03 BRITISH Firms $3,263.33 $ 652.67 NON-British Firms $ 907.01 $ 181.40 Partner Firms $3,250.46 $ 541.74 Supporter Firms $ 919.88 $ 229.97 Table 3: Abnormal Returns (AR) and Cumulative Abnormal Returns (CAR) in the event window. Panel A: Equally-Weighted Returns. ALL FIRMS BRITISH Event Day AR CAR AR CAR -15 -0.26% -0.26% -0.75% -0.75% -14 0.66% 0.40% 0.83% 0.08% -13 0.10% 0.50% 0.30% 0.38% -12 1.27% 1.77% 0.54% 0.92% -11 -1.37% 0.40% -1.15% -0.23% -10 -0.31% 0.09% -1.08% -1.30% -9 0.25% 0.33% 0.90% -0.40% -8 -0.28% 0.05% -0.10% -0.50% -7 -0.06% -0.01% 0.49% -0.01% -6 -0.39% -0.40% -0.33% -0.33% -5 -1.50% -1.89% -2.21% -2.54% -4 0.17% -1.72% -0.48% -3.02% -3 -0.42% -2.15% 0.05% -2.97% -2 -0.01% -2.16% -1.04% -4.01% -1 -1.13% -3.28% -1.01% -5.02% 0 0.75% -2.54% 1.53% -3.49% 1 -0.21% -2.74% -0.82% -4.31% 2 0.62% -2.12% 0.86% -3.45% 3 -0.32% -2.44% -0.72% -4.17% 4 -0.07% -2.51% 0.14% -4.04% 5 -0.25% -2.77% -0.45% -4.48% 6 -0.05% -2.81% 0.31% -4.18% 7 0.80% -2.01% 0.44% -3.74% 8 -0.31% -2.32% -0.36% -4.10% 9 0.73% -1.60% 1.23% -2.87% 10 -0.12% -1.71% -1.00% -3.88% 11 -0.06% -1.78% 0.17% -3.71% 12 0.94% -0.84% 1.05% -2.66% 13 -0.58% -1.42% -0.53% -3.19% 14 0.50% -0.92% 0.58% -2.61% 15 0.31% -0.61% -0.85% -3.46% t-stat Day-0 1.77 * 4.16 *** for AR t-stat (-15,15) (0.24) (1.13) for CAR Panel B: Value-Weighted returns. The weights are computed by using the market capitalizations of sample firms on the event-days. ALL FIRMS BRITISH Event Day AR CAR AR CAR -15 0.80% 0.80% 1.51% 1.51% -14 0.26% 1.06% 0.44% 1.95% -13 0.50% 1.55% 1.28% 3.23% -12 0.73% 2.29% -0.61% 2.63% -11 -0.78% 1.51% -0.69% 1.94% -10 0.34% 1.85% -0.17% 1.77% -9 -0.44% 1.41% 0.51% 2.28% -8 0.51% 1.92% 0.82% 3.11% -7 -0.26% 1.66% 0.59% 3.69% -6 -0.11% 1.55% -0.45% 3.25% -5 -1.18% 0.37% -1.19% 2.05% -4 0.47% 0.84% -0.47% 1.58% -3 -0.06% 0.78% 0.83% 2.41% -2 -1.01% -0.23% -2.67% -0.26% -1 -0.47% -0.70% -0.06% -0.32% 0 0.85% 0.15% 1.11% 0.79% 1 -0.16% -0.01% -0.38% 0.41% 2 0.57% 0.56% 1.19% 1.60% 3 -0.07% 0.49% -0.37% 1.23% 4 -0.17% 0.33% 0.26% 1.49% 5 -0.97% -0.65% -1.67% -0.18% 6 -0.15% -0.80% 0.12% -0.06% 7 0.70% -0.11% 0.53% 0.47% 8 -0.49% -0.60% -0.18% 0.29% 9 0.54% -0.05% 0.16% 0.45% 10 -0.77% -0.82% -0.89% -0.44% 11 -0.36% -1.18% -0.32% -0.75% 12 0.71% -0.47% 1.15% 0.40% 13 0.37% -0.10% 0.19% 0.59% 14 0.95% 0.84% 1.21% 1.80% 15 -0.81% 0.03% -1.12% 0.68% t-stat Day-0 2.13 ** 5.01 *** for AR t-stat (-15,15) 0.94 3.05 *** for CAR NON-BRITISH PARTNERS Event Day AR CAR AR CAR -15 0.23% 0.23% -0.60% -0.60% -14 0.48% 0.71% 1.25% 0.64% -13 -0.10% 0.61% -0.29% 0.35% -12 2.01% 2.62% 0.54% 0.89% -11 -1.60% 1.03% -1.84% -0.95% -10 0.46% 1.48% 0.09% -0.86% -9 -0.41% 1.07% 0.88% 0.02% -8 -0.47% 0.60% 0.06% 0.08% -7 -0.61% -0.01% 0.82% 0.90% -6 -0.45% -0.46% -1.11% -0.21% -5 -0.79% -1.24% -1.21% -1.42% -4 0.82% -0.42% -0.07% -1.49% -3 -0.90% -1.32% 0.53% -0.95% -2 1.02% -0.30% -0.95% -1.91% -1 -1.25% -1.55% -1.06% -2.97% 0 -0.03% -1.58% 0.92% -2.05% 1 0.41% -1.18% -0.26% -2.31% 2 0.38% -0.79% 0.58% -1.73% 3 0.09% -0.70% -0.19% -1.92% 4 -0.29% -0.99% 0.20% -1.72% 5 -0.06% -1.05% -0.88% -2.60% 6 -0.40% -1.45% -0.44% -3.04% 7 1.17% -0.28% 1.51% -1.53% 8 -0.26% -0.54% -1.33% -2.86% 9 0.22% -0.32% 1.71% -1.15% 10 0.76% 0.45% -1.75% -2.90% 11 -0.30% 0.15% 0.29% -2.60% 12 0.82% 0.97% 1.32% -1.29% 13 -0.63% 0.34% -0.76% -2.04% 14 0.43% 0.77% 0.61% -1.43% 15 1.47% 2.24% -0.66% -2.09% t-stat Day-0 (0.33) 2.74 ** for AR t-stat (-15,15) (0.09) (0.37) for CAR Panel B: Value-Weighted returns. The weights are computed by using the market capitalizations of sample firms on the event-days. NON-BRITISH PARTNERS Event Day AR CAR AR CAR -15 -0.25% -0.25% 1.16% 1.16% -14 -0.01% -0.26% 0.72% 1.88% -13 -0.66% -0.92% 0.94% 2.81% -12 2.71% 1.79% -0.30% 2.52% -11 -0.92% 0.87% -1.00% 1.51% -10 1.11% 1.97% 0.05% 1.56% -9 -1.84% 0.13% 0.55% 2.11% -8 0.04% 0.17% 0.72% 2.83% -7 -1.50% -1.33% 0.56% 3.39% -6 0.39% -0.94% -0.73% 2.66% -5 -1.16% -2.09% -0.98% 1.68% -4 1.85% -0.24% -0.40% 1.28% -3 -1.38% -1.62% 0.82% 2.10% -2 1.44% -0.18% -2.29% -0.19% -1 -1.08% -1.26% -0.29% -0.48% 0 0.46% -0.80% 1.01% 0.53% 1 0.18% -0.62% -0.39% 0.14% 2 -0.34% -0.97% 1.08% 1.22% 3 0.37% -0.60% -0.30% 0.92% 4 -0.79% -1.39% 0.20% 1.12% 5 0.06% -1.33% -1.64% -0.52% 6 -0.56% -1.89% -0.05% -0.57% 7 0.94% -0.96% 0.93% 0.37% 8 -0.94% -1.90% -0.47% -0.11% 9 1.11% -0.79% 0.37% 0.26% 10 -0.59% -1.38% -1.07% -0.81% 11 -0.43% -1.81% -0.24% -1.04% 12 0.05% -1.75% 1.23% 0.19% 13 0.63% -1.12% 0.06% 0.26% 14 0.56% -0.56% 1.08% 1.34% 15 -0.35% -0.91% -0.99% 0.34% t-stat Day-0 1.18 2.27 ** for AR t-stat (-15,15) (1.12) 1.41 for CAR SUPPORTERS Event Day AR CAR -15 0.25% 0.25% -14 -0.23% 0.03% -13 0.69% 0.71% -12 2.38% 3.09% -11 -0.67% 2.42% -10 -0.91% 1.51% -9 -0.71% 0.80% -8 -0.80% 0.01% -7 -1.38% -1.37% -6 0.69% -0.68% -5 -1.92% -2.60% -4 0.52% -2.07% -3 -1.86% -3.93% -2 1.40% -2.53% -1 -1.23% -3.76% 0 0.49% -3.26% 1 -0.13% -3.40% 2 0.69% -2.71% 3 -0.51% -3.22% 4 -0.49% -3.70% 5 0.69% -3.01% 6 0.54% -2.47% 7 -0.25% -2.73% 8 1.22% -1.51% 9 -0.76% -2.26% 10 2.32% 0.06% 11 -0.60% -0.54% 12 0.37% -0.17% 13 -0.32% -0.49% 14 0.34% -0.15% 15 1.76% 1.62% t-stat Day-0 0.95 for AR t-stat (-15,15) (1.50) for CAR Panel B: Value-Weighted returns. The weights are computed by using the market capitalizations of sample firms on the event-days. SUPPORTERS Event Day AR CAR -15 0.11% 0.11% -14 -0.60% -0.49% -13 -0.34% -0.83% -12 2.68% 1.85% -11 -0.35% 1.50% -10 0.90% 2.40% -9 -2.31% 0.10% -8 0.11% 0.20% -7 -1.81% -1.61% -6 1.07% -0.53% -5 -1.56% -2.09% -4 2.11% 0.02% -3 -1.73% -1.72% -2 1.42% -0.30% -1 -0.82% -1.11% 0 0.54% -0.57% 1 0.28% -0.29% 2 -0.39% -0.68% 3 0.37% -0.31% 4 -0.86% -1.17% 5 0.28% -0.89% 6 -0.35% -1.24% 7 0.24% -1.00% 8 -0.52% -1.52% 9 0.88% -0.63% 10 -0.20% -0.84% 11 -0.60% -1.44% 12 -0.28% -1.72% 13 0.94% -0.78% 14 0.69% -0.09% 15 -0.46% -0.55% t-stat Day-0 (0.83) for AR t-stat (-15,15) (0.88) for CAR Note: *** p <0.01; ** p <0.05 ; * p<0.1 Table 4. Ratio of Day t=0 trading volume to the average trading volume over the event window. Panel A: Partner Firms Day ADS BMW BP BT -1 0.57 0.46 1.11 0.82 0 0.71 0.54 1.07 0.73 1 0.75 0.74 0.58 0.59 t-stat day 0 -2.70 ** -4.03 *** 1.99 ** -8.18 *** Day EDF LLOY Avg. t-stat -1 0.98 0.73 0.78 -2.39 ** 0 0.62 2.06 0.95 -0.21 1 0.84 1.05 0.76 -3.68 *** t-stat day 0 -4.21 *** 14.1 *** Panel B: Supporter Firms Day ADEN CSCO TCG -1 1.00 1.02 1.15 0 1.12 0.90 1.11 1 0.97 1.20 0.66 t-stat day 0 1.93 * -2.21 ** 2.13 ** Day UPS Avg. t-stat -1 0.71 0.97 -0.38 0 1.18 1.08 1.39 1 1.14 0.99 -0.08 t-stat day 0 2.55 ** Panel C: British vs. Non-British Firms Day British Non- Difference British -1 0.96 0.75 0.21 0 1.12 0.89 0.23 1 0.74 0.96 -0.21 t-stat day 0 4.23 *** -3.00 *** 5.45 *** Note: *** p <0.01; ** p <0.05; * p <0.1 Table 5. Cash ratios, Institutional Holdings percentages, and Insider Holdings percentages for all sample firms. Company Cash Ratio Institutional Insider Name Difference Holdings Holdings Difference Difference ADEN -6.53% -19.36% -3.48% ADS -2.93% 11.30% -13.93% BMW 16.30% -11.43% 20.00% BP -2.50% 25.39% -9.14% BT -1.55% -4.79% -18.69% CSCO -14.89% -6.85% -11.71% EDF -1.93% 13.95% -0.10% LLOY -3.46% -2.31% -0.05% TCG -7.51% -0.07% -0.22% UPS -3.97% -4.11% -2.79% Average -2.90% 0.17% -4.01% t-stat -1.115 0.039 -1.129 Note: *** p<0.01; ** p<0.05; * p<0.1
|Printer friendly Cite/link Email Feedback|
|Author:||Baim, Dean V.; Goukasian, Levon; Misch, Marilyn B.|
|Publication:||International Journal of Sport Finance|
|Date:||May 1, 2015|
|Previous Article:||Testing profitability in the NBA season wins total betting market.|
|Next Article:||A theoretical comparison of the economic impact of large and small events.|