The Market Value of Sport Sponsorship

kudo_m.pdf

Table 5-11. Mean abnormal returns (MAR) for the full 81 event date in PGA, LPGA, and NASCAR
Mean
Abnormal Positive: Patell Generalized
Day N Return Negative Z Sign Z
-3 81 0.19% 39:42 0.262 -0.365
-2 81 0.04% 48:33) 1.783* 1.635$
-1 81 0.01% 35:46 0.460 -1.254
0 81 0.17% 43:38 0.589 0.524
+1 81 0.07% 40:41 0.259 -0.143
+2 81 0.08% 41:40 0.693 0.079
+3 81 -0.08% 36:45 -1.139 -1.032
The symbols $,*,**, and *** denote statistical significance at the 0.10,
0.05, 0.01 and 0.001 levels, respectively, using a generic one-tail test.
The symbols (,< or ),> etc. correspond to $,* and show the direction
and generic one-tail significance of the generalized sign test.


Table 5-12. Mean cumulative abnormal return (MCAR) for the full 81 event date in PGA, LPGA, and NASCAR
Mean
Cumulative Precision
Abnormal Weighted Positive: Patell Generalized
Days N Return CAAR Negative Z Sign Z
(0,+1) 81 0.24% 0.17% 40:41 0.600 -0.143
(0,+2) 81 0.32% 0.31% 42:39 0.890 0.302
(0,+3) 81 0.24% 0.08% 40:41 0.201 -0.143
The symbols $,*,**, and *** denote statistical significance at the 0.10, 0.05, 0.01 and 0.001
levels, respectively, using a generic one-tail test. The symbols (,< or ),> etc. correspond
to $,* and show the direction and generic one-tail significance of the generalized sign test.









not allow the researcher to generate the results obtained from this research and

conclude the results represent all title sponsorship in the PGA, LPGA, and NASCAR.

Finally, the determination of the level of congruence between title sponsor and

sponsee can lead to the existence of bias in the results. In this study, the image

congruence level was determined by the researcher after a careful discussion with an

expert. In the future research, the level of image congruence should be determined by

an extensive survey of the general public and it would enhance the validity of perceived

image congruence level, and the conclusion that image congruence level between title

sponsor and sponsored event plays an important role in stock price changes. Also, in

the future research, other sport entities that have title sponsorships such as professional

tennis tournament (ATP & VVTP) and college football (NCAA) should be chosen for the

topic. Title sponsorship in other countries' sport and markets would be an interesting

topic as well.

Repetition of this study in the future is also meaningful. Title sponsorship fee is

generally decided by the supply and demand relationship between sport entity and

potential corporate sponsors. Thus, when the equilibrium occurs in the title sponsorship

supply and demand curve, it is the time that the title sponsorship fee equals the benefits

the title sponsor receive. In this respect, any stock price change significance recorded

in this study show discrepancy between demand and supply in title sponsorship, and

the gap would be fixed based on the market principle in due course. Therefore, once the

title sponsorship fee and its trend have been changed by the sport entity, the results in

event study analysis would be different from those in this study. The results gained in

this study only reflect the current situation between title sponsors and their stock price









suggests that the event is harmful to a company's future value. If the results brought no

cumulative abnormal return, it would be implying the event is seen neither a positive nor

a negative for the future value of the company.

Sponsorship Effectiveness Using Event Study Analysis

Until now, several studies have conducted event study analysis in sport

sponsorship context. For example, Farell and Frame (1997), Miyazaki and Morgan

(2001), Oztur et al. (2004), Samitas, Kenourgios, and Zounis (2008), and Spais and Filis

(2006) examined the stock price impacts of sponsoring the Olympic Games. Also,

Cornwell, Pruitt, and Van Ness (2001) and Pruitt et al. (2004) studied the NASCAR

sponsorship effectiveness. Likewise, Becker-Olsen (2003) and Clark et al. (2002) did

about sponsorship of the stadium naming rights. The study of Cornwell, Pruitt, and Clark

(2005) analyzed effects of the official product sponsorship in major league sports

including the MLB, NBA, NFL, NHL, and PGA. Furthermore, Spais and Filis (2008)

researched sponsorship effects of an Italian soccer team. As well, Russell, Mahar, and

Drewniak (2005) studied stock market reactions to the publicity of athletic endorsers.

And finally, Caldron-Martinez et al. (2005) explored the difference between commercial

sponsorship and philanthropic sponsorship by employing event study analysis. None of

them to date, however, has focused on effectiveness of title sponsorship in professional

sport by employing event study analysis.











Table 4-4. NASCAR Nationwide series title sponsorship sample


Sport
NASCAR (Nationwide)
NASCAR (Nationwide)
NASCAR (Nationwide)
NASCAR (Nationwide)
NASCAR (Nationwide)
NASCAR (Nationwide)
NASCAR (Nationwide)
NASCAR (Nationwide)
NASCAR (Nationwide)
NASCAR (Nationwide)
NASCAR (Nationwide)
NASCAR (Nationwide)
NASCAR (Nationwide)
NASCAR (Nationwide)
NASCAR (Nationwide)
NASCAR (Nationwide)
NASCAR (Nationwide)
NASCAR (Nationwide)
NASCAR (Nationwide)
NASCAR (Nationwide)
NASCAR (Nationwide)
NASCAR (Nationwide)
NASCAR (Nationwide)
NASCAR (Nationwide)
NASCAR (Nationwide)
NASCAR (Nationwide)
NASCAR (Nationwide)
NASCAR (Nationwide)


Company Name
KELLOGG
CVS
SARA LEE
KROGER COMPANY
KRAFT FOODS
PEPSI
UNITED ONLINE
HERSHEY FOODS
KRAFT FOODS
ITT INDUSTRIES
WINN DIXIE STORES
NEWELL RUBBERMAID
GLAXOSMITHKLINE
VIACOM
DOLLAR GENERAL
SBC COMMUNICATIONS
USG
YELLOW ROADWAY
DOMINOS PIZZA
GLAXOSMITHKLINE
GLAXOSMITHKLINE
GENUINE PARTS
BANCO SANTANDER
COCA COLA
COCA COLA
DAIMLERCHRYSLER
UNILEVER
KROGER COMPANY


Title Sponsorship Name
Cheez-lt 250
CVS Pharmacy 200 Presented by Bayer
Sam's Club Presents Hill Bros. Coffee 300
Kroger 300 Presented by Oreo
Kroger 300 Presented by Oreo
Tropicana Twister 300
NetZero 250
Koolerz 300
Meijer 300 Presented by Oreo
Goulds Pumps ITT Industries 200
Winn-Dixie 200
Sharpie Professional 250
Goody's Headache Powder 200
SpongeBob SquarePants Movie 300
Dollar General 300
SBC 250
USG Durock 300
United Way 300 Presented by Yellow Transportation
Domino's Pizza 250
Nicorette 300
Goody's 250
NAPA Auto Parts 200
RoadLoans.com 200
Winn-Dixie 250 Powered by Coca-Cola
Dollar General 300 Powered by Coca-Cola
Missouri-Illinois Dodge Dealers 250
Lipton Tea 250
Kroger On Track for the Cure 250


Announcement Date
11/20/2000
3/31/2001
6/13/2001
4/16/2002
4/16/2002
4/18/2002
6/17/2002
2/12/2003
3/17/2003
4/16/2003
8/23/2003
11/13/2003
2/11/2004
5/27/2004
1/21/2005
4/7/2005
5/16/2005
5/20/2005
8/9/2005
1/25/2006
5/14/2006
4/16/2007
6/1/2007
12/7/2007
12/7/2007
1/25/2008
4/17/2008
8/27/2008


Event Date
3/26/2001
5/14/2001
7/16/2001
6/17/2002
6/17/2002
7/15/2002
7/29/2002
2/18/2003
6/16/2003
5/19/2003
9/2/2003
3/29/2004
2/23/2004
10/15/2004
10/14/2005
6/27/2005
7/11/2005
10/11/2005
8/22/2005
3/20/2006
7/24/2006
8/6/2007
9/24/2007
7/7/2008
7/11/2008
7/21/2008
5/2/2008
10/27/2008









in the estimation period. The null hypothesis under the Generalized Sign Z test is that

the fraction of positive returns is the same as that in the estimation period.

Table 5-2 shows the results of the mean cumulative abnormal return (MCAR)

and their associated statistics tests over three different event windows surrounding the

title sponsorship. The event windows include t = (0, 1), (0, 2), and (0, 3). Becker-Olsen

(2003) noted that several event windows need to be assessed in order to obtain the full

effects of the event.

According to the results obtained, there has not been detected any abnormal

stock market returns around the sponsorship announcement dates. This means that

stock market has seen title sponsorship announcements in the PGA, LPGA, and

NASCAR as neither positive nor negative events (at least statistically significant level)

for sponsoring companies. Therefore, H, was not confirmed. In event study analysis,

however, it must be emphasized that no significant reaction on stock price does not

necessarily mean that the company has not received enough benefits or suffered a net

loss by the title sponsorship. Rather, it should mean that the company is receiving

financial benefits from the event exactly same amount as the cost they paid for the

event. That is, stock market has seen title sponsors in the PGA, LGPA, and NASCAR

are likely to receive some benefits that are financially equivalent to the title sponsorship

fee investments in the future due to the new relationship with the sporting event. In

other words, market assumes title sponsorship fees are set for fair prices.

The study of Cornwell et al. (2005a) demonstrated that stock price changes

following the sponsorship announcements vary among the different sports entities (i.e,

NFL, MLB, NHL, NBA, and PGA). Also, past research only focused on the NASCAR









seen NASCAR sponsors (Cornwell et al., 2001; Pruitt et al., 2004). This study

demonstrated it from a perspective of title sponsor's stock price increases and added

additional evidence that NASCAR can be a special marketing platform for businesses.

Also, in this study, it was demonstrated that the difference of NASCAR event

competition level (i.e., Sprint series or Nationwide series) plays an important role in the

stock price changes, and that more prestigious event's (i.e., Sprint series) title sponsors

are likely to be favored by market. In the case of golf event, title sponsors for the PGA

experienced significant negative stock market returns from both announcement date

and following title sponsorship event date. These results are consistent with opinions by

a sport critique (Harig, 2009) and one of the PGA's title sponsor (Mell, 2009) noting that

title sponsorship fee in the PGA are just too expensive. Annual 5% increase of title

sponsorship fees in PGA may be one of the reasons that caused this negative

significance. Or it may be just indicating a conflict of title sponsorship objectives

between title sponsor who regards the PGA tour event as more hospitality opportunity to

strengthen relationship with key customers and stock market who only seeks tangible

and direct financial returns from the company's investment. Interestingly, as opposed to

the result of PGA, the stock prices of title sponsors for LPGA reacted significantly

positive following the new title sponsorship announcement. This finding implies that title

sponsors for LGPA are seen by stock investors to be likely to increase their company's

performance including sales thanks to the title sponsorship, rather than what they paid

for the title sponsorship fee. In the other words, the stock market sees that title

sponsorships for PGA are overvalued and those of LPGA are undervalued. For a more

practical suggestion, the results of this study imply that title sponsorship fee for PGA









General Accounting Office report. (1993). 1992 Olympic sponsorship-profit or loss is
unknown. Retrieved November 17, 2008, from LEGISTORM web site:
http://www.legistorm. com/score_gao/show/id/23325. html

Giannoulakis, C. & Stotlar, D. (2006). Evolution of Olympic sponsorship and its impact on
the Olympic movement. Retrieved October 20, 2009 from
http://findarticles.com/p/articles/mi_7091/is_2006_Oct/ai_n28503594/pg_2/?tag=c
ontent;coll

Giannoulakis, C., Stotlar, D., & Chatziefstathiou, D. (2008). Olympic sponsorship:
evolution, challenges and impact of the Olympic Movement. International Journal
of Sport Marketing & Sponsorship, 9(4), 256-270.

Grunert, K. G. (1996) Automatic and strategic processes in advertising effects.
Journal of Marketing, 60(4), 88-101.

Gwinner, K. (1997). A model of image creation and image transfer in event
sponsorship. International Marketing Review, 14(3), 145-158.

Gwinner, K., & Eaton, J. (1999). Building brand image through event sponsorship: The
role of image transfer. Journal of Advertising, 28(4), 47-57.

Harig, B. (2009). Tours' sponsorship woe creating uncertainty. Retrieved October 28,
2009 from ESPN.com:
http://sports.espn.go.com/golf/notebook?page=notebook/bb-090805

IEG Sponsorship Report. (2000). Thatlook.com sets sights on more sponsorship.
Sponsorship Report, 19(5), 3.

IEG Sponsorship Report. (2006). Projection: sponsorship growth to increase for fifth
straight year. retrieved November 17, 2008, from the IEG Sponsorship Report
Sample Issue: http://www.sponsorship.com/documents/SR_Promolssue_01-
07.pdf

IEG Press Release. (2009). Sponsorship spending to rise 2.2% in 2009. Retrieved
September 25, 2009 from http://www.sponsorship.com/About-IEG/Press-
Room/Sponsorship-Spending-To-Rise-2.2-Percent-in-2009.aspx

Ko, Y. J., Kim, K., Claussen, C. L., & Kim T. H. (2008). The effects of sport involvement,
sponsor awareness and corporate image on intention to purchase sponsors'
products. International Journal of Sports Marketing and Sponsorship, 9(2), 79-94.

MacKinlay, A. C. (1997). Event studies in economics and finance. Journal of Economics
Literature, 35(1), 13-39.

McWilliams, A., & Siegel, D. (1997). Event studies in management research: Theoretical
and empirical issues. Academy of Management Journal, 40(3), 626-657.









APPENDIX
SPONSOR-EVENT FIT SCALE

Speed & Thompson's (2000) sponsor-event fit scale items

5 Questions used to determine the level of congruence are;

Q1. There is a logical connection between the event and the sponsor.
Q2. The image of the event and the image of the sponsor are similar.
Q3. The sponsor and the event fit together well.
Q4. The company and the event stand for similar things.
Q5. It makes sense to me that this company sponsors this event.

7 point Liker type scale

Strongly Disagree Strongly Agree

1 2 3 4 5 6 7


As noted in the data collection section, the answer of the 5 questions was averaged and then the
average was used as the congruence level between sponsor and event. This procedure was taken
in all 84 title sponsorship samples.









In summary, LPGA and NASCAR Sprint sponsors received significant stock

market returns following the title sponsorship announcements, while PGA sponsors

experienced significantly negative decreases in the stock prices and NASCAR

Nationwide sponsors did not neither positive nor negative. These results confirmed that

stock market reactions toward sport sponsorship are different among sport entities

(Cornwell el al, 2005) and therefore, research should be conducted separately based on

the sports in order to fully measure the impact of title sponsorship on stock prices.

Event Date

The second stage of event study analysis verifies whether title sponsorship event

following the title sponsorship announcement in the PGA, LPGA, and NASCAR brings

significance in the title sponsors' stock prices. Table 5-11 reports the results of the

mean abnormal returns (MAR) and their associated statistics tests for the interval from t

= -3 to +3 for the overall 81 samples of title sponsorship announcements. 1 sample from

the PGA and 2 samples from the LGPA were not included for this event date analysis

because the EVENUS program could not retrieve the precise stock price information

from the date. Table 5-12 shows the results of the mean cumulative abnormal return

(MCAR) and their associated statistics tests over three different event windows (0, 1),

(0, 2), and (0, 3) same as the announcement date.

Table 5-11 shows that title sponsors in the PGA, LPGA, and NASCAR

experienced significant stock market returns at 5% statistical significant level on 2 days

before the title sponsorship event. Please note that unlike the case of announcement

date, stock market's significant reaction before the event does not mean information

leakage and after the event does not mean market's delay of reaction or interpretation

because the event date has been made public (usually long ago before the event) since









Hypothesis 6: In the NASCAR title sponsorship, title sponsors of the Sprint Cup

Series are likely to have more stock market returns than do those of the

Nationwide Series.

Industry Segment (H7)

Industry Segment refers to the industrial sector to which the title sponsor

corporation belongs. Industrial Segments included in this study are high-technology,

automobile, food, banking, investment, and pharmacy.

Several studies (Caldron-Martinez et al., 2005; Clark et al., 2002; Cornwell et al.,

2005a) demonstrated that high-technology companies in the computer, internet,

telecommunication, and electricity sectors experienced positive reactions following the

announcements of their sponsorship more than did the other types of companies such

as banking, insurance, construction, food and so on. For example, Clark et al. (2002)

reported that high technology companies sponsoring stadium naming rights

experienced over 4% greater net increases in stock prices following the stadium

sponsorship announcements than did non-high technology companies. Likewise, The

study of Cornwell et al. (2005a) showed that high technology companies sponsoring

major-league sports' official products experienced about 11% greater net increase in

shareholder wealth following the announcements than did not-high technology

companies. Furthermore, Caldron-Martinez et al. (2005) found that among various types

of sponsorship firms, the sector of electricity, which could be categorized into high

technology companies in the above two studies, was a determinant of abnormal stock

market returns. Actually in their study, among seven industrial sectors including

construction, automobile, communication, petroleum, banking, insurance and electricity,

only the sector of electricity brought a statistically significant difference in gaining









On the other hand, as an untraditional approach in an effort to measure

sponsorship effectiveness, Meenaghan (1991) suggested to measure sponsorship

effectiveness in terms of sales impact or other indicators of company's performance

(i.e., market outcome) such as corporate accounting. However, cost-benefit analysis of

sponsorship (i.e., comparing revenue changes and its associated costs) is in most

cases very complicated and as a result, brings ambiguous results (Miyazaki & Morgan,

2001). For instance, a research effort by the United States General Accounting Office

("GAO") to evaluate the sponsorship effectiveness of the U.S. Postal Service in 1992

Olympics eventually resulted in "unknown" (General Accounting Office Report, 1993).

The conclusion of GAO was that GAO could not determine whether the Olympic

sponsorship produced a profit or loss because GAO could not verify the key revenues

and costs attributed solely to the Olympic sponsorship (General Accounting Office

Report, 1993). In summary, while proving the market value of sponsorship is the best

way to justify sponsorship investment as a useful marketing technique (Cornwell &

Maigan, 1998), measuring sponsorship effects in terms of company's performance as a

final market outcome is a very challenging task (Calderon-Martinez et al., 2005).

Fortunately, however, this difficulty can be approached by employing a methodological

technique called event study analysis, which uses stock price fluctuations of a

corporation in order to investigate the market reactions to the announcement of

sponsorship agreement between the corporation and the event (Miyazaki & Morgan,

2001).

Event Study Analysis

Event study analysis is an analysis of "whether there was a statistically significant

reaction in financial markets to past occurrences of a given type of event that is









performances. In this regard, repetition of this study would be beneficial to evaluate the

current title sponsorship fee in the PGA, LPGA, and NASCAR, and the results could be

used as a guidance to accommodate the current sponsorship fee or to decide

appropriate title sponsorship fee in the following year.









Spais, G. S., & Filis, G. N. (2006). Stock market reaction on Olympic sponsorship
announcement using event-study method. Journal of Korean Academy of
Marketing Science, 16(1), 95-108.

Spais, G. S., & Filis, G. N. (2008). Measuring stock market reaction to sponsorship
announcements: The case of Fiat and Juventus. Journal of Targeting,
Measurement and Analysis for Marketing, 16(3), 169-180.

Speed, R., & Thompson, P. (2000). Determinants of sports sponsorship response.
Journal of the Academy of Marketing Science, 28(2), 226-238.

Sponsor Map. (2008). Strong growth for the global sponsorship industry in 2008.
Retrieved November 17, 2008, from the Sponsor Map Web site:
http://www.sponsormap.com/?p=72

Stipp, H. (1998). The impact of Olympic sponsorship on corporate image. International
Journal of Advertising, 17(1), 75-87.

Stipp, H., & Schiavone, N. P. (1996). Modeling the impact of Olympic sponsorship on
corporate image. Journal of Advertising Research, 36(4), 22-28.

Tripodi, J. (2001). Sponsorship a confirmed weapon in the promotional armoury.
International Journal of Sports Marketing and Sponsorship, 20(1), 1-20.









should be lowered and for LPGA it could be raised when considering what title sponsors

paid for and what they are likely to benefit by the new established relationship with the

sport entity. Otherwise, PGA should make up the sponsors' negative results in stock

market returns by demonstrating additional financial benefit that stock market could not

take into account as stock price increases but that title sponsor could get due to the

sponsorship. By doing so, current title sponsorship fee for PGA could keep sounding

"reasonable". Emphasizing on non financial or not soon reflected as figures but surely

fruitful later benefits would be another strategy for PGA keeping the current sponsorship

fee. Such examples include emphasizing on hospitality opportunity with key and

potential customers, enhancement of employee's loyalty to the company, and

enhancement of corporate image through the partnership with PGA.

Consistent with past findings, congruence was an important determinant of the

stock price increases for title sponsors. This result confirmed that the already

established value of high image congruence between sponsor and sponsored event

relationship in title sponsorship too. As well as the congruence level, this study also

confirmed the past finding that high-technology companies are likely to get higher stock

market returns from the announcement of title sponsorship agreement.

As a new attempt in event study analysis, this study also focused on actual

sporting event date following the new title sponsorship announcement. As a result,

some significance level of abnormal stock returns have been observed in title

sponsorship event date as well as the announcement date. By these findings, it is

suggested that the importance of taking into account the stock price changes around

the event date besides the announcement date in order to fully measure the









5-16 Mean cumulative abnormal return (MCAR) for event date in LPGA................ 64

5-17 Mean abnormal returns (MAR) for event date in NASCAR (Sprint).................... 65

5-18 Mean cumulative abnormal return (MCAR) for event date in NASCAR
(S print) ......... ......... .. ........................................... .. ..... .. 65

5-19 Mean abnormal returns (MAR) for event dates in NASCAR (Nationwide).......... 66

5-20 Mean cumulative abnormal return (MCAR) for event dates in NASCAR
(N atio nw id e ) ....... ........................................................................ 6 6

5-2 1 M multiple R egression A nalysis................................................... ... .................. 67

5-22 ANOVA for industry segment.... ......................................................... ............... 68

5-23 Regression for high-technology company ................................................ 68
































2010 Masaki Kudo









may influence consumer perceptions, efficacy and market valuation (Becker-Olsen,

2003). For example, naming right sponsorship is likely to gain a greater number of

exposures due to the longer duration while event sponsorship tends to have more

supporting advertising messages due to the shorter duration (Becker-Olsen, 2003).

Also, when it comes to sphere of influence, naming right sponsorship seems to have

advantage over traditional sponsorship because of the longer duration, more exposure

through nationwide television media, and larger target audiences (Becker-Olsen, 2003).

For instance, regarding the difference in duration, one week or even one month after a

PGA Tour, people are more likely to remember the name of title sponsor of the tour,

while they are not as likely to remember the name of individual sponsors each player

has. It seems that the "high visibility" the naming right sponsor has plays an important

role in title sponsorship too. In fact, Stipp (1998) noted that high visible naming right

sponsor has advantage in building stronger positive brand associations in consumer

perception over traditional sponsorship. As an example of the "high visibility" a sponsor

gained through their title sponsorship, Thatlook, a referral service for elective cosmetic

surgery, recorded more than 4 million Internet hits in the three days following its title

sponsorship of a NASCAR Winston Cup, which was about 16 times as a normal day's

hits (IEG Sponsorship Report, 2000).

In an exchange for such "high visibility" and other benefits title sponsorship gain,

the right fee for title sponsorship is generally expensive, especially when it comes with

nationwide or global level television exposure. As a matter of fact, in the PGA Tour,

annual 5% increases of title sponsorship fees have been considered as a standard part

of title sponsorships (Show, 2009). Also, despite the current economic recession of









CHAPTER 2
REVIEW OF LITERATURE

Title Sponsorship

Since sponsorship particularly focusing on sport title sponsorship has not been

well developed, anecdotal evidences were collected from various sponsorship-related

journal articles and business resources (e.g., naming rights sponsorship, which is

considered one type of title sponsorship). Based on the definition of sponsorship, "the

acquisition of rights to affiliate or directly associate with a product or event for the

purpose of deriving benefits related to that affiliation or association" (Mullin et al., 2000,

p.254), title sponsorship can be defined as the acquisition of rights to take part in the

official name of the event for the purpose of deriving benefits related to that name

sharing.

Then why do corporations seek title sponsorship? According to Clark, Cornwell,

and Pruitt (2002), it is natural that there is not much difference between corporate

objectives that managers hold for the naming right sponsorship and those held for other

event and activity type of sponsorship. The similarity they have is that both naming right

sponsorship and traditional event sponsorship utilize their rights of sponsorship to

maximize brand awareness and seek strong brand associations through the repeated

parings (Becker-Olsen, 2003). Specifically, Roy and Cornwell (1999) identified the six

objectives that managers hold for the naming right sponsorship; Image Enhancement,

Less Cluttered Communications Environment, Awareness, Brand Positioning, part of an

Integrated Marketing Communications Plan, and Direct On-site Sales. On the other

hand, there are some differences between the objectives of naming right sponsorship

and those of traditional sponsorship in its scope, duration and value. Those differences









Morgan, 2001). While these methods measuring such as media exposure, awareness,

and image enhancement are relatively easy to adopt for sponsoring companies

(Crompton, 2004), these methods may not be able to provide the real effectiveness in

terms of the commercial aspect on sponsorship (Pham, 1991). Media exposure method,

for example, is problematic because it is not the ultimate purpose of sponsorship

(Pham, 1991) and achievement of a certain level of media exposure does not

necessarily lead to a high recall rate or positive attitude change in consumer behavior

(Speed & Thompson, 2000). In fact, Crompton (2004) pointed out that "[advertising]

managers will be required to design evaluation research that measures the extent to

which [sponsorship] benefits are required, rather than being able to rely on the much

easier measures that evaluate media exposure, awareness and image enhancement"

(p. 280). And continuously, "the conventional wisdom appears to be that there is now

more pressure on senior managers to demonstrate accountability for sponsorship

investments by showing their potential for increasing a company's profitability" (p. 268).

That is, demonstrating the commercial value of sponsorship as a final market outcome

has been required for sponsoring companies to justify their high sponsorship fee

investment (Cornwell & Maigan, 1998; Crompton, 2004).

As described previously, the traditional approach to measuring sponsorship

effectiveness, which mainly focused on media exposure or image transfer, does not

necessarily justify the high sponsorship fee. Instead of such traditional approaches,

demonstrating positive results at the level of company's performance stage provide

additional "accountability" that exposure or image transfer related research may not be

able to identify in evaluating sponsorship effectiveness. Specifically, Cornwell and









The estimation of the above equation allows the calculation of the daily abnormal

stock market returns (AR) as follows (McWilliams & Siegel, 1997):

ARit = Rit (ai + bi Rmt)

where

ai and bi are the estimations obtained with the parameter estimates obtained from the

regression of Rit on Rmt over an estimation period (T). In this study, the researcher and

used a maximum of 255 daily return observations for the period around the

announcement, starting a day -251 and ending at day +3 relative to the event. That is,

the first 248 days (-251 to -4) was designated for the estimation window, and the

following 7 days (-3 to +3) was designated for the event window. The purpose the event

window was set from -3 to +3 is to obtain the full effects of the event. The 3 days

period accounts for information leakage prior to the announcement, and the + 3 days

period accounts for information spread after the announcement. For the market model

parameter estimates, while some research use the ordinary least squares (OLS) model,

this study employed the Scholes-Williams standardized cross-sectional market model

(Scholes &Williams, 1977) because the Scholes-Williams standardized cross-sectional

market model is designed to eliminate the problems associated with nonsynchronous

trading that sometimes occurs in event-based studies with firms of widely varying

market values (Cornwell et al., 2005a).

Following the standard practice (Cornwell et al., 2005a; Pruitt et al., 2004), all

statistical calculations were performed using the EVENTUS program developed by

Cowan Research, L.L.C. The EVENTUS program is specifically designed for the event













Table 4-2. LPGA title sponsorship sample
Sport Company Name
LPGA SYBASE Sybase Big
LPGA OFFICE DEPOT The Office
LPGA TYCO INTERNATIONAL Tyco/ADT
LPGA SAFEWAY Safeway Pi
LPGA H S B C HOLDINGS HSBC Wor
LPGA CANADIAN NATIONAL RAILWAY CN Canadia
LPGA HONDA MOTOR Honda LPG
LPGA NAVISTAR INTERNATIONAL Navistar LP
LPGA H S B C HOLDINGS HSBC Wor
LPGA BELL MICROPRODUCTS Bell Micro
LPGA PROCTER & GAMBLE P&G Beaut


Title Sponsorship Name
Apple Classic
Depot Hosted by Amy Alcott
Championship
ng
en's World Match Play Championship
in Women's Open
A Thailand
GA Classic
en's Champions
LPGA Classic
y NW Arkansas Classic


Announcement Date
1/9/2001
3/26/2001
5/9/2001
7/2/2003
1/25/2005
10/19/2005
5/17/2006
7/6/2006
8/30/2007
1/10/2008
1/31/2008


Event Date
7/19/2001
4/12/2001
11/15/2001
3/18/2004
6/30/2005
8/10/2006
10/20/2006
9/27/2007
2/28/2008
9/11/2008
7/4/2008











Table 5-3. Mean abnormal returns (MAR) for announcement date in PGA
Mean
Abnormal Positive: Patell Generalized
Day N Return Negative Z Sign Z
-3 23 0.25% 16:7> 1.123 1.985*
-2 23 0.04% 9:14 -0.343 -0.935
-1 23 -0.19% 10:13 -0.352 -0.518
0 23 0.00% 8:15( -0.577 -1.353$
+1 23 0.06% 12:11 0.844 0.316
+2 23 -0.48% 8:15( -2.697** -1.353$
+3 23 0.18% 14:9 0.058 1.150
The symbols $,*,**, and *** denote statistical significance at the 0.10,
0.05, 0.01 and 0.001 levels, respectively, using a generic one-tail test.
The symbols (,< or ),> etc. correspond to $,* and show the direction
and generic one-tail significance of the generalized sign test.






Table 5-4. Mean cumulative abnormal return (MCAR) for announcement date in PGA
Mean
Cumulative Precision
Abnormal Weighted Positive: Patell Generalized
Days N Return CAAR Negative Z Sign Z
(0,+1) 23 0.06% 0.08% 13:10 0.188 0.733
(0,+2) 23 -0.42% -0.72% 8:15( -1.403$ -1.353$
(0,+3) 23 -0.24% -0.71% 9:14 -1.186 -0.935
The symbols $,*,**, and *** denote statistical significance at the 0.10, 0.05, 0.01 and 0.001
levels, respectively, using a generic one-tail test. The symbols (,< or ),> etc. correspond
to $,* and show the direction and generic one-tail significance of the generalized sign test.









Confounding events are major news or other significant information that might have an

impact on the stock prices during the event window (Becker-Olsen, 2003). In general,

the longer the event window, the more difficult it is for researcher to justify the above

assumption and claim that confounding effects had been controlled (McWilliams &

Siegel, 1997).

In short, while event study analysis has been widely employed in management

and academic research, there is a fact that the usefulness of this methodology depends

heavily on several strong assumptions (Brown & Warner, 1985). By violating these

assumptions, the results of research may be biased and inaccurate; therefore, the

conclusion may be problematic, or even it is possible to manipulate the conclusion by

employing inappropriate techniques (Spais & Filis, 2008).

Basically, event study analysis involves measuring how a certain event (i.e.,

announcement of sponsorship agreement in this study) influences movement of

particular stock prices. According to Miyazaki and Morgan (2001), "it is important to

compare the movement of the individual stock to the market as a whole in order to

determine whether a movement of large magnitude in the individual stock price

coincides with what of the market or whether it was due to firm-specific information"

(p.11). The abnormal stock market returns (i.e., returns not consistent with the pattern of

change as established by past company and market activity) are calculated as the

difference between the actual and the expected returns. Thus, a positive abnormal

return (i.e., a significant increase beyond normally expected returns) indicates that the

event is typically seen as beneficial to a company's market value. On the other hand, a

negative abnormal return (i.e., a significant decrease beyond normally expected returns)









equivalency calculation to assess the sponsorship effectiveness, this type of

measurement seems to be the most widely adopted tool in the business world

(Poknywczynski, 2000). Poknywczynski (2000) reported that while 70% of companies

they researched were using media exposure as a tool for their sponsorship impact

measurement, only 40% of companies used the impact on awareness and image, and

only 7% of companies used the impact of sales for their measurement.

The latter approach, corporate image transfer, is typically measured through

surveys directed at the consumers. The rationale behind the corporate image transfer is

that pre-existing associations held in consumers' memory with respect to a sponsorship

event become linked in memory with the sponsoring company; therefore, the image of

sponsorship event is transferred into the company or product (Gwinner, 1997; Gwinner

& Eaton, 1999). Gwinner and Eaton (1999), for example, demonstrated that

sponsorship results in corporate image transfer and when sponsoring brand are

matched on either on the image or function with the event, its positive image transfer is

enhanced. Also, Grunert (1996) showed that a simple exposure leads to a better

evaluation of a product or brand under the situation that the consumers automatically

respond to the advertising. However, this type of measurement also has the limitation in

that it does not differentiate between the impacts of a sponsorship per se and the

impacts of other forms of advertising (Calderon-Martinez et al., 2005; Miyazaki &

Morgan, 2001). The difficulty of differentiation of its effects from sponsorship per se and

other promotional type of marketing seems to be a common problem in evaluating

sponsorship effectiveness (Pham, 1991).











Table 5-19. Mean abnormal returns (MAR) for event dates in NASCAR (Nationwide)
Mean
Abnormal Positive: Patell Generalized
Day N Return Negative Z Sign Z
-3 28 -0.36% 13:15 -0.629 -0.255
-2 28 -0.24% 16:12 0.728 0.880
-1 28 -0.45% 11:17 -0.789 -1.011
0 28 0.32% 18:10) 1.233 1.636$
+1 28 -0.07% 11:17 -0.322 -1.011
+2 28 -0.10% 14:14 0.294 0.124
+3 28 -0.44% 12:16 -2.801** -0.633
The symbols $,*,**, and *** denote statistical significance at the 0.10,
0.05, 0.01 and 0.001 levels, respectively, using a generic one-tail test.
The symbols (,< or ),> etc. correspond to $,* and show the direction
and generic one-tail significance of the generalized sign test.






Table 5-20. Mean cumulative abnormal return (MCAR) for event dates in NASCAR (Nationwide)
Mean
Cumulative Precision
Abnormal Weighted Positive: Patell Generalized
Days N Return CAAR Negative Z Sign Z
(0,+1) 28 0.25% 0.24% 14:14 0.644 0.124
(0,+2) 28 0.15% 0.32% 16:12 0.696 0.880
(0,+3) 28 -0.29% -0.42% 13:15 -0.798 -0.255
The symbols $,*,**, and *** denote statistical significance at the 0.10, 0.05, 0.01 and 0.001
levels, respectively, using a generic one-tail test. The symbols (,< or ),> etc. correspond
to $,* and show the direction and generic one-tail significance of the generalized sign test.









CHAPTER 1
INTRODUCTION

Sport sponsorship is defined as "the acquisition of rights to affiliate or directly

associate with a product or event for the purpose of deriving benefits related to that

affiliation or association" (Mullin, Hardy, & Sutton, 2000, p. 254). In 2008, the

International Events Group estimated that $43.5 billion was spent around the world, and

of that, $16.8 billion was spent in North America by companies' sponsorship activities

including sporting events, music events, festivals, and so on (Sponsor Map, 2008). Of

the $43.5 billion sponsorship money, roughly 65-70% has been reportedly invested into

sport sponsorship (IEG Sponsorship Report, 2006). This fact clearly indicates that

utilizing sponsorship has become one of the most important advertising channels for

companies to differentiate themselves among competitors. In reality, sponsorship

spending overall is still projected to rise 3.9% in 2009 (IEG Press Release, 2009),

regardless of a severe economic recession, which generally results in the dramatic

budget cuts in marketing expenditures including sponsorship activities.

According to Tripodi (2001), this uprising trend of sport sponsorship is expected to

continue. This would be definitely true when tracking the amount of sponsorship

revenue that the International Olympic Committee (IOC) has gained through The

Worldwide Olympic Partners program. The Worldwide Olympic Partners have been

known as the top category of Olympic sponsorship programme and was created one

year after the Los Angeles Olympic in 1984, when the IOC experienced a tremendous

financial success and realized that corporate sponsors provided the Olympic with

substantial profits and that financing through sponsorship became an integral part of the

Olympic Movement (Giannoulakis & Stotlar, 2006). In fact, the IOC generated $95









abnormal stock market returns at a probability < 0.01 level. These findings suggest the

following hypothesis:

Hypothesis 7: Title sponsors whose business segment is high-technology in

computer, Internet, and electricity industries are likely to have more stock market

returns than do non-high-tech companies.











Table 5-5. Mean abnormal returns (MAR) for announcement date in LPGA
Mean
Abnormal Positive: Patell Generalized
Day N Return Negative Z Sign Z
-3 11 0.08% 4:7 -0.432 -0.882
-2 11 -0.23% 5:6 -1.401$ -0.279
-1 11 -0.08% 5:6 0.336 -0.279
0 11 0.10% 4:7 -0.455 -0.882
+1 11 0.24% 9:2> 0.830 2.133*
+2 11 -0.81% 4:7 -1.605$ -0.882
+3 11 -0.31% 5:6 -1.297$ -0.279
The symbols $,*,**, and *** denote statistical significance at the 0.10,
0.05, 0.01 and 0.001 levels, respectively, using a generic one-tail test.
The symbols (,< or ),> etc. correspond to $,* and show the direction
and generic one-tail significance of the generalized sign test.






Table 5-6. Mean cumulative abnormal return (MCAR) for announcement date in LPGA
Mean
Cumulative Precision
Abnormal Weighted Positive: Patell Generalized
Days N Return CAAR Negative Z Sign Z
(0,+1) 11 0.34% 0.15% 6:5 0.265 0.324
(0,+2) 11 -0.47% -0.49% 4:7 -0.710 -0.882
(0,+3) 11 -0.78% -1.00% 4:7 -1.263 -0.882
The symbols $,*,**, and *** denote statistical significance at the 0.10, 0.05, 0.01 and 0.001
levels, respectively, using a generic one-tail test. The symbols (,< or ),> etc. correspond
to $,* and show the direction and generic one-tail significance of the generalized sign test.











Table 5-13. Mean abnormal returns (MAR) for event date in PGA
Mean
Abnormal Positive: Patell Generalized
Day N Return Negative Z Sign Z
-3 22 0.23% 10:12 0.026 -0.293
-2 22 0.02% 13:9 0.474 0.987
-1 22 0.28% 9:13 0.544 -0.719
0 22 -0.61% 9:13 -1.455$ -0.719
+1 22 -0.03% 11:11 -0.108 0.134
+2 22 -0.70% 6:16< -1.785* -1.999*
+3 22 0.09% 10:12 0.303 -0.293
The symbols $,*,**, and *** denote statistical significance at the 0.10,
0.05, 0.01 and 0.001 levels, respectively, using a generic one-tail test.
The symbols (,< or ),> etc. correspond to $,* and show the direction
and generic one-tail significance of the generalized sign test.






Table 5-14. Mean cumulative abnormal return (MCAR) for event date in PGA
Mean
Cumulative Precision
Abnormal Weighted Positive: Patell Generalized
Days N Return CAAR Negative Z Sign Z
(0,+1) 22 -0.65% -0.49% 8:14 -1.105 -1.146
(0,+2) 22 -1.35% -1.05% 7:15( -1.933* -1.572$
(0,+3) 22 -1.26% -0.96% 8:14 -1.523$ -1.146
The symbols $,*,**, and *** denote statistical significance at the 0.10, 0.05, 0.01 and 0.001
levels, respectively, using a generic one-tail test. The symbols (,< or ),> etc. correspond
to $,* and show the direction and generic one-tail significance of the generalized sign test.









the value of sponsorship is very difficult task for both sponsorship industry and

academic researcher (Giannoukakis et al., 2008; Oztur et al., 2004; Miyazaki & Morgan,

2001). As noted, in the sponsorship industry, 40% of companies engaged in

sponsorship activity reported they have nothing spent on the research for sponsorship

effectiveness (Crompton, 2004). Also, in the academic field, research regarding

sponsorship measurement tends to traditionally focus on either media exposure or

corporate image transfer (Oztur et al., 2004; Miyazaki & Morgan, 2001).

The former method, media exposure, has been conducted through the

quantification of media exposure and coverage by either measuring or counting the

times the sponsor is referred to, the times its logo appears, or the duration of the

appearance of the sponsor name or logo in the media (Speed & Thompson, 2000). In

most cases, such media exposure is then translated into money, in order to calculate

the advertising cost that would be required to achieve an equivalent exposure or

coverage in the traditional forms of advertising (Calderon-Martinez et al., 2005).

However, this approach has several limitations. For instance, Pham (1991) argued that

for the sponsoring company, media coverage is not the ultimate purpose of

sponsorship; therefore, it should not be used for the effectiveness evaluation. As a

matter of fact, achievement of a certain level of media coverage does not necessarily

imply a high recall rate or a positive attitude change in consumer behavior (Ozturk et al.,

2004; Speed & Thompson, 2000). Therefore, Pham (1991) pointed out that the

exposure method cannot provide or analyze the real effectiveness at least in terms of

the commercial aspect on sponsorship. However, despite the fact that there are

substantial limitations when using media exposure and following advertisement









terms of delivered concepts, the possibility of mismatch between the sponsored event

and the sponsoring brand is relatively high (Pruitt et al., 2004). Therefore,

Hypothesis 3: Title sponsors whose title is accompanied with the corporate name

(i.e., corporate-level sponsorship) are likely to have more stock market returns

than do those whose title is named by the corporation's product or service (i.e.,

brand-level sponsorship).

Prestige of Event (H4) (Hs)

As noted previously, for the purpose of this study, prestige of event is defined by

three elements of the event; (a) Amount of Purse, (b) Traditionality, and (c) Event

Competition Level (Sprint/Nationwide). Amount of Purse and Traditionality are used for

the PGA and LGPA. Event Competition Level is used for the NASCAR.

Amount of purse and Traditionality: In the PGA and LPGA title sponsorship, in

an effort to evaluate the effects of differences in tournament prestige on sponsorship

returns, Amount of Purse and Traditionality were included as potential variables.

Amount of Purse refers to the total amount of purse the title sponsor raised for the

tournament. Traditionality refers to the length of the tournament having been held.

According to Speed and Thompson (2000), perceived prestige of the sponsored event

is a determinant that positively associates with the sponsorship response. For example,

a special and high-status event such as the Olympics is likely to create such

opportunities for sponsors since the audience usually has a high regard toward such an

event, regardless the level of personal regard for the sponsor (Stipp & Schiavone,

1996). Thus, prestige of the PGA and LPGA tournament, which is measured by amount

of purse and traditionality in this study, is likely to be seen in a positive manner in the

stock market. Therefore,









CHAPTER 5
RESULTS

This study is comprised of two parts: (1) event study analysis to examine stock

price fluctuations of new title sponsors in both title sponsorship announcement date and

the title sponsorship event date (Hi), and (2) exploration of the selected moderator

variables that might influence stock price fluctuations by using the results of (1) (H2

through H7).

Event Study Analysis

Announcement Date

The first stage of event study analysis verifies whether announcement of new title

sponsorship agreement in the PGA, LPGA, and NASCAR brings significance in the title

sponsors' stock prices. Table 5-1 reports the results of the mean abnormal returns

(MAR) and their associated statistics tests for the interval from t = -3 to +3 for the overall

84 samples of title sponsorship announcements. Table 5-1 also reports the number of

announcements in the sample (N), the number of companies registering positive and

negative abnormal return changes (Positive:Negative), and the associated statistics

tests (Patell Z and Generalized Sign Z) for this fraction for each event day. The Patell

test, which usually referred as the Z statistic and known as the standardized abnormal

returns test, assumes cross sectional independence among the samples. The null

hypothesis for the Patell test is that the mean abnormal returns for each event day

should be approximate zero. The Generalized Sign Z test is a nonparametric test for the

significance of a differing proportion of positive or negative returns in the sample period

when compared with the estimation period (Pandey, Shanahan, & Hansen, 2005). That

is, this test controls for the normal asymmetry of positive and negative abnormal returns











Table 5-7. Mean abnormal returns (MAR) for announcement date in NASCAR (Sprint)
Mean
Abnormal Positive: Patell Generalized
Day N Return Negative Z Sign Z
-3 22 -0.04% 10:12 -0.593 -0.246
-2 22 0.41% 11:11 1.212 0.181
-1 22 -0.12% 12:10 -1.133 0.607
0 22 0.53% 15:7> 1.435$ 1.888*
+1 22 -0.21% 11:11 -0.692 0.181
+2 22 -0.27% 9:13 -0.412 -0.673
+3 22 -0.15% 11:11 -0.198 0.181
The symbols $,*,**, and *** denote statistical significance at the 0.10,
0.05, 0.01 and 0.001 levels, respectively, using a generic one-tail test.
The symbols (,< or ),> etc. correspond to $,* and show the direction
and generic one-tail significance of the generalized sign test.






Table 5-8. Mean cumulative abnormal return (MCAR) for announcement date in NASCAR (Sprint)
Mean
Cumulative Precision
Abnormal Weighted Positive: Patell Generalized
Days N Return CAAR Negative Z Sign Z
(0,+1) 22 0.32% 0.19% 15:7> 0.525 1.888*
(0,+2) 22 0.05% 0.09% 12:10 0.191 0.607
(0,+3) 22 -0.10% 0.03% 9:13 0.066 -0.673
The symbols $,*,**, and *** denote statistical significance at the 0.10, 0.05, 0.01 and 0.001
levels, respectively, using a generic one-tail test. The symbols (,< or ),> etc. correspond
to $,* and show the direction and generic one-tail significance of the generalized sign test.











Table 5-9. Mean abnormal returns (MAR) for announcement date in NASCAR (Nationwide)
Mean
Abnormal Positive: Patell Generalized
Day N Return Negative Z Sign Z
-3 28 -0.08% 14:14 0.130 0.162
-2 28 0.25% 12:16 0.132 -0.594
-1 28 0.12% 13:15 0.430 -0.216
0 28 0.14% 14:14 0.254 0.162
+1 28 -0.28% 12:16 -1.339$ -0.594
+2 28 0.03% 13:15 0.657 -0.216
+3 28 -0.21% 12:16 -0.853 -0.594
The symbols $,*,**, and *** denote statistical significance at the 0.10,
0.05, 0.01 and 0.001 levels, respectively, using a generic one-tail test.
The symbols (,< or ),> etc. correspond to $,* and show the direction
and generic one-tail significance of the generalized sign test.






Table 5-10. Mean cumulative abnormal return (MCAR) for announcement date in NASCAR (Nationwide)
Mean
Cumulative Precision
Abnormal Weighted Positive: Patell Generalized
Days N Return CAAR Negative Z Sign Z
(0,+1) 28 -0.13% -0.28% 14:14 -0.767 0.162
(0,+2) 28 -0.10% -0.11% 15:13 -0.247 0.540
(0,+3) 28 -0.31% -0.33% 13:15 -0.640 -0.216
The symbols $,*,**, and *** denote statistical significance at the 0.10, 0.05, 0.01 and 0.001
levels, respectively, using a generic one-tail test. The symbols (,< or ),> etc. correspond
to $,* and show the direction and generic one-tail significance of the generalized sign test.










Table 5-22. ANOVA for industry segment


Industry Segment


High-technology


Retails


Services


Total


Minimum


-.047


-.050


-.054


-.054


Cumulative Abnormal Returns (CAR)

Mean Maximum Std. Deviation


.0046 .038 .0177


.0031 .061 .0204


-.0066 .053 .0223


.0003 .061 .0205


F-statistic 2.667


Significance .076


Table 5-23. Regression for high-technology company
N
Coefficient t
F-Statistic
Significance
R2
Adjusted R2
*shows statistical significance at 5% level


84(high-tech 19, other 65)
2.266
5.135
*0.026
0.059
0.047









sponsors demonstrated that they experienced significant abnormal stock price

increases following the sponsorship announcements (Pruitt et al., 2004). These

research imply that there is a possibility that insignificant events may conceal some

important trends in the data by crossing off some significance in the data with their

insignificances when averaging all data of events regardless of the sport entity.

Therefore, the researcher conducted event study analysis in the PGA, LPGA, and

NASCAR respectively. The each result is as follows.

PGA

Table 5-3 and Table 5-4 show mean abnormal returns (MAR) and mean

cumulative abnormal returns (MCAR) for the title sponsorship announcements in PGA.

As shown in the Table 5-3, title sponsors for PGA experienced negative stock market

returns on the whole including statistically significant level at 10% on the title

sponsorship announcement date and at 5% level on the 2 days after the announcement

date. Table 5-4 also shows statistically negative significance on the (0, +2) window.

These results indicate that market sees entering title sponsorship for PGA tour brings

financially net negative for the company and that title sponsorship fee for PGA tour may

be too expensive when considering what they get by the sponsorship.

LPGA

Table 5-5 and Table 5-6 show MAR and MCAR for the LPGA title sponsors.

Interestingly, as opposed to the results in PGA, title sponsors in the LPGA experienced

significant stock price increases on the +1 day after the announcement. In fact, 9 of 11

companies received positive abnormal returns on the day. While Patell test reported

some negative significance at 10% level around the announcement date, they are far

more than 2 days. In general, the longer the event window, the more difficult it is to









effectiveness of title sponsorship in terms of stock price changes. When considering

that title sponsorship has advantage over traditional sport sponsorship and other

advertising due to the longer duration, high visibility via nationwide broadcasting, and

larger target audiences (Becker-Olsen, 2003; Stipp, 1998), it is reasonable that the

actual title sponsorship event day following the new sponsorship agreement had a

positive impact on their stock prices. Specifically, the results of NASCAR Nationwide

series sponsors in the title sponsorship event date were not consistent with those in the

announcement date in that only the event date produced significantly positive stock

market returns. The different results between announcement date and event date in the

NASCAR Nationwide sponsors suggest that research should focus not only on

announcement date but also event date in order to fully measure the impact of title

sponsorship on stock prices.

It is a common sense in both practice and research that measuring effectiveness

of sponsorship is a really challenging issue. In addressing this subject, this study

attempted to measure the effectiveness of sport sponsorship by regarding sponsor's

stock price change as representing one of the final outcomes of the sponsorship, which

cannot be obtained from more traditional approaches such as exposure value

conversion and consumer surveys. The researcher hopes that the results of this study

contribute to the literature in this regard and will be used to further development of sport

sponsorship in the future.

Limitations of the Study

The main limitations of this study are the following three: (1) event study analysis,

(2) the selection of the companies (i.e., title sponsors), and (3) the determination of the

congruence level between title sponsor and sport event.









sponsorship deal seems to have a big impact either positively or negatively on their

stock prices when considering the amount of the sponsorship exposure and media

coverage on the event day.

Purpose of the Study

Accordingly, the purposes of this study are: (1) to examine effectiveness (i.e.,

stock price changes) of title sponsorship in sport events by comparing stock price

changes of corporate sponsors in different sport events sponsorship (i.e., Professional

Golfers' Association (PGA), Ladies Professional Golf Association (LPGA) and National

Association for Stock Car Auto Racing (NASCAR)) on both sponsorship announcement

date and sporting event date following the announcement and (2) to explore the roles of

selected moderator variables in the stock price changes by focusing on Image

Congruence, Corporate vs Brand Name, Industry Segment, Amount of Purse,

Traditionality, and Event Competition Level (Sprint/Nationwide). For the purposes of this

study, the following research questions were developed: (1) Is there any linkage

between sport sponsorship involvement and corporate performance (i.e., stock price

changes)? (2) Are there any differences in sponsorship effectiveness (i.e., stock price

changes) among selected types of sport events (i.e., the PGA, LPGA, and NASCAR)?

And (3) Are there any other variables that influenced the stock price changes?

Significance of the Study

This research would make a contribution to fill two major gaps existing in

sponsorship literature and sport management practice: (1) the lack of research in title

sponsorship effectiveness, and (2) the need for more research measuring sponsorship

effectiveness at the level of corporate performance. The results of this research should









claim that the significance was caused by the event because of the increased possibility

of confounding effects (McWilliams & Siegel, 1997).When considering that the positive

significance was recorded on the + 1day which is immediately after the announcement,

with 5 % statistical significant level, this positive significance would better reflect the

market reaction toward title sponsorship in the LPGA, than do those negative

significance at 10 % level in far more than 2 days. This result indicates that stock

market sees favorably entering title sponsorship for the LPGA and assumes that the title

sponsors are likely to get financial benefits more than what they paid for the rights fee.

NASCAR (Sprint series)

Table 5-7 and Table 5-8 show MAR and MCAR for the NASCAR Sprint series title

sponsors. The results present that NASCAR Sprint series sponsors received significant

stock market returns on the announcement day. The positive significance of MCAR on

the (0, +1) window also indicates that stock market has favorably seen being title

sponsor for the NASCAR Sprint series for increasing future profits of the company.

NASCAR (Nationwide series)

Table 5-9 and Table 5-10 show MAR and MCAR for the NASCAR Nationwide

series title sponsors. Interestingly, unlike the Sprint case, MAR and MCAR did not

report any significance in the Nationwide series with an exception of +1 day negative

significance at 10% significance level. These different results between Sprint series

sponsors and Nationwide series sponsors indicates that the level of competition in

NASCAR matters in gaining stock market returns and that stock market are only

favorable for title sponsors in the higher level event (i.e., Sprint series sponsors), while

lower level sponsors (i.e., Nationwide series sponsors) are seen as neutral.









ACKNOWLEDGMENTS

I gratefully thank the chair and member of my supervisory committee for their

mentoring throughout my thesis. Dr. Ko, Dr. Connaughton, and Dr. Walker, without your

supports and advices, this thesis would not have been completed. I owe my deepest

gratitude to them and it is my greatest honor to meet and have worked with you whom I

respect as a researcher, as an educator, and as a person, here half way across the

globe from where I come from. A special thanks to the University of Florida, and the

Department of Tourism, Recreation, and Sport Management for providing me with the

resources to complete this thesis.











Table 5-21. Multiple Regression Analysis
ALL PGA & LPGA
Variable Coefficient t-statistics
4*A O------


Corp/Brand
Purse
Tradition
Competition Level
N
F-Statistic
Significance


-t. OU5
-0.509


84
*12.382
0.000


0.232
0.215


Adjusted R2


1.084
0.227


32
1.160
0.327
0.074


NASCAR


0.564
50
0.318
0.575


0.007


0.100 -0.140
*shows statistical significance at less than 1% level


Significance
0.000
0.612
0.287
0.822
0.757









5 R E S U L T S .................................................................................................. 4 7

Event Study A analysis ......... .............. ............. ......... ......................... 47
Announcement Date.............................................. ............... 47
P G A .............................................................................................. 4 9
LPGA ......................... .................. 49
NASCA R (Sprint series)................................................ .................... 50
NASCAR (Nationwide series) ........... .................. .. ........... 50
Event Date................................ ............ ............... 51
PGA ...................... ........... .......... ............... 52
LPGA ............. ......... ......... ....................... 52
NASCA R (Sprint series)................................................ .................... 53
NASCAR (Nationwide series) ........... .................. .. ........... 53
Multiple Regression and ANOVA Test ....... ...................... ........ 54
Multiple Regressions ............... .... ......... ............... 54
ANOVA for Industry Segment.......................... .................... 56

6 DISCUSSION AND LIMITATION .................. .............. ........ 69

D is c u s s io n ........... ................... ................. .................................... 6 9
Limitations of the Study...... ............................ ........ 72

APPENDIX SPONSOR-EVENT FIT SCALE ................................... 76

LIST OF REFERENCES .......................... ......... ......... 77

BIOGRAPHICAL SKETCH ................ ......... ................. 81
























6









(H4) and Traditionality (H5) were done for PGA and LPGA sponsors, and Event

Competition Level (Sprint/Nationwide) (H6) was done for NASCAR sponsors.

Table 5-21 presents the results of the regressions. The most striking result shown

in Table 5-21 is that Image Congruence showed statistically significance at p<0.01

(F=12.382, R2= 0.232, Adjusted R2= 0.215). The regression analysis for Image

Congruence and Corporate vs Brand Name reports that those variables explained a

total 23% variance on the abnormal stock market returns (R2= 0.232). Therefore, H2

was confirmed. This result is consistent with numerous prior studies that highlighted the

importance of the degree of congruence in the relationship between sponsor and

sponsored event in gaining better consumers' responses toward sponsorship (Cornwell

et al., 2005b; Gwinner & Eaton, 1999; McDaniel, 1999). Prior studies using event study

analysis tool also demonstrated that congruence level between sponsor and sponsee is

the determinant in gaining positive abnormal stock market returns (Calderon-Martinez et

al., 2005; Cornwell et al., 2001, Cornwell et al., 2005a). The result shown in this study

demonstrated that in title sponsorship also, image congruence level between sponsor

and sponsee is the determinant for abnormal stock market returns. On the other hand,

other variables including Corporate vs Brand Name, Amount of Purse, Traditionality,

and Event Competition Level (Sprint/Nationwide) did not indicate significance in the

analysis (p>0.05, F = 1.160, R2 = 0.074, Adjusted R2 = 0.100 for Amount of Purse and

Traditionality regression, and p>0.05, F=0.318, R2= 0.007, Adjusted R2 = -0.140 for

Event Competition Level regression). Therefore, H3,H4 ,H5 ,and H6 were rejected.

Regarding the H6, however, this result is conflict with those obtained from event studies

conducted for the NASCAR Sprint series sponsors and the NASCAR Nationwide series









First, employing event study analysis is both a strong and weak point in this

study. While the literature and prevalence of this method itself have demonstrated the

validity and reliability of this methodology, the researcher who uses this method needs

to rely on some strong assumptions under the study. The existence of confounding

effects around the event date is one of the most controversial assumptions in event

study analysis. While the researcher carefully tried to choose the event that had no

confounding effects likely to impact the stock prices around the date, it seems

impossible to affirm that there were no confounding effects around the date, and

therefore, that the significance of stock price changes obtained from this study were

caused 100% by the title sponsorship. Also, the researcher needs to mention that one

of the assumptions that the event should be unexpected to public conflicts the second

event study series in this study focusing on the event date following the announcement

date. As noted, unlike the announcement date, the event date is available for everyone

and this fact is incompatible with the assumption that the market is supposed to know

the information from the announcement for the first time and event date in event study

analysis should be unexpected to market before the event date.

Second, the selection of title sponsors can lead to the existence of bias in the

results. For example, in the PGA 2007 tour, there were 14 new title sponsorships. Out

of 14, four companies are private and their stocks are unavailable to public, or they are

exchanged in foreign markets. Five companies were not included because specific

announcement dates were not obtained from the Web search. As a result, remaining

five companies of 14 from this year's PGA tour were used for the analysis. This fact did









According to Brown and Warner (1985), who provided a comprehensive outline

of event study analysis, great care should be taken to precisely determine the date of

the first trading day following the sponsorship announcement. To do so, an extensive

search by marking use of electronic data bases such as Lexis-Nexis, Factiva, U.S.

newspapers and periodicals was conducted. Table 1 presents lists of the title sponsor,

title sponsorship event, sponsorship announcement date, and following event date

analyzed in this study (Table 4-1 for PGA, Table 4-2 for LPGA, Table 4-3 for NASCAR

Sprint series, and Table 4-4 for NASCAR Nationwide series). In an effort to enhance the

relevance, only those sponsorships initiated between January 1, 2000 and December

31, 2008 were included in the analysis. The total number of samples is 84. While small

sample size is quite common in the management literature using event study analysis

(McWilliams & Siegel, 1997), 84 samples in this study are considered to be a large

enough sample. For example, among 12 studies using event study analysis in sport-

related literature, the largest samples were 53 by Cornwell et al. (2005a), and the

average of those 12 studies was 28 samples.

Event Date

As noted previously, besides the announcement day of the title sponsorship

agreement, this study also examines stock price reactions in the sporting event day

whose title is named after the company following the sponsorship agreement. When

considering that naming right sponsorship has advantage over traditional advertising

because of the longer duration, high visibility via nationwide broadcasting, and larger

target audiences (Becker-Olsen, 2003; Stipp, 1998), the actual sporting event day

following the new sponsorship agreement seems to have some impacts either positively

or negatively on their stock prices. Therefore, it would be important to take into account

37









Data Analysis Procedures


Event Study Analysis

The basic procedures followed the standard event study methodology proposed

by Brown and Warner (1985), and McWilliams and Siegel (1997). The event study

analysis is designed to measure the impact of a specific event has on the stock price of

a selected company by comparing actual stock returns of the company to estimated

stock returns. The actual stock returns were obtained from the CRSP databases. The

estimated stock returns is what the stock price would have been in the absence of the

event and is determined by the stock price against a market index such as the CRSP

value weighted index or the Standard & Poor's 500 on the event day (Becker-Olsen,

2003). In this study, the CRSP value weighted index was used as a market index. By

calculating the difference between the actual stock returns and the estimated stock

returns, the researcher is able to obtain the abnormal stock market returns, which are

assumed to reflect the stock market's reaction to the event. The first step is estimating a

market model for each company. The equation is as follows (McWilliams & Siegel,

1997): The rate of return on the share price of firm i on day t is expressed as

Rit = ai + pi Rmt + E it

where

Rit = the rate of return on the share price of firm i on day t,

Rmt= the rate of return on a market portfolio of stocks (such as the CRSP value

weighted index) on day t,

a = the intercept term,

3 = the systematic risk of stock i, and

E it = the error term, with E(E it) = 0.











Table 5-17. Mean abnormal returns (MAR) for event date in NASCAR (Sprint)
Mean
Abnormal Positive: Patell Generalized
Day N Return Negative Z Sign Z
-3 22 0.47% 11:11 0.547 0.192
-2 22 -0.05% 13:9 -0.031 1.046
-1 22 -0.01% 10:12 0.061 -0.234
0 22 0.34% 15:7> 1.181 1.900*
+1 22 0.46% 15:7> 1.216 1.900*
+2 22 0.36% 12:10 0.231 0.619
+3 22 0.27% 11:11 0.325 0.192
The symbols $,*,**, and *** denote statistical significance at the 0.10,
0.05, 0.01 and 0.001 levels, respectively, using a generic one-tail test.
The symbols (,< or ),> etc. correspond to $,* and show the direction
and generic one-tail significance of the generalized sign test.






Table 5-18. Mean cumulative abnormal return (MCAR) for event date in NASCAR (Sprint)
Mean
Cumulative Precision
Abnormal Weighted Positive: Patell Generalized
Days N Return CAAR Negative Z Sign Z
(0,+1) 22 0.80% 0.66% 14:8) 1.695* 1.473$
(0,+2) 22 1.15% 0.73% 16:6> 1.517$ 2.326**
(0,+3) 22 1.42% 0.82% 14:8) 1.477$ 1.473$
The symbols $,*,**, and *** denote statistical significance at the 0.10, 0.05, 0.01 and 0.001
levels, respectively, using a generic one-tail test. The symbols (,< or ),> etc. correspond
to $,* and show the direction and generic one-tail significance of the generalized sign test.









major assumptions that need to be addressed. The first assumption is that markets are

efficient. In an efficient market, stock prices correctly and quickly respond to every

single piece of information available to the public (Miyazaki & Morgan, 2001). That is,

public announcements or communication of events influencing future market value of

the company (e.g., announcements of mergers, new product information, and celebrity

endorsement) should be incorporated into investors' mind as a decision-making factor

(Miyazaki & Morgan, 2001). As a result, any information change assumed to be

important for the future value of the company results in significant changes in the stock

price. In other words, stock prices should represent the present value of the company

such as discounted future income, expenditures, and firm's strategy (Miyazaki &

Morgan, 2001). The second assumption is that the event is unanticipated to the public.

The market previously did not have any information on the event, and the investors

should gain the information for the first time from the announcement (McWilliams &

Siegel, 1997). Abnormal stock market return has been assumed to be the result of the

market response to the new information. When an event was anticipated or leaked to

the public and investors in advance of the official announcement, such information

leakage would make use of event study analysis problematic (McWilliams & Siegel,

1997). The third assumption is that researcher can isolate the effects of an event from

the effects of other events. In event study analysis, this assumption is perhaps the most

critical one (McWilliams & Siegel, 1997). This assumption theoretically allows

researcher to isolate market returns derived from sponsorship from those from other

promotions by eliminating the impact of other events (Calderon-Martinez et al., 2005).

That is, researcher needs to be careful of confounding effects during the event window.





























To my parents, Masanori and Sachiko Kudo, who gave me an opportunity to pursue a
master's degree abroad and an opportunity to grow as a person in a wonderful place,
the University of Florida, with incredibly great people. I am deeply indebted to them for
their continued support and unwavering faith in me.









clearly categorized into commercial sponsorship (Cornwell et al., 2005a; Miyazaki &

Morgan, 2001; Pruitt et al., 2004; Spais & Filis, 2006) are consistent with the conclusion

led by Caldron-Martinez et al. (2005), in that the announcements of commercial

sponsorship were positively associated with the abnormal stock market returns. In this

point, since samples of title sponsorship used for this study obviously belong to

commercial sponsorship, abnormal stock market return would be expected as the result

of sponsorship announcements. Also, business behavior surrounding sport sponsorship

such as rapid growth of sponsorship market and increase of the sponsorship fee that

occurred during the past several decades implies that corporations find enough cost

and communication advantages in sport sponsorship over other traditional advertising

(Cornwell et al., 2005a). Based on these past findings, the following hypothesis is

presented:

Hypothesis 1: Announcements of title sponsorship in NASCAR, PGA, and LPGA

positively influence on stock price of sponsoring companies.

In addition to the above central hypothesis that would answer the most important

research question established for this study, it is important to explore and identify the

possible existence of moderators (variables) that may influence the result of

sponsorship (i.e., stock price changes). Therefore, several potential variables are taken

into account based on two perspectives; one originated from sponsoring company's

characteristics including Image Congruence, Corporate vs Brand Name, and Industry

Segment (e.g., high technology, banking, food, etc), and another derived from

sponsored event's characteristics. In this study, the researcher use event prestige as

sponsored event's characteristics and for the purpose of this study, event prestige is









2009, the PGA Tour decided to retain the steady fees for their title sponsorship rights

(Show, 2009). In fact, the president of one golf course that lost their title sponsor was

optimistic and said immediately after the title sponsor's withdrawal announcement that

"we are 100% confident that we'll have a new title sponsor in place for the January 2010

golf tournament" and "we've got five or six great (corporate) names we're talking to, all

of which you'd recognize" (Ferguson, 2009).

However, the situation surrounding the PGA Tour has changed due to a severe

economic recession in 2009. The PGA lost their four title sponsors in a year; two

tournaments supported by Buick, one by the U.S. Bank, and one by Stanford Financial.

Of those, the 50-year relationship with Buick was the oldest and the most established of

the PGA Tour title sponsorships. Also, the CEO of Deutsche Bank Americas, the top

manager sponsoring the Deutsche Bank Championship, criticized the PGA Tour's

annual rising purses saying that "there's pressure on the PGA Tour to reduce the

purses it will demand sponsors to fund in the coming years" (Mell, 2009). Also, a sport

critique noted that even though "no PGA Tour event can exist much longer without a

title sponsor," the seven million-plus price tag for just "operating a tournament today is

simply too much for a local, nonprofit entity to raise" (Harig, 2009). These situations

clearly suggest that the importance of measurement of title sponsorship effectiveness

and the research results should be used as a reference to determine the fair price of

title sponsorship.

Measurement of Sponsorship Effectiveness

In general, academic marketing research has been accused of its insufficient

attention with respect to sponsorship regardless of its market size, and particularly

measuring the effectiveness of sponsorship (Spais & Filis, 2008). However, evaluating









TABLE OF CONTENTS

page

ACKNOWLEDGMENTS ............. ..................... ............... 4

LIST O F TA BLES ......... ...... ...................................................................... 7

CHAPTER

1 INTRODUCTION ........................... .......... ......... 11

Purpose of the Study .......................... ........ ......... 15
Significance of the Study .............. .... .............. ........... ............ ............ 15
Lim stations of the Study ......... .......... ......... ............... ....... ........ 16
Definitions of Terms .......................... ......... ......... 16

2 REVIEW OF LITERATURE .............. ............ ..... ............... 18

Title Sponsorship ................ ......... ................. 18
Measurement of Sponsorship Effectiveness................................. 20
Event Study Analysis ................ ... ......... .. .. .. ......... 23
Sponsorship Effectiveness Using Event Study Analysis ............. .............. 27

3 THEORETICAL BACKGROUND AND HYPOTHESES DEVELOPMENT.............. 28

Hypotheses Development........................ ..... .............................. 28
M ain H hypothesis (H 1) ....................... .................. ..... ......... .............. 28
Im age Congruence (H2) .................................. ........................ .............. 30
C corporate vs Brand Nam e (H 3) ................................................. ........... ...... 31
Prestige of Event (H4) (H5) ....................... ..... ............. ................. 32
Event Competition Level (Sprint/Nationwide) (H6)................... .................... 33
Industry Segment (H7) ..................... ........................................ 34

4 M E T H O D .............. ..... ............ ................. ............................................... 3 6

Data Collection Procedures and Description of the Data .............. ..... ....... 36
Announcement Date.............................................. ............... 36
Event Date................................................ ........ 37
Possible M oderators......................... ...... .......... ............... 38
Im ag e co ng rue nce (H 2).......................... ....... ................................... 3 8
Corporate vs brand name (H3) and event competition level
(Sprint/Nationwide)(H6)............. ............... ............. 39
Am ount of purse (H4) and traditionality (Hs) ............................................. 39
Industry segment (H7)...................... ............... 39
D ata A analysis P rocedures............................................................................. 40
Event Study Analysis ....... ......................... .................... 40
Multiple Regression Analysis and ANOVA.............. ...... .................. 42









CHAPTER 6
DISCUSSION AND LIMITATION

Discussion

Title sponsorship has been increasingly employed as a marketing communication

tool, but its effectiveness and returns are extremely difficult to measure. Likewise, the

literature has not examined the impact of title sponsorship on the company's

performance regardless of the market size. This study focused on the performance of

title sponsorship in the PGA, LGPA, and NASCAR, by analyzing stock investors'

response on corporate sponsor's title sponsorship announcement. Also, this study

investigated the company performance of title sponsorship in sporting event date

following the new title sponsorship announcement as well as the announcement date,

for the first time in the sports related literature.

The main conclusion of the study is that title sponsorships generally appear to

neither positively nor negatively affect on the company's stock prices. However, in event

study analysis, it must be emphasized that no significant reaction on stock price does

not mean that the company has not received enough benefit or suffered a net loss by

the title sponsorship. Rather, it should mean that market assumes that the company has

received financial benefits from the relationship with the event exactly for the same

amount as the cost they paid for the sponsorship fee. In this respect, as long as the

results in event study are neutral, the title sponsorship should be accepted and the

decision whether entering title sponsorship or other alternative communication tools

would be left to the extent of marketing preference of the company. Also, as

demonstrated by the event study conducted for each sport entity respectively, not all

title sponsorships are equal. As past studies have demonstrated, market has favorably









Hypothesis 4: In the PGA and LPGA title sponsorship, amount of purse in the

tournament is likely to associate positively with the stock market returns.

Hypothesis 5: In the PGA and LPGA title sponsorship, traditionality of the

tournament (i.e., the length of the tournament having been held) is likely to

associate positively with the stock market returns.

Event Competition Level (Sprint/Nationwide) (H6)

In the NASCAR title sponsorship, in an effort to evaluate the effects of differences

in competition level of event on sponsorship returns, Event Competition Level was

considered as a potential variable. Event Competition Level refers to whether the

NASCAR event belongs to the Sprint Cup series or the Nationwide series. In the

NASCAR event, the Sprint Cup series is the highest level of competition and the

Nationwide is the second level of competition. The study of Pruitt et al. (2004) found that

sponsorship announcement with a highly successful team, which was determined by the

acquired winning points in last year's races, is more positively received in stock market

than that with a lower-tier team. That is, successful teams in the NASCAR are more

likely to have abnormal stock market returns (Pruitt et al., 2004). The rationale behind

this is that when considering the sphere of television exposure and other media

coverage, successful teams are much more likely to gain such benefits. In this respect,

title sponsor of higher level competition (i.e., Sprint Cup series) are more likely to have

abnormal stock market returns than are that of lower level competition (i.e., Nationwide

series) because they are likely to gain broader media coverage. Also, higher level

competition is considered as more prestigious event in a general sense; thus,

sponsorship responses are likely to result in a more positive way (Speed & Thompson,

2000). Therefore,









sponsors respectively (See Table 5-7, Table 5-8, Table 5-9, Table 5-10). As noted, the

results examined each level individually showed that stock market are only favorable for

title sponsors in the higher level event (i.e., Sprint series sponsors), while lower level

sponsors (i.e., Nationwide series sponsors) are seen as neutral. Therefore, this result

reports the level of competition in NASCAR matters in gaining positive stock market

returns, though statistic test comparing those did not show the significance.

ANOVA for Industry Segment

For the examination of whether Industry Segment (H7) in title sponsors is the

determinant of the abnormal stock market returns, ANOVA test was carried out among

3 industry segments including High-technology, Retails, and Services.

Table 5-22 presents the result of ANOVA test. As shown in the Table, there were

not any significant differences among the selected business segments (p>0.05,

F=2.667, M = .0046 for High-technology, M = .0031 for Retails, and M = -.0066 for

Services). However, since past research found statistical significance for high-

technology company by regression analysis by coding dummy variable 1 for High-

technology company and otherwise 0 (Cornwell et al., 2001; Cornwell et al., 2005a;

Pruitt et al., 2004), the researcher conducted a regression analysis following their

method. For the purpose of this analysis, the dummy variable High-tech was coded as 1

(N=19 which refers to the first categorization before the grouping for the ANOVA), and

otherwise 0 (N=65 including all other segments besides high-tech companies in the first

grouping). Table 7-a shows the result of the regression analysis. The test found the

significance at 5% level in the relationship between high-technology companies and

their positive abnormal stock market returns (p<0.05, F=5.135, R2= 0.059, Adjusted R2=

0.047). Therefore, H7 was partially confirmed.











Table 5-15. Mean abnormal returns (MAR) for event date in LPGA
Mean
Abnormal Positive: Patell
Day N Return Negative Z


-3 9
-2 9
-1 9
0 9


-0.335
0.511
0.295
0.155
-0.30,
1.025
-0.26'


4:5
4:5
4:5
5:4
3:6
7:2>
4:5


-0.403
2.117*
0.535
-0.658
-0.567
2.469**
-0.991


Generalized
Sign Z
-0.293
-0.293
-0.293
0.374
-0.960
1.707*
-0.293


The symbols $,*,**, and *** denote statistical significance at the 0.10,
0.05, 0.01 and 0.001 levels, respectively, using a generic one-tail test.
The symbols (,< or ),> etc. correspond to $,* and show the direction
and generic one-tail significance of the generalized sign test.


Table 5-1




Days
(0,+1)
(0,+2)
(0,+3)


6. Mean cumulative abnormal return (MCAR) for event date in LPGA
Mean
Cumulative Precision
Abnormal Weighted Positive: Patell Gen
N Return CAAR Negative Z
9 -0.16% -0.50% 5:4 -0.866
9 0.86% 0.51% 7:2> 0.718
9 0.60% 0.10% 6:3 0.126
The symbols $,*,**, and *** denote statistical significance at the 0.10,
0.05, 0.01 and 0.001 levels, respectively, using a generic one-tail test.
The symbols (,< or ),> etc. correspond to $,* and show the direction
and generic one-tail significance of the generalized sign test.


eralized
Sign Z
0.374
1.707*
1.041









defined by three elements of the event; (a) Amount of Purse, (b) Traditionality, and (c)

Event Competition Level (Sprint/Nationwide). These potential variables are discussed

along with the hypotheses in the following sections.

Image Congruence (H2)

Image Congruence refers to a perceived image fit between the sponsoring

company and the sponsored sport event. Numerous studies highlighted the importance

of the degree of congruence found in the relationship between sponsor and sponsored

event in predicting consumers' response to sponsorship (Cornwell, Weeks, & Roy,

2005; McDaniel, 1999). For example, Gwinner and Eaton (1999) concluded that when

event and sponsor brand are matched on either an image or functional basis, the image

transfer process, which are most likely sought by sponsoring company as the primary

purpose of the sponsorship, is enhanced. According to Cornwell et al. (2005b),

sponsors and products that have been perceived as being closely related to the event

by the consumers have a number of advantages over unrelated sponsors. In fact,

McDaniel (1999) stated that the nature of linkage or congruence found in the

relationship between sponsor and event is an important determinant for successful

sponsorship activity. Specifically, in the literature of event study analysis, Cornwell et al.

(2001) demonstrated that the NASCAR sponsors whose business segment has a direct

tie to automotive industry experienced 3 percent greater increases in their stock prices

on an average than those do not have the congruence. Also, Cornwell et al. (2005a)

showed that the Major League Baseball (MLB), National Basketball Association (NBA),

National Hockey League (NHL), National Football League (NFL), and PGA sponsors

that had congruence with their sponsored sports increased their stock prices about 11%









NASCAR (Sprint series)

Table 5-17 and Table 5-18 show MAR and MCAR for the NASCAR Sprint series

title sponsors. As expected, on the event 0 day (in most samples, it is Monday which is

the first market available day after the NASCAR event) observed significant abnormal

stock market returns at 10% level. Also, all MCAR windows report that the title sponsors

are receiving significant stock price increases after the event date including statistical

significance 5% level on the (0, +2) window. These results are consistent with those in

the announcement date and confirmed that stock market has seen being title sponsor

for the NASCAR Sprint series brings financially net surplus for the company.

NASCAR (Nationwide series)

Table 5-19 and Table 5-20 show MAR and MCAR for the NASCAR Nationwide

series title sponsors. Table 5-19 reports mixed results that are event 0 day with

significant positive stock price increases and +3 day with significant negative stock price

decreases. As noted earlier, when considering that event window nearer to t = 0 day

would better reflect the market's reaction toward the event than do further day from the

0 day because of the controlling problem of the confounding effects. (McWilliams &

Siegel, 1997), the positive results recorded on the 0 day should be seen as primary

stock market reaction toward the event. Then, this result is interesting because title

sponsors in the NASCAR Nationwide series received neither positive nor negative

returns from the sponsorship announcement (Table 2-d & Table 3-d). That is, the result

is not consistent with those in the announcement date and demonstrating that stock

market positively responds only for the event day, not for the announcement day for the

NASCAR Nationwide title sponsors.









Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Science

THE MARKET VALUE OF SPORT SPONSORSHIP:
THE INFLUENCE OF TITLE SPONSORSHIP ON SPONSORING COMPANY'S
STOCK PRICES

By

Masaki Kudo

August 2010

Chair: Yong Jae Ko
Major: Sport Management

Title sponsorship has been increasingly employed as a marketing communication

tool, but its effectiveness and returns are extremely difficult to measure. Likewise, the

literature has not examined the impact of title sponsorship on the company's

performance regardless of the market size. The objective of this study is to analyze the

impact of title sponsorship in the PGA, LGPA, and NASCAR, by examining the

sponsor's stock price fluctuations on both the title sponsorship announcement date and

the following title sponsorship event date, and subsequently to explore the possible

moderators in the stock price changes.

The event study analysis conducted for each sport respectively demonstrates that

title sponsors in the LPGA and NASCAR obtained significant stock increases on both

the announcement date and event date. On the other hand, title sponsors in the PGA

experienced significant stock price decreases on the both dates. That is, the stock

market sees that title sponsorships for the PGA are overvalued and those of the LPGA

and NASCAR are undervalued.









study analysis using the CRSP databases and the SAS statistics software. Specific

procedure was followed the EVENTUS study guide manual 8.0 (Cowan, 2007).

Multiple Regression Analysis and ANOVA

To examine H2 through H7, multiple regression analysis and ANOVA test were

executed. In this stages, the cumulative abnormal return (CAR) (i.e., stock price

change) recorded by each title sponsor over the event windows t = 0 to t = +1 served as

the dependent variable, while selected possible moderators served as the independent

variables. For these tests, the SPSS 18.0 program was used.











Table 5-1. Mean abnormal returns (MAR) for the full 84 announcements in PGA, LPGA, and NASCAR
Mean
Abnormal Positive: Patell Generalized
Day N Return Negative Z Sign Z
-3 84 0.14% 46:38 0.809 0.969
-2 84 0.22% 36:48 0.135 -1.213
-1 84 -0.06% 40:44 -0.793 -0.340
0 84 0.24% 42:42 0.467 0.096
+1 84 -0.06% 40:44 -0.854 -0.340
+2 84 -0.11% 36:48 -0.844 -1.213
+3 84 -0.03% 45:39 -0.469 0.751
The symbols $,*,**, and *** denote statistical significance at the 0.10,
0.05, 0.01 and 0.001 levels, respectively, using a generic one-tail test.
The symbols (,< or ),> etc. correspond to $,* and show the direction
and generic one-tail significance of the generalized sign test.


Table 5-2.


Mean cumulative abnormal return (MCAR) for the full 84 announcements in PGA, LPGA, and NASCAR
Mean
Cumulative Precision


Abnormal Weighted Positive: Patell Generalized
Days N Return CAAR Negative Z Sign Z
(0,+1) 84 0.18% -0.08% 41:43 -0.273 -0.122
(0,+2) 84 0.07% -0.24% 43:41 -0.711 0.314
(0,+3) 84 0.04% -0.34% 37:47 -0.850 -0.995
The symbols $,*,**, and *** denote statistical significance at the 0.10, 0.05, 0.01 and 0.001
levels, respectively, using a generic one-tail test. The symbols (,< or ),> etc. correspond
to $,* and show the direction and generic one-tail significance of the generalized sign test.















Table 4-3. NASCAR Sprint series title sponsorship sample
Sport Company Name Title Sponsorship Name


NASCAR (Sprint)
NASCAR (Sprint)
NASCAR (Sprint)
NASCAR (Sprint)
NASCAR (Sprint)
NASCAR (Sprint)
NASCAR (Sprint)
NASCAR (Sprint)
NASCAR (Sprint)
NASCAR (Sprint)
NASCAR (Sprint)
NASCAR (Sprint)
NASCAR (Sprint)
NASCAR (Sprint)
NASCAR (Sprint)
NASCAR (Sprint)
NASCAR (Sprint)
NASCAR (Sprint)
NASCAR (Sprint)
NASCAR (Sprint)
NASCAR (Sprint)
NASCAR (Sprint)


DAIMLERCHRYSLER
HARRAH'S ENTERTAINMENT
PEPSI
AARON RENTS
SIEMENS
MBNA
FORD MOTOR
ALLSTATE
USG
BANK OF AMERICA
SONY
TOYOTA MOTOR
DIRECT GROUP
DIAGEO
NEWELL RUBBERMAID
ADVANCED MICRO DEVICES
3M
LOWES COMPANIES
CITIZENS BANKING
COCA COLA
PEPSI
SUNOCO


UAW-Daimler Chrysler 400
Harrah's 500
Tropicana 400
Aaron's 499
Sylvania 300
Bass Pro Shops MBNA 500
UAW-Ford 500
Allstate 400 at the Brickyard
USG Sheetrock 400
Bank of America 500
Sony HD 500
Toyota/Save Mart 350
DirecTV 500
Crown Royal 400
Lenox Industrial Tools 300
AMD at the Glen
3M Performance 400
Kobalt Tools 500
Citizens Bank 400
Coke Zero 400 presented by Coca-Cola
Pepsi 500
Sunoco Red Cross Pennsylvania 500


Announcement Date
10/19/2000
11/17/2000
3/7/2001
4/8/2002
1/30/2003
2/4/2003
2/22/2005
4/28/2005
5/16/2005
6/29/2005
9/1/2005
1/23/2006
3/7/2006
4/17/2006
6/22/2006
8/3/2006
1/31/2007
3/5/2007
4/9/2007
12/7/2007
2/19/2008
7/2/2008


Event Date
3/5/2001
4/2/2001
7/16/2001
4/22/2002
9/15/2003
3/10/2003
10/3/2005
8/8/2005
7/11/2005
10/16/2006
9/6/2005
6/25/2007
4/3/2006
5/8/2006
7/17/2006
8/14/2006
8/21/2007
3/19/2007
6/18/2007
7/7/2008
9/2/2008
8/4/2008













4-1. PGA title sponsorship sample


Table
Sport
PGA
PGA
PGA
PGA
PGA
PGA
PGA
PGA
PGA
PGA
PGA
PGA
PGA
PGA
PGA
PGA
PGA
PGA
PGA
PGA
PGA
PGA
PGA


Company name
SEI INVESTMENTS
VERIZON COMMUNICATIONS
ELECTRONIC DATA SYSTEM
WACHOVIA
COCA COLA
BANK OF AMERICA
FORD MOTOR
DEUTSCHE BANK
FRIEDMAN BILLINGS RAMSEY GRP
GENERAL MOTORS
FRIEDMAN BILLINGS RAMSEY GRP
ICOS
LILLY ELI
U.S. BANK
BARCLAYS
ST PAUL TRAVELERS
INTERCONTINENTAL HOTELS GRP
AT&T
BANCO POPULAR
WAL MART STORES
NORTHERN TRUST
ROYAL BANK CANADA
HEWLETT PACKARD


Title sponsorship name
SEI Pennsylvania Classic
Genuity Championship
EDS Byron Nelson Championship
Wachovia Championship
The Tour Championship presented by Coca-Cola
Bank of America Colonial
Ford Championship at Doral
Deutsche Bank Championship
FBR Capital Open
Buick Championship
FBR Open
Cialis Western Open
Cialis Western Open
U.S. Bank in Milwaukee
Barclays Classic
Travelers Championship
Crowne Plaza Invitational at Colonial
AT&T National
Puerto Rico Open presented by Banco Popular
Children's Miracle Network Classic presented by Wal-Mart
Northern Trust Open
RBC Canadian Open
HP Byron Nelson Championship


*1 Eventus program could retrieve stock price data before 12/31/2008 and the event was played in March 2009


Announcement date
1/10/2000
11/20/2000
4/12/2001
5/6/2002
5/11/2002
5/20/2002
9/16/2002
11/14/2002
3/19/2003
7/29/2003
10/8/2003
1/23/2004
1/23/2004
3/31/2004
10/12/2004
4/11/2006
7/25/2006
3/7/2007
4/9/2007
7/24/2007
10/15/2007
11/1/2007
10/2/2008


Event date
9/14/2000
3/1/2001
5/15/2003
5/8/2003
10/31/2002
5/22/2003
3/6/2003
8/29/2003
6/5/2003
2/12/2004
1/29/2004
7/1/2004
7/1/2004
7/22/2004
6/23/2005
6/21/2007
5/24/2007
7/5/2007
3/20/2008
11/1/2007
2/14/2008
7/24/2008
N/A*1









higher than sponsors that did not have linkage with the sponsored sports. These past

findings suggest following:

Hypothesis 2: Image congruence between sport event and title sponsors is likely

to have positive association with the stock price change.

Corporate vs Brand Name (H3)

Corporate vs Brand Name refers to whether the name of the title sponsorship is

accompanied with the main corporate name or the name of the title sponsor is

accompanied with the corporate divisional brand name. The study of Thjomoe, Olson,

and Bronn (2002) that analyzed decision making process regarding sponsorship

activities in 400 Norwegian firms demonstrated that more professional company in

sponsorship activities tend to use sponsorship to improve their whole corporate image,

not a particular product when compared with a less professional company. Pruitt et al.

(2004) supported this result in the NASAR sponsors by event study analysis. They

found that stock market investors have more favorably seen sponsorship undertaken in

the name of an entire corporation than sponsorship tagged along with single product or

division of a corporation. The rationale behind this is that the NASCAR fans who have

been known as having extraordinary loyalty for the teams and sponsoring corporations

are more likely to be influenced in a positive way for sponsoring corporations by

corporate-level sponsorship that offers the corporation more opportunities to link fan's

support across a corporation's whole products and services rather than single

product/service-level sponsorship where the benefits only concentrate the

product/service (Pruitt et al., 2004). Also, sponsorship at the corporate-level is likely to

offer more chances of contacts for positive sponsor fit comparing to sponsorship at the

brand-level (Pruitt et al., 2004). Because brand-level sponsorship is highly specialized in









Industry Segment: Industry segment refers to the industrial sector to which the

title sponsor corporation belongs.

Amount of Purse: Amount of purse refers to the total amount of purse the event

raised for the tournament.

Traditionality: Traditionality refers to the length (year) of the tournament having

been held.

Event Competition Level (Sprint/Nationwide): Event competition level refers to

whether the NASCAR event belongs to the Sprint Cup series or the Nationwide series.

Specifically, this thesis has been organized as follows: first, a review of the

literature with respect to title sponsorship and measurement of sponsorship

effectiveness is presented, along with the challenges in measurement of sponsorship

effectiveness. Also, in the literature review, the method of Event Study Analysis is

introduced and previous research using this method is reviewed by focusing on the

advantages and disadvantages of this methodology; Second, the research hypotheses

are presented with the theoretical backgrounds based on previous research; Third, the

data collection method and data analysis procedures are presented; Fourth, the

empirical results obtained in the study are described; Finally, a discussion is presented

by focusing on research and managerial implications, main conclusions, and limitations

of this study.









The examination of possible moderators demonstrates that high image

congruence between title sponsor and sponsored sport event and whether or not they

are high-technology companies are the key determinants in stock price increases. Also,

the difference of NASCAR event competition level (i.e., Sprint series or Nationwide

series) plays an important role in the stock price changes, and more prestigious event

(i.e., Sprint series) sponsors are favorably seen by the market.









LIST OF REFERENCES


Becker-Olsen, K. (2003). Questioning the name game: An event study analysis of
stadium naming rights sponsorship announcements. International Journal of
Sports Marketing & Sponsorship, 5(3), 181-192.

Brown, S., & Warner, J. (1985). Using daily stock returns: The case of event studies.
Journal of Financial Economics, 14, 3-31.

Calderon-Martinez, A., Mas-Ruiz, F. J., & Nicolau-Gonzalbez, J. L. (2005). Commercial
and philanthropic sponsorship: Direct and interaction effects on company
performance. International Journal of Market Research, 47(1), 75-94.

Clark, J. M., Cornwell, T. B., & Pruitt, S. W. (2002). Corporate stadium sponsorship,
signaling theory, agency conflicts and shareholder wealth. Journal of Advertising
Research, 42(6), 16 32.

Copeland, R., Frisby, W., & McCarville, R. (1996). Understanding the sport sponsorship
process from a corporate perspective. Journal of Sport Management, 10(1), 32-48.

Cornwell, T. B., & Maigan, I. (1998). An international review of sponsorship research,
Journal of Advertising. 27(1), 1-22.

Cornwell, T. B., Pruitt, S. W., & Van Ness, R. (2001). An exploratory analysis of the value
of winning in motorsports: Sponsorship-linked marketing and shareholder wealth.
Journal of Advertising Research, 41, 17-31.

Cornwell, T. B., Pruitt, S. W., & Clark, J. M. (2005a). The Relationship between major-
league sports' official sponsorship announcements and the stock prices of
sponsoring firms. Journal of the Academy of Marketing Science, 33(4), 401-412.

Cornwell, T. B., Weeks, C., & Roy, D. P. (2005b). Sponsorship-linked marketing: opening
the blackbox. Journal of Advertising, 34(2), 23-45.

Cowan, A. R. (2007). Eventus 8.0 User's Guide, Iowa State University Press, Ames IA:
Cowan Research.

Crompton, J. (2004) Conceptualization and alternate operationalizations of the
measurement of sponsorship effectiveness in sport. Leisure Studies, 23(3), 267-
281.

Farell, K. A., & Frame, W. S. (1997). The value of Olympic sponsorships: Who is
capturing the gold? Journal of Market Focused Management, 2(2), 171-182.

Ferguson, D. (2009). Buick ends title sponsorships with all PGA Tour events, including
Buick Invitational. Retrieved October 15, 2009 from Sports News:
http://blog.taragana.com/sports/2009/08/04/buick-ends-title-sponsorships-with-all-
pga-tour-events-including-buick-invitational-17579/









Meenaghan, J. (1991). The role of sponsorship in the marketing communications
mix. International Journal of Advertising, 10, 35-47.

Mell, R. (2009). Deutsche bank CEO scrutinized PGA Tour purses. Retrieved October
25, 2009 from:http://www.thegolfchannel.com/tour-insider/deutsche-bank-ceo-
scrutinizes-pga-tour-purses-31959/

Miyazaki, A.D., & Morgan, A.G. (2001). Assessing the market value of event sponsorship.
Journal of Advertising Research, 41(1), 9-16.

Mullin, B., Hardy, S., & Sutton, W. (2000). Sport Marketing. Champaign. Illinois / Leeds,
UK: Human Kinetics

Oztur, M.A., Kozub, F.M., & Kocak, S. (2004). Impact of sponsorship on companies that
supported the 2002 Salt Lake City winter Paralympics. International Journal of
Sports Marketing & Sponsorship, 5(4), 282-295.

Pandey, V. K., Shanahan, K. J., & Hansen, S. W. (2005). The relationship between
shareholder wealth effects, diversity, and publicity as a marketing strategy. Journal
of the Academy of Marketing Science, 33(4), 423-431.

Pham, M. (1991). Evaluation of sponsorship effectiveness; a model and some
methodological considerations. Gestion 2000. 47-64.

Poknywczynski, J. (2000). Sport sponsors know what they want. Sports Business
Journal, 5(11), 2

Pruitt, S. W., Cornwell, T. B., & Clark, J.M. (2004). The NASCAR phenomenon: auto
racing sponsorship and shareholder wealth. Journal of Advertising Research,
44(3), 281-296.

Roy, D. P., & Cornwell, T. B. (1999). Managers' use of sponsorship in building brands:
service and product firms contrasted. International Journal of Sports Marketing and
Sponsorship, 1(6), 345-360.

Russell, M., Mahar, J., & Drewniak, B. (2005). Examination of stock market responses to
publicity surrounding athletic endorsers. The Marketing Management Journal,
15(2), 67-79.

Samitas, A., Kenourgios, D., & Zounis, P. (2008). Athens' Olympic games 2004 impact of
sponsor's stock returns, applied Financial Economics, 18(19), 1569-1580.

Sander, D.M., & Shani, D. (1993). Sponsorship and the Olympic games: the consumer
perspective. Sport Marketing Quarterly, 2(3), 38-43.

Show, J. (2009). Prices steady for golf title sponsors. Street & Smith's Sportsbusiness
Journal, 12(6), 4.









In summary, LPGA and NASCAR Sprint sponsors received significant stock

market returns after the title sponsorship event, while PGA sponsors experienced

significantly negative decreases after the event date. These results are consistent with

those they received from the announcement date. However, the result of NASCAR

Nationwide sponsors is not consistent with those in the announcement date in that only

event date produced significantly positive stock market returns. The different results

between announcement date and event date in the NASCAR Nationwide sponsors

suggest that research focusing on not only announcement date but also event date is

necessary in order to fully measure the impact of title sponsorship on stock prices.

Multiple Regression and ANOVA Test

The second purpose of this study is to explore the existence of possible

moderators in title sponsors' stock price changes following the announcement by

employing the results obtained above event studies. For the purse of this examination,

multiple regression analysis and ANOVA test were carried out as follows. As noted,

CAR window (0, +1) in the event studies for announcement date served as the

dependent variable, while hypothesized possible moderators served as the independent

variables.

Multiple Regressions

For the examination of whether Image Congruence (H2), Corporate vs Brand

Name (H3) and prestige of event including Amount of Purse (H4), Traditionality (H5),

and Event Competition Level (Sprint/Nationwide) (H6) were significant predictors of the

abnormal stock market returns, multiple regression analysis for such moderators was

carried out. As noted earlier, Image Congruence (H2) and Corporate vs Brand Name

(H3) were examined for all PGA, LPGA, and NASCAR title sponsors, Amount of Purse









LIST OF TABLES


Table page

4-1 P G A title sponsorship sam ple.................................................. .... .. ............... 43

4-2 LPGA title sponsorship sam ple.............................. ............................ ..... 44

4-3 NASCAR Sprint series title sponsorship sample .............. ....... .... ............. 45

4-4 NASCAR Nationwide series title sponsorship sample ........... ...... ........ 46

5-1 Mean abnormal returns (MAR) for the full 84 announcements in PGA, LPGA,
and NASCAR......................................... ........... 57

5-2 Mean cumulative abnormal return (MCAR) for the full 84 announcements in
PG A LPG A and NA SCA R ........................................ .......................... 57

5-3 Mean abnormal returns (MAR) for announcement date in PGA...................... 58

5-4 Mean cumulative abnormal return (MCAR) for announcement date in PGA...... 58

5-5 Mean abnormal returns (MAR) for announcement date in LPGA....................... 59

5-6 Mean cumulative abnormal return (MCAR) for announcement date in LPGA.... 59

5-7 Mean abnormal returns (MAR) for announcement date in NASCAR (Sprint)..... 60

5-8 Mean cumulative abnormal return (MCAR) for announcement date in
NASCAR (Sprint) ............................. .............................. .............. 60

5-9 Mean abnormal returns (MAR) for announcement date in NASCAR
(Nationw ide) ............... ... ... ... .. ..................... 61

5-10 Mean cumulative abnormal return (MCAR) for announcement date in
NA SCA R (Nationw ide) ....................................... .... .......... .............. 61

5-11 Mean abnormal returns (MAR) for the full 81 event date in PGA, LPGA, and
NASCA R .................. .............. ............... .......... .......... 62

5-12 Mean cumulative abnormal return (MCAR) for the full 81 event date in PGA,
LPGA, and NASCAR ............ ......... ........ ........................ 62

5-13 Mean abnormal returns (MAR) for event date in PGA .................. ............ .. 63

5-14 Mean cumulative abnormal return (MCAR) for event date in PGA.................... 63

5-15 Mean abnormal returns (MAR) for event date in LPGA................. ............... 64









Maigan (1998) stated that demonstrating the market value of sponsorship is apparently

the best way to legitimize sponsorship investment for sponsoring company. To date,

however, only few studies have been conducted to analyze the impact of sponsorship

on the company's actual financial performance (Calderon-Martinez, Mas-Ruiz, &

Nicolau-Gonzalbez, 2005).

In light of the increasing need for a systematic measurement tool for sponsorship

effectiveness on the corporations' performance level, several scholars have attempted

to examine the impacts of sport sponsorship by focusing on stock price changes

(Cornwell, Pruitt, & Van Ness, 2001; Cornwell, Pruitt, & Clark, 2005; Farell & Frame,

1997; Samitas, Kenourgios, & Zounis, 2008). This methodology, a quantitative analysis

of the stock price fluctuations recorded at the time of initiation of the programs (i.e., the

time of sponsorship announcements), is called as Event Study Analysis, which has

been widely accepted in the field of finance, accounting, marketing, and management

(Pruitt, Cornwell, & Clark, 2004). While event study analysis is only recently to be

introduced to marketing research field (Ozur et al., 2004), several studies employed this

methodology in sport sponsorship contexts such as the Olympics, NASCAR, stadium

naming right, major league official products, athletic endorsement, and European soccer

teams. None of them, however, has addressed the value of title sponsorship in

professional sport event contexts. Also, most of the past research that adopted event

study analysis method only focused on a single sport event rather than examining the

relative impacts of sport sponsorship across multiple sports. Furthermore, most of the

past research has exclusively focused on the announcement day of the new

sponsorship agreement while the actual sporting event day following the new









CHAPTER 4
METHOD

Data Collection Procedures and Description of the Data

Announcement Date

The primary data used in this study are daily stock returns of companies' having

announced title sponsorship agreements in the PGA, LPGA, and NASCAR. The initial

title sponsorship lists for the PGA, LPGA, and NASCAR were drawn from the Web

pages of each sport organization during 2009. In this study, renewal announcements of

title sponsorship were not included; thus, only title sponsors that entered a title

sponsorship for the sporting event for the first time were used for this study. The PGA,

LPGA, and NASCAR were selected for this study because they are some of the most

popular professional sports in the United States, and other popular sports, for example

the MLB, NFL, NBA, and NHL have rarely given title sponsorship to their games;

instead, the PGA, LPGA, and NASCAR chosen for this study have virtually put a title

sponsor name on every single game they have held. In this study, the researcher

included only title sponsors that have publicly-traded common stock on either the

NYSE, AMEX, or the NASDAQ exchange at the time of the sponsorship announcement.

As a result, both foreign and privately-held domestic companies were excluded from the

analysis. Also, from the obtained data, closing prices were picked to use in the data

analysis. The volume and highest-lowest prices were not used. The daily stock data

were obtained from the University of Chicago's Center for Research in Security Prices

(CRSP), which is well known as the most comprehensive database of stock price,

return, and volume data for the NYSE, AMEX and NASDAQ stock markets.









BIOGRAPHICAL SKETCH

Masaki Kudo received his Bachelor of Arts in sports science from Waseda

University, Japan in 2007. He then attended the University of Florida where he

completed his Master of Science in sport management in 2010. His major research

interests are measuring effectiveness of sport sponsorship, cultural difference in sport

sponsorship, and sport sponsorship as corporate social responsibility (CSR).









be of interest to both researchers and practitioners and be used for the further

development of sport sponsorship.

Limitations of the Study

The main limitations of this study are the following three: (1) event study analysis,

(2) the selection of the companies (i.e., title sponsors), and (3) the determination of the

congruence level between title sponsor and sport event. These are discussed in detail

in the discussion section

Definitions of Terms

Title Sponsorship: Title sponsorship is defined as the acquisition of rights to take

part in the official name of the event for the purpose of deriving benefits related to that

name sharing.

Event Study Analysis: Event study analysis is an analysis of "whether there was

a statistically significant reaction in financial markets to past occurrences of a given type

of event that is hypothesised to affect public organizations' market values" (Spais &

Filis, 2008, p. 173).

Abnormal Stock Market Return: Abnormal stock market returns refers to returns

not consistent with the pattern of change as established by past company and market

activity.

Image Congruence: Image congruence refers to a perceived image fit between

the sponsoring company and the sponsored sport event.

Corporate vs Brand Name: Corporate vs brand name refers to whether the

name of the title sponsorship is accompanied with the main corporate name or the

name of the title sponsor is accompanied with the corporate divisional brand name.









hypothesised to affect public organizations' market values" (Spais & Filis, 2008, p. 173).

Although event study analysis has a long history and been widely employed as a

research tool in the finance and economics field (MacKinlay, 1997), it is only recently to

be introduced to marketing research (Ozturk et al., 2004). In management, event study

analysis has been used to judge the effects of corporate control changes, such as

mergers, earnings announcements, issues of new debt, CEO turnover (McWilliams &

Siegel, 1997). In fact, this method is known as the standard assessment metric for the

measurement of the net economic value of any corporate event whose initial

announcement dates are precisely available (Pruitt et al., 2004). One of the primary

reasons why event study analysis has become so popular in both management and

academic research is that this method obviates the need of "analysis" of accounting-

based measures (McWilliams & Siegel, 1997). Although it is possible that managers

show or select some favorable accounting-based figures and manipulate the actual

corporate performance, stock prices are not as subject to such manipulation by insiders

(McWilliams & Siegel, 1997). Therefore, stock prices are supposed to reflect the true

value of a corporation (McWilliams & Siegel, 1997) and represent the corporate

performance in a better way than does the accounting sheet (Calderon-Martinez et al.,

2005).

While event study analysis can be a powerful technique, there are some

disadvantages in this method (Becker-Olsen, 2003; Spais & Filis, 2008). The

disadvantages of event study analysis are associated with the characteristics of this

method in that the validity of conclusions led by event study analysis highly depends on

its strong assumptions. The event study analysis is basically conducted under the three









the stock price changes around the event day too in order to fully measure the

effectiveness of title sponsorship in terms of stock price changes.

For the purpose of this study, the event day was set to the first day market

opened following the title sponsorship event day, otherwise the event was held on

weekday that market is opened as usual. That is, in the NASCAR whose events are

usually held on Saturday or Sunday, the event day was set to the next market available

day that is usually Monday (or sometimes Tuesday if the Monday was also market

closed day). In the PGA and LPGA, most of the event day was set to Thursday because

most tours officially start on Thursday. In such case, +2 day in the event windows

indicates Monday, which will be likely impacted on stock prices due to the great number

of title sponsorship exposure and recognition by Sunday's media coverage. That is, in

the NASCAR, some stock market reactions are likely to be observed on the event 0 day

if any, and in the PGA and LPGA, those are likely to be seen on the event +2 day if any.

Possible Moderators

In addition to the primary analysis on stock prices, several variables to be used

for exploring the possible moderators influencing on the stock prices were collected and

categorized as following:

Image congruence (H2)

To determine the level of Image Congruence between title sponsor and sponsored

event, the researcher employed questionnaire developed by Speed and Thompson

(2000), which comprised of 5 questions asking sponsor-event fit with a 7 Likert (see

Appendix A). After careful discussion with an expert, the questionnaire was answered in

all 84 samples and the average of 5 questions was each employed as the congruence

level between title sponsor and sponsored event for all 84 samples.

38









CHAPTER 3
THEORETICAL BACKGROUND AND HYPOTHESES DEVELOPMENT

Hypotheses Development

For the first time, this study aims at comparing the market value of title

sponsorship in professional sports, specifically in the PGA, LPGA, and NASCAR,

employing stock price fluctuations as a measurement by focusing on both sponsorship

announcement day and sporting event day, and subsequently, the possible existence of

moderators affecting the stock price fluctuations in title sponsorship is explored.

Specifically, whether the sponsoring company's characteristics such as congruence

between company and sponsored event, and the sponsored sport event's

characteristics such as prestige of the event influenced their stock prices are

investigated in this study in an effort to identify the possible moderators in title

sponsorship effectiveness.

Main Hypothesis (H1)

Previous researches of sponsorship effectiveness using event study analysis as

the measurement have produced mixed results including both investors' positive

perspectives ( Cornwell et al., 2001; Cornwell et al., 2005a, Miyazaki & Morgan, 2001)

and negative perspectives (Farell & Frame, 1997; Oztur et al., 2004) toward

corporations' sponsorship activities. For instance, the study of Caldron-Martinez et al.

(2005) demonstrated that while commercial sponsorship generates abnormal stock

market returns, philanthropic sponsorship does not bring such benefits. Thus, they

concluded that corporations trying to improve their performances in terms of stock price

value through sponsorship activities had better choose commercial sponsorship rather

than philanthropic sponsorship. The results of several studies whose study subjects are









million (from 9 companies) in the 1988 Seoul Olympics; $175 million (froml2

companies) in 1992 Barcelona; $350 million (from10 companies) in 1996 Atlanta; $550

million (from 11 companies) in 2000 Sydney; $650 million (from 11 companies) in 2004

Athens; and $866 million (from 12 companies) in 2008 Beijing (Crompton, 2004;

Giannoulakis, Stotlar & Chatziefstathiou, 2008). Besides this sponsorship revenue of

the IOC, the host country's Olympic Committee has also separately generated

significant amount of revenue from their sponsorship programs designed for local

businesses. In the Sydney Olympics, for example, the Sydney Organizing Committee

individually generated $492 million from their sponsorship programme (Crompton,

2004). That is to say, in Sydney Olympics, the total revenues that both IOC and the host

committee gained through the corporate sponsorships were over $1 billion.

To the contrary to this tremendous sponsorship growth, it has been known that

measuring sponsorship effectiveness is very challenging task (Ko, Kim, Claussen &

Kim, 2008; Spais & Filis, 2008); as a result, research with respect to sponsorship

effectiveness is still in the infancy stage (Oztur, Kozub & Kocak, 2004). While interest in

evaluating the impact of sponsorship has increased (Copeland, Frisby, & McCarville,

1996; Crompton, 2004), most sponsoring corporations and properties do not actually

have established procedures and systems for measuring sponsorship effectiveness

(Miyazaki & Morgan, 2001). In reality, about 40% of companies engaged in major sport

sponsorship activity reported that they do not spend their budgets on research to

measure the impact of sponsorship (Crompton, 2004).

Also, traditional research in sponsorship measurement tends to only focus on

either media exposure or corporate image transfer (Oztur et al., 2004; Miyazaki &









Corporate vs brand name (H3) and event competition level (Sprint/Nationwide)(H6)

These were coded as dummy variables for the analysis. For Corporate vs Brand

Name, 1 was coded for corporate name title sponsors and 0 was coded for brand name

title sponsors in all samples. For Event Competition Level, title sponsors for Sprint

series were coded as 1 and those for Nationwide series were coded as 0 in all NASCAR

samples.

Amount of purse (H4) and traditionality (Hs)

Specific figures were obtained from Web search and used for the analysis. For

example, the total amount of purse and traditionality of the SEI Pennsylvania Classic in

the PGA 2000 tour were $3,200,000 and 1 (which means this was inaugural year of this

tour event) respectively. Such specific figures were collected in all samples in the PGA

and LPGA.

Industry segment (H7)

First, based on their business segment, 84 title sponsors were categorized into 6

categories as followings; (1) High-technology, (2) Automobile, (3) Food & Retail, (4)

Banking, (5) Investment, and (6) Service. The sample numbers for each category were

19 for High-technology, 13 for Automobile, 23 for Food & Retail, 13 for Banking, 6 for

Investment, and 13 for Service. Second, due to small sample size in the first

categorization, the researcher further categorized them into (a) High-technology by

combining above (1) High-technology and (2) Automobile, (b) Food & Retail, and

Services by combining (4) Banking, (5) Investment, and (6) Service. Thus, the final

sample number for ANOVA was 32 for High-technology, 23 for Retails, and 29 for

Services respectively.









THE MARKET VALUE OF SPORT SPONSORSHIP:
THE INFLUENCE OF TITLE SPONSORSHIP ON SPONSORING COMPANY'S
STOCK PRICES

















By

MASAKI KUDO


A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE

UNIVERSITY OF FLORIDA

2010









the announcement day. That is, this analysis was conducted under the situation that

market knows the event day in advance and, in such condition, demonstrated that

stocks of title sponsors were significantly bought by market before 2 days of the event.

As well as the announcement date, the researcher conducted further event study

analysis focusing on each PGA, LPGA, NASCAR Sprint, and NACAR Nationwide

sponsors respectively and the results were obtained as follows.

PGA

Table 5-13 and Table 5-14 show MAR and MCAR for the PGA title sponsors. As

expected, +2 day (in most samples, it is Monday which is the first day market available

after the PGA and LPGA event) observed significant market reaction which is negative

at 5% statistical significant level. In fact, 16 of 22 title sponsors in the PGA received

negative abnormal returns on the next market available day following the title

sponsorship event. These results are consistent with those in the announcement date

and confirmed that stock market has seen being title sponsor for the PGA tour brings

financially net negative for the company.

LPGA

Table 5-15 and Table 5-16 show MAR and MCAR for the LPGA title sponsors. As

expected again, +2 day observed significant market reaction which is positive at 5%

level. In fact, 7 of 9 title sponsors in the LPGA received abnormal stock market returns

on Monday and the mean 1.02% is the most striking increases observed in this study.

These results are consistent with those in the announcement date and put an another

evidence that title sponsors in the LPGA are favorably seen by stock market and that

market believes being title sponsors in the LPGA produces more financial value than

the cost of sponsorship.

PAGE 1

THE MARKET VALUE OF SPORT SPONSORSHIP: THE INFL UENCE OF TITLE SPONSOR SHIP ON SPONSORING COMPANYS STOCK PRICES By MASAKI KUDO A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORID A IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2010 1

PAGE 2

2010 Masaki Kudo 2

PAGE 3

To my parents, Masanori and Sac hiko Kudo, who gave me an opportunity to pursue a masters degree abroad and an opportunity to grow as a person in a wonderful place, the University of Florida, with incredibly great people. I am deeply indebted to them for their continued support and unwavering faith in me. 3

PAGE 4

ACK NOWLEDGMENTS I gratefully thank the c hair and member of my supervisory committee for their mentoring throughout my thesis. Dr. Ko, Dr Connaughton, and Dr. Wa lker, without your supports and advices, this thesis would not have been completed. I owe my deepest gratitude to them and it is my greatest honor to meet and have worked with you whom I respect as a researcher, as an educator and as a person, here half way across the globe from where I come from. A special than ks to the University of Florida, and the Department of Tourism, Recreation, and S port Management for providing me with the resources to complete this thesis. 4

PAGE 5

TABL E OF CONTENTS page ACKNOWLEDGMENTS ..................................................................................................4 LIST OF TABLES ............................................................................................................7 CHA PTER 1 INTRODUC TION ....................................................................................................11 Purpose of the Study ..............................................................................................15 Significance of the Study ........................................................................................15 Limitations of the Study ...........................................................................................16 Definitions of Terms ................................................................................................16 2 REVIEW OF LITERATURE ....................................................................................18 Title Sponsorship ....................................................................................................18 Measurement of Sponsorship Effectiveness ...........................................................20 Event Study Analysis ..............................................................................................23 Sponsorship Effectiveness Using Event Study Analysis .........................................27 3 THEORETICAL BACK GROUND AND HYPOTHESE S DEVELOPMENT ..............28 Hypotheses Development .......................................................................................28 Main Hypothesis (H )1.......................................................................................28 Image Congruence (H )2...................................................................................30 Corporate vs Brand Name (H )3........................................................................31 Prestige of Event (H ) (H )4 5...............................................................................32 Event Competition Level (Sprint/Nationwide) (H )6............................................33 Industry Segment (H )7......................................................................................34 4 METH OD ................................................................................................................36 Data Collection Procedures and Description of the Data ........................................36 Announcement Date .........................................................................................36 Event Date ........................................................................................................37 Possible Moderators .........................................................................................38 Image congruence (H )2..............................................................................38 Corporate vs brand name (H ) and event competition level (Sprint/Nationwide)(H)3 6...........................................................................39 Amount of purse (H ) and traditionality (H )4 5..............................................39 Industry segment (H )7................................................................................39 Data Analysis Procedures .......................................................................................40 Event Study Analysis ........................................................................................40 Multiple Regression Analysis and ANOVA .......................................................42 5

PAGE 6

5 RESULT S...............................................................................................................47 Event Study Analysis ..............................................................................................47 Announcement Date .........................................................................................47 PGA ...........................................................................................................49 LPGA .........................................................................................................49 NASCAR (Sprint series) .............................................................................50 NASCAR (Nationwide series) ....................................................................50 Event Date ........................................................................................................51 PGA ...........................................................................................................52 LPGA .........................................................................................................52 NASCAR (Sprint series) .............................................................................53 NASCAR (Nationwide series) ....................................................................53 Multiple Regression and ANOVA Test ....................................................................54 Multiple Regressions ........................................................................................54 ANOVA for Industry Segment ...........................................................................56 6 DISCUSSION A ND LIMITATION ............................................................................69 Discussion ..............................................................................................................69 Limitations of the Study ...........................................................................................72 APPENDIX SPONSOR-EVE NT FIT SC ALE ................................................................76 LIST OF REFERENCES ...............................................................................................77 BIOGRAPHICAL SKETCH ............................................................................................81 6

PAGE 7

LIST OF TABLES Table page 4-1 PGA title sponsorship sample .............................................................................434-2 LPGA title sponsorship sample ...........................................................................444-3 NASCAR Sprint series title sponsorship sample ................................................454-4 NASCAR Nationwide series title sponsorship sample ........................................465-1 Mean abnormal returns (MAR) for t he full 84 announcements in PGA, LPGA, and NASCAR ......................................................................................................575-2 Mean cumulative abnormal return (MCAR) for the full 84 announcements in PGA, LPGA, and NASCAR ................................................................................575-3 Mean abnormal returns (MAR) for announcement date in PGA .........................585-4 Mean cumulative abnormal return (MCAR) for announcement date in PGA ......585-5 Mean abnormal returns (MAR) for announcement date in LPGA .......................595-6 Mean cumulative abnormal return (MCAR) for announcement date in LPGA ....595-7 Mean abnormal returns (MAR) for announcement date in NASCAR (Sprint) .....605-8 Mean cumulative abnormal return (MCAR) for announcement date in NASCAR (Sprint) ................................................................................................605-9 Mean abnormal returns (MAR) for announcement date in NASCAR (Nationwide) .......................................................................................................615-10 Mean cumulative abnormal return (MCAR) for announcement date in NASCAR (Nationwide) .......................................................................................615-11 Mean abnormal returns (MAR) for the full 81 event date in PGA, LPGA, and NASCAR ............................................................................................................625-12 Mean cumulative abnormal return (MCAR) for the full 81 event date in PGA, LPGA, and NASCAR ..........................................................................................625-13 Mean abnormal returns (MAR) for event date in PGA ........................................635-14 Mean cumulative abnormal return (MCAR) for event date in PGA .....................635-15 Mean abnormal returns (MAR) for event date in LPGA ......................................64 7

PAGE 8

5-16 Mean cumulative abnormal return (MCAR) for event date in LPGA ...................645-17 Mean abnormal returns (MAR) for event date in NASCAR (Sprint) ....................655-18 Mean cumulative abnormal return (MCAR) for event date in NASCAR (Sprint) ................................................................................................................655-19 Mean abnormal returns (MAR) for ev ent dates in NASCAR (Nationwide) ..........665-20 Mean cumulative abnormal return (MCAR) for event dates in NASCAR (Nationwide) .......................................................................................................665-21 Multiple Regression Analysis ..............................................................................675-22 ANOVA for industry segment ..............................................................................685-23 Regression for high-technology company ..........................................................68 8

PAGE 9

Abstract of Thesis Pres ented to the Graduate School of the University of Fl orida in Partial Fulf illment of the Requirements for t he Degree of Master of Science THE MARKET VALUE OF SPORT SPONSORSHIP: THE INFLUENCE OF TITLE SPONSOR SHIP ON SPONSORING COMPANYS STOCK PRICES By Masaki Kudo August 2010 Chair: Yong Jae Ko Major: Sport Management Title sponsorship has been increasingly employed as a marketing communication tool, but its effectiveness and re turns are extremely difficult to measure. Likewise, the literature has not examined the impact of title sponsorship on the companys performance regardless of the ma rket size. The objective of this study is to analyze the impact of title sponsorship in the PG A, LGPA, and NASCAR, by examining the sponsors stock price fluctuations on both the title sponsorship announcement date and the following title sponsorship event date, and subsequently to explore the possible moderators in the stock price changes. The event study analysis conducted for each sport respectively demonstrates that title sponsors in the LPGA and NASCAR obtained significant stock increases on both the announcement date and event dat e. On the other hand, title sponsors in the PGA experienced significant stock price decreases on the both dates. That is, the stock market sees that title spons orships for the PGA are overvalued and those of the LPGA and NASCAR are undervalued. 9

PAGE 10

10 The examination of possible moder ators demonstrates that high image congruence between title sponsor and sponsored sport event and whether or not they are high-technology companies are the key det erminants in stock price increases. Also, the difference of NASCAR event competition level (i.e., Sprint series or Nationwide series) plays an important role in the stock price changes, and more prestigious event (i.e., Sprint series) sponsors ar e favorably seen by the market.

PAGE 11

CHA PTER 1 INTRODUCTION Sport sponsorship is defined as "the acquisi tion of rights to affiliate or directly associate with a product or event for the pur pose of deriving benefits related to that affiliation or association" (Mullin, Hard y, & Sutton, 2000, p. 254). In 2008, the International Events Group esti mated that $43.5 bil lion was spent around the world, and of that, $16.8 billion was spent in North Amer ica by companies sponsorship activities including sporting events, music events, fe stivals, and so on (Sponsor Map, 2008). Of the $43.5 billion sponsorship money, roughly 65-70% has been reportedly invested into sport sponsorship (IEG Sponsorship Report, 2006). This fact clearly indicates that utilizing sponsorship has become one of the most important advertising channels for companies to differentiate themselves among competitors. In reality, sponsorship spending overall is still proj ected to rise 3.9% in 2009 (IEG Press Release, 2009), regardless of a severe economic recession, which generally results in the dramatic budget cuts in marketing expenditures including sponsorship activities. According to Tripodi (2001), this uprising trend of sport sponsorship is expected to continue. This would be definitely true when tracking the amount of sponsorship revenue that the International Olympi c Committee (IOC) has gained through The Worldwide Olympic Partners program. T he Worldwide Olympic Partners have been known as the top category of Olympic sponsorship programme and was created one year after the Los Angeles Olympic in 1984, when the IOC experienced a tremendous financial success and realized that corpor ate sponsors provided the Olympic with substantial profits and that financing through sponsorship bec ame an integral part of the Olympic Movement (Giannoulakis & Stotlar, 2006). In fact, the IOC generated $95 11

PAGE 12

million (from 9 companies) in the 1988 Seou l Olympics; $175 million (from12 companies) in 1992 Barcelona; $ 350 million (from10 companies) in 1996 Atlanta; $550 million (from 11 companies) in 2000 Sydney; $650 million (from 11 companies) in 2004 Athens; and $866 million (from 12 companies ) in 2008 Beijing (Crompton, 2004; Giannoulakis, Stotlar & Chatzi efstathiou, 2008). Besides this sponsorship revenue of the IOC, the host countrys Olympic Committee has also separately generated significant amount of revenue from thei r sponsorship programs designed for local businesses. In the Sydney Olympics, for example, the Sydney Organizing Committee individually generated $492 millio n from their sponsorship programme (Crompton, 2004). That is to say, in Sydney Olympics, the total revenues that both IOC and the host committee gained through the corporate sponsorships were over $1 billion. To the contrary to this tremendous spons orship growth, it has been known that measuring sponsorship effectiveness is very challenging task (K o, Kim, Claussen & Kim, 2008; Spais & Filis, 2008); as a resul t, research with respect to sponsorship effectiveness is still in the infancy stage ( Oztur, Kozub & Kocak, 2004). While interest in evaluating the impact of sponsorship has increased (Copeland, Frisby, & McCarville, 1996; Crompton, 2004), most sp onsoring c orporations and pr operties do not actually have established procedures and systems for measuring sponsorship effectiveness (Miyazaki & Morgan, 2001). In reality, about 40% of companies engaged in major sport sponsorship activity reported that they do not spend their budgets on research to measure the impact of spons orship (Crompton, 2004). Also, traditional research in sponsorsh ip measurement tends to only focus on either media exposure or corporate image transfer (Oztur et al., 2004; Miyazaki & 12

PAGE 13

Morgan, 2001). While these methods measuri ng such as media exposure, awareness, and image enhancem ent are relatively easy to adopt for sponsoring companies (Crompton, 2004), these methods may not be abl e to provide the real effectiveness in terms of the commercial aspect on sponsor ship (Pham, 1991). Media exposure method, for example, is problematic because it is not the ultimate purpose of sponsorship (Pham, 1991) and achievement of a certai n level of media exposure does not necessarily lead to a high recall rate or positive attitude change in consumer behavior (Speed & Thompson, 2000). In fact, Crompton (2004) pointed out that [advertising] managers will be required to desi gn evaluation research that measures the extent to which [sponsorship] benefits are required, rather than being able to rely on the much easier measures that ev aluate media exposure, aw areness and image enhancement (p. 280). And continuously, the conventiona l wisdom appears to be that there is now more pressure on senior managers to demon strate accountability for sponsorship investments by showing their potential for increasing a companys profitability (p. 268). That is, demonstrating the commercial value of sponsorship as a final market outcome has been required for sponsoring companies to justify their hi gh sponsorship fee investment (Cornwell & Ma igan, 1998; Crompton, 2004). As described previously, the traditiona l approach to measuring sponsorship effectiveness, which mainly focused on medi a exposure or image transfer, does not necessarily justify the high sponsorship f ee. Instead of such traditional approaches, demonstrating positive results at the level of companys performance stage provide additional accountability that exposure or image transfer related research may not be able to identify in evaluating sponsorshi p effectiveness. Specifically, Cornwell and 13

PAGE 14

Maigan (1998) stated that dem onstrating the market value of sponsorship is apparently the best way to legitimize sponsorship in vest ment for sponsoring company. To date, however, only few studies have been conducted to analyze the impact of sponsorship on the companys actual financial perfo rmance (Calderon-Martinez, Mas-Ruiz, & Nicolau-Gonzalbez, 2005). In light of the increasing need for a systematic measurement tool for sponsorship effectiveness on the corporations performanc e level, several scholars have attempted to examine the impacts of sport sponsorsh ip by focusing on stock price changes (Cornwell, Pruitt, & Van Ness, 2001; Cornwe ll, Pruitt, & Clark, 2005; Farell & Frame, 1997; Samitas, Kenourgios, & Zounis, 2008). Th is methodology, a quantitative analysis of the stock price fluctuations recorded at the time of initia tion of the programs (i.e., the time of sponsorship announcements), is called as Event Study Analysis, which has been widely accepted in the field of finance, accounting, marketing, and management (Pruitt, Cornwell, & Clark, 2004). While ev ent study analysis is only recently to be introduced to marketing research field (Ozur et al., 2004), se veral studies employed this methodology in sport sponsorship contexts such as the Olympics, NASCAR, stadium naming right, major league offi cial products, athletic endorsement, and European soccer teams. None of them, however, has address ed the value of title sponsorship in professional sport event contexts. Also, mo st of the past research that adopted event study analysis method only focused on a singl e sport event rather than examining the relative impacts of sport sponsorship across multiple sports. Furthermore, most of the past research has exclusively focu sed on the announcement day of the new sponsorship agreement while the actual sporting event day following the new 14

PAGE 15

sponsorship deal seems to have a big impact ei ther positively or negatively on their stock prices when considering the amount of the sponsorship exp osure and media coverage on the event day. Purpose of the Study Accordingly, the purposes of this study are: (1) to examine effectiveness (i.e., stock price changes) of title sponsorship in sport events by comparing stock price changes of corporate sponsors in different sport events sponsorship (i.e., Professional Golfers Association (PGA), Ladies Professi onal Golf Association (LPGA) and National Association for Stock Car Auto Racing (NASCAR)) on both sponsorship announcement date and sporting event date following the announc ement and (2) to explore the roles of selected moderator variables in the st ock price changes by focusing on Image Congruence, Corporate vs Brand Name, Industry Segment, Amount of Purse, Traditionality, and Event Competition Level (Spr int/Nationwide). For the purposes of this study, the following research questions were developed: (1) Is there any linkage between sport sponsorship involvement and corporate performance (i .e., stock price changes)? (2) Are there any differences in sp onsorship effectiveness (i.e., stock price changes) among selected types of sport events (i.e., the PGA, LPGA, and NASCAR)? And (3) Are there any other variables that influenced the stock price changes? Significance of the Study This research would make a contributi on to fill two major gaps existing in sponsorship literature and spor t management practice: (1) the lack of research in title sponsorship effectiveness, and (2) the need fo r more research measuring sponsorship effectiveness at the level of corporate performance. The results of this research should 15

PAGE 16

be of interest to both researchers and practitioners and be used for the further development of sport sponsorship. Limitations of the Study The main limitations of this study are the following three: (1) event study analysis, (2) the selection of the companies (i.e., title sponsors), and (3) the determination of the congruence level between title sponsor and s port event. These are discussed in detail in the discussion section Definitions of Terms Title Sponsorship: Title sponsorship is defined as the acquisition of rights to take part in the official name of t he event for the purpose of deriv ing benefits related to that name sharing. Event Study Analysis: Event study analysis is an analysis of whether there was a statistically significant r eaction in financial markets to past occurrences of a given type of event that is hypothesised to affect public organizations market values (Spais & Filis, 2008, p. 173). Abnormal Stock Market Return: Abnormal stock market retu rns refers to returns not consistent with the pattern of change as established by past company and market activity. Image Congruence: Image congruence refers to a perceived image fit between the sponsoring company and the sponsored sport event. Corporate vs Brand Name: Corporate vs brand name re fers to whether the name of the title sponsorship is accompani ed with the main co rporate name or the name of the title sponsor is accompanied with the corporate di visional brand name. 16

PAGE 17

17 Industry Segment: Industry segment refers to the industrial sector to which the title sponsor corporation belongs. Amount of Purse: Amount of purse refers to the total amount of purse the event raised for the tournament. Traditionality: Traditionality refers to the lengt h (year) of the tournament having been held. Event Competition Level (Sprint/Nationwide): Event competition level refers to whether the NASCAR event belongs to the Sprint Cup series or the Nationwide series. Specifically, this thesis has been organized as follows: first, a review of the literature with respect to title sponsor ship and measurement of sponsorship effectiveness is presented, along with the c hallenges in measurem ent of sponsorship effectiveness. Also, in the literature revi ew, the method of Event Study Analysis is introduced and previous research using this method is reviewed by focusing on the advantages and disadvantages of this methodol ogy; Second, the research hypotheses are presented with the theoretical backgrounds based on pr evious research; Third, the data collection method and data analysis pr ocedures are presented; Fourth, the empirical results obtained in the study are described; Finally, a discussion is presented by focusing on research and managerial implic ations, main conclusions, and limitations of this study.

PAGE 18

CHA PTER 2 REVIEW OF LITERATURE Title Sponsorship Since sponsorship particularly focusing on sport title sponsorship has not been well developed, anecdotal evidences were collected from various sponsorship-related journal articles and business resources (e .g., naming rights sponsorship, which is considered one type of title sponsorship). Ba sed on the definition of sponsorship, the acquisition of rights to affilia te or directly associate wit h a product or event for the purpose of deriving benefits related to that aff iliation or association (Mullin et al., 2000, p.254), title sponsorship can be defined as the acquisition of rights to take part in the official name of the event fo r the purpose of deriv ing benefits related to that name sharing. Then why do corporations seek title sponsorship? According to Clark, Cornwell, and Pruitt (2002), it is natural that there is not much difference between corporate objectives that managers hold for the naming ri ght sponsorship and those held for other event and activity type of sponsorship. The sim ilarity they have is that both naming right sponsorship and traditional event sponsorship utilize their rights of sponsorship to maximize brand awareness and seek strong brand associations through the repeated parings (Becker-Olsen, 2003). Specifically, Roy and Cornwell (1999) identified the six objectives that managers hold for the nami ng right sponsorship; Image Enhancement, Less Cluttered Communications Environment, Aw areness, Brand Positioning, part of an Integrated Marketing Communications Plan, and Direct On-site Sales. On the other hand, there are some differences between t he objectives of naming right sponsorship and those of traditional sponsorship in its scope, duration and value. Those differences 18

PAGE 19

may influence consumer perceptions, effi cacy and market valuation (Becker-Ol sen, 2003). For example, naming right sponsorship is likely to gain a greater number of exposures due to the longer duration while event sponsorship tends to have more supporting advertising messages due to the shorter duration (Becker-Olsen, 2003). Also, when it comes to sphere of influenc e, naming right sponsorship seems to have advantage over traditional sponsorship because of the longer duration, more exposure through nationwide television media, and lar ger target audiences (Becker-Olsen, 2003). For instance, regarding the difference in dur ation, one week or even one month after a PGA Tour, people are more likely to remember the name of title sponsor of the tour, while they are not as likely to remember the name of individual sponsors each player has. It seems that the high visibility t he naming right sponsor has plays an important role in title sponsorship t oo. In fact, Stipp (1998) noted that high visible naming right sponsor has advantage in buildin g stronger positive brand associations in consumer perception over traditional sponsorship. As an example of the high visibility a sponsor gained through their title sponsorship, Thatlook, a referral service for elective cosmetic surgery, recorded more than 4 million Internet hits in the three days following its title sponsorship of a NASCAR Winston Cup, wh ich was about 16 times as a normal days hits (IEG Sponsorship Report, 2000). In an exchange for such high visibility and other benefits title sponsorship gain, the right fee for title sponsorship is general ly expensive, especially when it comes with nationwide or global level television exposure As a matter of fact, in the PGA Tour, annual 5% increases of title sponsorship fees have been considered as a standard part of title sponsorships (Show, 2009). Also, despite the current economic recession of 19

PAGE 20

2009, the PGA Tour decided to retain the steady fees for their title sponsorship rights (Show, 2009). In fact, t he president of one golf course that lost their title sponsor was optimistic and said immediately after the title sponsors withdrawal announc ement that we are 100% confident that well have a new title sponsor in place for the January 2010 golf tournament and weve got five or six great (corporate) names were talking to, all of which youd recognize (Ferguson, 2009). However, the situation surrounding the PGA Tour has changed due to a severe economic recession in 2009. The PGA lost th eir four title sponsors in a year; two tournaments supported by Buick, one by the U.S. Bank, and one by Stanford Financial. Of those, the 50-year relationship with Buick was the oldest and the most established of the PGA Tour title sponsorships. Also, t he CEO of Deutsche Bank Americas, the top manager sponsoring the Deutsche Bank Championship, criticized the PGA Tours annual rising purses saying that theres pr essure on the PGA Tour to reduce the purses it will demand sponsors to fund in the coming years (Mell, 2009). Also, a sport critique noted that even though no PGA Tour event can exist much longer without a title sponsor, the seven million-plus price t ag for just operating a tournament today is simply too much for a local, nonprofit entity to raise (Harig, 2009). These situations clearly suggest that the impor tance of measurement of ti tle sponsorship effectiveness and the research results should be used as a reference to determine the fair price of title sponsorship. Measurement of Sponsorship Effectiveness In general, academic marketing research has been accused of its insufficient attention with respect to sponsorship regar dless of its market size, and particularly measuring the effectiveness of sponsorship (Spais & Filis, 2008). However, evaluating 20

PAGE 21

the value of sponsorship is very difficu lt task for both sponsorship industry and academic researcher (Giannoukakis et al., 2 008; Oztur et al., 2004; Miyazaki & Morgan, 2001). As noted, in the sponsorship i ndustry, 40% of companies engaged in sponsorship activity reported they have not hing spent on the research for sponsorship effectiveness (Crompton, 2004). Also, in t he academic field, research regarding sponsorship measurement tends to traditi onally focus on either media exposure or corporate image transfer (O ztur et al., 2004; Miyazaki & Morgan, 2001). The former method, media expos ur e, has been conducted through the quantification of media expos ure and coverage by either measuring or counting the times the sponsor is referred to, the time s its logo appears, or the duration of the appearance of the sponsor name or logo in the media (Speed & Thompson, 2000). In most cases, such media exposure is then tr anslated into money, in order to calculate the advertising cost that would be requir ed to achieve an equivalent exposure or coverage in the traditional forms of advertising (Calderon-Martinez et al., 2005). However, this approach has several limitati ons. For instance, Pham (1991) argued that for the sponsoring company, media cove rage is not the ultimate purpose of sponsorship; therefore, it should not be us ed for the effectiveness evaluation. As a matter of fact, achievement of a certain level of media coverage does not necessarily imply a high recall rate or a positive attitude change in consumer behavior (Ozturk et al., 2004; Speed & Thompson, 2000) Therefore, Pham (1991) pointed out that the exposure method cannot provide or analyze the real effectiveness at least in terms of the commercial aspect on sponsorship. Howeve r, despite the fact that there are substantial limitations w hen using media exposure and following advertisement 21

PAGE 22

equivalenc y calculation to assess the s ponsorship effectiveness, this type of measurement seems to be the most wi dely adopted tool in the business world (Poknywczynski, 2000). Poknywczynski (2000) reported that while 70% of companies they researched were using media exposure as a tool for their sponsorship impact measurement, only 40% of companies used the impact on awareness and image, and only 7% of companies used the impac t of sales for their measurement. The latter approach, corporate image trans fer, is typically measured through surveys directed at the consumers. The rati onale behind the corporat e image transfer is that pre-existing associations held in consumers memory with respect to a sponsorship event become linked in memory with the s ponsoring company; ther efore, the image of sponsorship event is transferred into the company or product (Gwinner, 1997; Gwinner & Eaton, 1999). Gwinner and Eaton (1999), for example, demonstrated that sponsorship results in corporate im age transfer and when sponsoring brand are matched on either on the image or function with the event, it s positive image transfer is enhanced. Also, Grunert (1996) showed that a simple exposure leads to a better evaluation of a product or brand under the situation that t he consumers automatically respond to the advertising. However, this type of measurement also has the limitation in that it does not differentiate between the impacts of a sponsorship per se and the impacts of other forms of advertising (Calderon-Martinez et al., 2005; Miyazaki & Morgan, 2001). The difficulty of di fferentiation of its effects from sponsorship per se and other promotional type of marketing s eems to be a common problem in evaluating sponsorship effectiveness (Pham, 1991). 22

PAGE 23

On the other hand, as an untraditional approach in an effort to measure sponsorship effectiveness, Meenaghan (1991) suggested to measure sponsorship effectiveness in terms of sales impact or ot her indicators of companys performance (i.e., market outcome) such as corporate a ccounting. However, cost-benefit analysis of sponsorship (i.e., comparing revenue changes and its associated costs) is in most cases very complicated and as a result, bri ngs ambiguous results (Miyazaki & Morgan, 2001). For instance, a research effort by t he United States General Accounting Office (GAO) to evaluate the sponsorship effect iveness of the U.S. Postal Service in 1992 Olympics eventually resulted in unknown (General Accounting Office Report, 1993). The conclusion of GAO was that GAO c ould not determine whether the Olympic sponsorship produced a profit or loss becaus e GAO could not verify the key revenues and costs attributed solely to the Olympic sponsorship (General Accounting Office Report, 1993). In summary, while proving the market value of sponsorship is the best way to justify sponsorship investment as a useful marketing technique (Cornwell & Maigan, 1998), measuring sponsorship effects in terms of companys performance as a final market outcome is a very challengi ng task (Calderon-Martinez et al., 2005). Fortunately, however, this difficulty can be approached by employi ng a methodological technique called event study analysis, which uses stock price fluctuations of a corporation in order to in vestigate the market reacti ons to the announcement of sponsorship agreement between the corporat ion and the event (Miyazaki & Morgan, 2001). Event Study Analysis Event study analysis is an analysis of whether there was a statistically significant reaction in financial markets to past occurre nces of a given type of event that is 23

PAGE 24

hypothesis ed to affect public organizations market values (Spais & Filis, 2008, p.173). Although event study analys is has a long history and been widely employed as a research tool in the finance and economics fiel d (MacKinlay, 1997), it is only recently to be introduced to marketing research (Oztur k et al., 2004). In management, event study analysis has been used to judge the effects of corporate control changes, such as mergers, earnings announcements, issues of new debt, CEO turnover (McWilliams & Siegel, 1997). In fact this method is known as the st andard assessment metric for the measurement of the net ec onomic value of any corpor ate event whose initial announcement dates are precisely available (P ruitt et al., 2004). O ne of the primary reasons why event study analysis has bec ome so popular in both management and academic research is that this method obv iates the need of analysis of accountingbased measures (McWilliams & Siegel, 1997) Although it is possible that managers show or select some favorable accounti ng-based figures and manipulate the actual corporate performance, st ock prices are not as subject to such manipulation by insiders (McWilliams & Siegel, 1997). Theref ore, stock prices are suppos ed to reflect the true value of a corporation (McWilliams & Si egel, 1997) and represent the corporate performance in a better way than does the acc ounting sheet (Calderon-Martinez et al., 2005). While event study analysis can be a powerful technique, there are some disadvantages in this method (Becker-Olsen, 2003; Spais & Filis, 2008). The disadvantages of event study analysis are a ssociated with the charac teristics of this method in that the validity of conclusions l ed by event study analysis highly depends on its strong assumptions. The event study analysis is basically conducted under the three 24

PAGE 25

major assumptions that need to be address ed. The first assumption is that markets are efficient. In an efficient ma rket, stock prices correctly and quickly respond to every single piece of information available to t he public (Miyazaki & Morgan, 2001). That is, public announcements or communication of event s influencing future market value of the company (e.g., announcements of merger s, new product information, and celebrity endorsement) should be incorporated into invest ors mind as a decision-making factor (Miyazaki & Morgan, 2001). As a resul t, any information change assumed to be important for the future value of the company results in significant changes in the stock price. In other words, stock prices should represent the present value of the company such as discounted future income, expendit ures, and firms strategy (Miyazaki & Morgan, 2001). The second assumption is that the event is unanticipated to the public. The market previously did not have any information on the event, and the investors should gain the information for the first time from the announc ement (McWilliams & Siegel, 1997). Abnormal stock market return has been assumed to be the result of the market response to the new information. Wh en an event was anticipated or leaked to the public and investors in advance of t he official announcem ent, such information leakage would make use of event study analysis problematic (McWilliams & Siegel, 1997). The third assumption is that researcher can isolate the effects of an event from the effects of other events. In event study analysis, this a ssumption is perhaps the most critical one (McWilliams & Siegel, 1997). Th is assumption theoretically allows researcher to isolate market returns deriv ed from sponsorship fr om those from other promotions by eliminating the impact of other events (Calderon-Martinez et al., 2005). That is, researcher needs to be careful of confounding effects during the event window. 25

PAGE 26

Confounding events are major news or other si gnificant information that might have an impact on the stock prices during the event window (Becker-Olsen, 2003). In general, the longer the event window, the more difficult it is for researcher to justify the above assumption and claim that confounding effects had been controlled (McWilliams & Siegel, 1997). In short, while event study analysis has been widely employed in management and academic research, there is a fact that the usefulness of this methodology depends heavily on several strong assumptions (Bro wn & Warner, 1985). By violating these assumptions, the results of research may be biased and inaccurate; therefore, the conclusion may be problematic, or even it is possible to manipulate the conclusion by employing inappropriate techniques (Spais & Filis, 2008). Basically, event study analysis involves measuring how a certain event (i.e., announcement of sponsorship agreement in this study) influences movement of particular stock prices. According to Miya zaki and Morgan (2001), it is important to compare the movement of the individual stock to the mark et as a whole in order to determine whether a movement of large ma gnitude in the individual stock price coincides with what of the market or whet her it was due to firm-specific information (p.11). The abnormal stock market returns (i.e., returns not consistent with the pattern of change as established by past company and market activity) are calculated as the difference between the actual and the expe cted returns. Thus, a positive abnormal return (i.e., a si gnificant increase beyond normally expe cted returns) indicates that the event is typically seen as beneficial to a companys market value. On the other hand, a negative abnormal return (i.e., a significant decrease beyond normally expected returns) 26

PAGE 27

27 suggests that the event is harmful to a comp anys future value. If the results brought no cumulative abnormal return, it would be implyi ng the event is seen neither a positive nor a negative for the future value of the company. Sponsorship Effectiveness Using Event Study Analysis Until now, several studies have c onducted event study analysis in sport sponsorship context. For example, Fa rell and Frame (1997), Miyazaki and Morgan (2001), Oztur et al. (2004), Samitas, K enourgios, and Zounis (2008), and Spais and Filis (2006) examined the stock pr ice impacts of sponsoring the Olympic Games. Also, Cornwell, Pruitt, and Van Ness (2001) and Pr uitt et al. (2004) studied the NASCAR sponsorship effectiveness. Likewise, Becke r-Olsen (2003) and Clark et al. (2002) did about sponsorship of the stadium naming right s. The study of Cornwell, Pruitt, and Clark (2005) analyzed effects of the official pr oduct sponsorship in major league sports including the MLB, NBA, NFL, NHL, and PGA. Furthermore, Spais and Filis (2008) researched sponsorship effects of an Italian soccer team. As well, Russell, Mahar, and Drewniak (2005) studied stock ma rket reactions to the public ity of athletic endorsers. And finally, Caldron-Martinez et al. (2005) explored the difference between commercial sponsorship and philanthropic sponsorship by employing event study analysis. None of them to date, however, has focused on effectiv eness of title sponsorship in professional sport by employing event study analysis.

PAGE 28

CHA PTER 3 THEORETICAL BACKGROUND AND HYPOTHESES DEVELOPMENT Hypotheses Development For the first time, this study aims at comparing the market value of title sponsorship in professional sports, specif ically in the PGA, LPGA, and NASCAR, employing stock price fluct uations as a measurement by focusing on both sponsorship announcement day and sporting event day, and subsequently, the possible existence of moderators affecting the stock price fluctuations in title sponsorship is explored. Specifically, whether the sponsoring comp anys characteristics such as congruence between company and sponsored even t, and the sponsored sport events characteristics such as prestige of the event influenced their stock prices are investigated in this study in an effort to identify the possible moderators in title sponsorship effectiveness. Main Hypothesis (H1) Previous researches of sponsorship e ffectiveness using event study analysis as the measurement have produced mixed resu lts including both in vestors positive perspectives ( Cornwell et al., 2001; Cornwell et al., 2005a, Miyazaki & Morgan, 2001) and negative perspectives (Farell & Frame, 1997; Oztur et al., 2004) toward corporations sponsorship activities. For inst ance, the study of Caldron-Martinez et al. (2005) demonstrated that while commercia l sponsorship generates abnormal stock market returns, philanthropic sponsorship does not bring such benefits. Thus, they concluded that corporations tr ying to improve their performan ces in terms of stock price value through sponsorship activities had better choose commercial sponsorship rather than philanthropic sponsorship. The results of several studies whose study subjects are 28

PAGE 29

clearly categorized int o commercial sponsor ship (Cornwell et al., 2005a; Miyazaki & Morgan, 2001; Pruitt et al., 2004; Spais & Filis, 2006) are consistent with the conclusion led by Caldron-Martinez et al. (2005), in that the announcements of commercial sponsorship were positively associated with t he abnormal stock market returns. In this point, since samples of title sponsorship used for this study obviously belong to commercial sponsorship, abnormal stock market return would be expected as the result of sponsorship announcements. Also, busin ess behavior surrounding sport sponsorship such as rapid growth of sponsorship mark et and increase of the sponsorship fee that occurred during the past several decades imp lies that corporations find enough cost and communication advantages in sport sponsors hip over other tradit ional advertisings (Cornwell et al., 2005a). Based on these past findings, the following hypothesis is presented: Hypothesis 1: Announcements of title sponsorship in NASCAR, PGA, and LPGA positively influence on stock price of sponsoring companies. In addition to the above central hypothesis that would answer the most important research question established for this study, it is important to explore and identify the possible existence of moderators (variabl es) that may influence the result of sponsorship (i.e., stock price changes). Ther efore, several potentia l variables are taken into account based on two perspectives; one originated from sponsoring companys characteristics including Image Congruence, Corporate vs Brand Name, and Industry Segment (e.g., high technology, banking, food, etc), and another derived from sponsored events characteristics. In this study, the researcher use event prestige as sponsored events characteristics and for the purpose of this study, event prestige is 29

PAGE 30

defined by three elements of the event; (a) Amount of Purse, (b) Traditionality, and (c) Event Competition Lev el (Sprint/Nationwide) These potential variables are discussed along with the hypotheses in the following sections. Image Congruence (H2) Image Congruence refers to a perceiv ed image fit between the sponsoring company and the sponsored sport event. Numer ous studies highlighted the importance of the degree of congruence found in the relationship between sponsor and sponsored event in predicting consumers response to sponsorship (Cornwell, Weeks, & Roy, 2005; McDaniel, 1999). For example, Gwinne r and Eaton (1999) concluded that when event and sponsor brand are matched on either an image or functional basis, the image transfer process, which are most likely soug ht by sponsoring company as the primary purpose of the sponsorship, is enhanced. According to Cornwell et al. (2005b), sponsors and products that have been perceived as being closely related to the event by the consumers have a number of advant ages over unrelated sponsors. In fact, McDaniel (1999) stated that the nature of linkage or congruence found in the relationship between sponsor and event is an important determina nt for successful sponsorship activity. Specifically, in the literature of event study analysis, Cornwell et al. (2001) demonstrated that the NASCAR sponso rs whose business segment has a direct tie to automotive industry ex perienced 3 percent greater increases in their stock prices on an average than those do not have the congr uence. Also, Cornwell et al. (2005a) showed that the Major League Ba seball (MLB), National Basketball Association (NBA), National Hockey League (NHL), National Football League (NFL), and PGA sponsors that had congruence with their s ponsored sports increased their stock prices about 11% 30

PAGE 31

higher than sponsors that did not have lin k age with the sponsored sports. These past findings suggest following: Hypothesis 2: Image congruence between sp ort event and title sponsors is likely to have positive association wi th the stock price change. Corporate vs Brand Name (H3) Corporate vs Brand Name refers to whether the name of the ti tle sponsorship is accompanied with the main co rporate name or the name of the title sponsor is accompanied with the corporate divisional brand name. The study of Thjomoe, Olson, and Bronn (2002) that analyzed decision making process regarding sponsorship activities in 400 Norwegian firms demonstr ated that more professional company in sponsorship activities tend to use sponsorsh ip to improve their whole corporate image, not a particular product when compared with a less professional company. Pruitt et al. (2004) supported this result in the NASAR sponsors by event study analysis. They found that stock market investors have more favorably seen sponsorship undertaken in the name of an entire co rporation than sponsorship tagged along with single product or division of a corporation. The rationale behi nd this is that the NASCAR fans who have been known as having extraordinary loyalty for the teams and sponsoring corporations are more likely to be influenced in a positive way for sponsoring corporations by corporate-level sponsorship that offers the corporation mo re opportunities to link fans support across a corporations whole products and services rather than single product/service-level sponsorship w here the benefits only concentrate the product/service (Pruitt et al., 2004). Also, sponsorship at the corporate-level is likely to offer more chances of contacts for positive sponsor fit comparing to sponsorship at the brand-level (Pruitt et al., 2004). Because brand-level sponsorship is highly specialized in 31

PAGE 32

terms of delivered con cepts, the possibilit y of mismatch betw een the sponsored event and the sponsoring brand is relatively high (Pruitt et al., 2004). Therefore, Hypothesis 3: Title sponsors whose title is accompanied with the corporate name (i.e., corporate-level sponsorship) are likely to have more stock market returns than do those whose title is named by the corporations product or service (i.e., brand-level sponsorship). Prestige of Event (H4) (H5) As noted previously, for the purpose of th is study, prestige of event is defined by three elements of the event; (a) Amount of Purse, (b) Tr aditionality, and (c) Event Competition Level (Sprint/Nat ionwide). Amount of Purse and Traditionality are used for the PGA and LGPA. Event Competition Level is used for the NASCAR. Amount of purse and Traditionality: In the PGA and LPGA title sponsorship, in an effort to evaluate the effects of differ ences in tournament prestige on sponsorship returns, Amount of Purse and Traditionalit y were included as potential variables. Amount of Purse refers to t he total amount of purse the title sponsor raised for the tournament. Traditionality refers to the length of the tournament having been held. According to Speed and Thompson (2000), perc eived prestige of the sponsored event is a determinant that positively associates wi th the sponsorship response. For example, a special and high-status event such as the Olympics is likely to create such opportunities for sponsors since the audience usually has a high regard toward such an event, regardless the level of personal r egard for the sponsor (Stipp & Schiavone, 1996). Thus, prestige of the PGA and LPGA tournament, which is measured by amount of purse and traditionality in this study, is lik ely to be seen in a positive manner in the stock market. Therefore, 32

PAGE 33

Hypothesis 4: In the PGA and LPGA title spons orship, amount of purse in the tournament is likely to associate positi vely with the stock market returns. Hypothesis 5: In the PGA and LPGA title spons orship, traditionality of the tournament (i.e., the lengt h of the tournament having been held) is likely to associate positively with the stock market returns. Event Competition Level (Sprint/Nationwide) (H6) In the NASCAR title sponsorship, in an effo rt to evaluate the e ffects of differences in competition level of event on sponsorsh ip returns, Event Competition Level was considered as a potential variable. Event Competition Level refe rs to whether the NASCAR event belongs to the Sprint Cup series or the Nationwide series. In the NASCAR event, the Sprint Cup series is the highest level of competition and the Nationwide is the second level of competition. The study of Pruitt et al. (2004) found that sponsorship announcement with a highly successf ul team, which was determined by the acquired winning points in last years races, is more positively received in stock market than that with a lower-tier team. That is, su ccessful teams in the NASCAR are more likely to have abnormal stock market returns (Pruitt et al., 2004). The rationale behind this is that when consider ing the sphere of television exposure and other media coverage, successful teams are much more likel y to gain such benefits. In this respect, title sponsor of higher level competition (i.e., Sprint Cup series) are more likely to have abnormal stock market returns th an are that of lower level competition (i.e., Nationwide series) because they are likely to gain br oader media coverage. Also, higher level competition is considered as more pres tigious event in a general sense; thus, sponsorship responses are likely to result in a more positive way (Speed & Thompson, 2000). Therefore, 33

PAGE 34

Hypothesis 6: In the NASCAR title sponsorship, title sponsors of the Sprint Cup Series are likely to have more stock market returns than do those of the Nationwide Series. Industry Segment (H7) Industry Segment refers to the industrial sector to which the title sponsor corporation belongs. Industrial Segments included in this study are high-technology, automobile, food, banking, investment, and pharmacy. Several studies (Caldron-Martinez et al ., 2005; Clark et al., 2002; Cornwell et al., 2005a) demonstrated that hi gh-technology companies in the computer, internet, telecommunication, and electricity sectors experienced positive reactions following the announcements of their sponsorship more than did the other types of companies such as banking, insurance, construction, food and so on. For example, Clark et al. (2002) reported that high technology companies sponsoring stadium naming rights experienced over 4% greater net increases in stock pr ices following the stadium sponsorship announcements than did non-high technology companies. Likewise, The study of Cornwell et al. (2005a) showed that high technology companies sponsoring major-league sports official products experi enced about 11% greater net increase in shareholder wealth following the announcem ents than did not-high technology companies. Furthermore, Caldron-Martinez et al. (2005) found that among various types of sponsorship firms, the sector of electr icity, which could be categorized into high technology companies in the above two st udies, was a determinant of abnormal stock market returns. Actually in their study among seven industrial sectors including construction, automobile, communication, petrol eum, banking, insurance and electricity, only the sector of electricity brought a stat istically significant difference in gaining 34

PAGE 35

35 abnormal stock market returns at a probability < 0.01 level. These findings suggest the following hypothesis: Hypothesis 7: Title sponsors whose business segment is high-technology in computer, Internet, and electricity industr ies are likely to have more stock market returns than do non-high-tech companies.

PAGE 36

CHA PTER 4 METHOD Data Collection Procedures an d Description of the Data Announcement Date The primary data used in this study are daily stock returns of companies having announced title sponsorship agreements in the PGA, LPGA, and NASCAR. The initial title sponsorship lists for the PGA, LPGA and NASCAR were drawn from the Web pages of each sport organization during 2009. In this study, renewal announcements of title sponsorship were not included; thus only title sponsors that entered a title sponsorship for the sporting event for the firs t time were used for th is study. The PGA, LPGA, and NASCAR were selected for this st udy because they are some of the most popular professional sports in the United St ates, and other popular sports, for example the MLB, NFL, NBA, and NHL have rarely given title sponsorship to their games; instead, the PGA, LPGA, and NASCAR chosen fo r this study have virtually put a title sponsor name on every single game they ha ve held. In this study, the researcher included only title sponsors that have publicly-traded common st ock on either the NYSE, AMEX, or the NASDAQ exchange at the time of t he sponsorship announcement. As a result, both foreign and privately-held do mestic companies were excluded from the analysis. Also, from the obtai ned data, closing prices were picked to use in the data analysis. The volume and highest-lowest pric es were not used. The daily stock data were obtained from the Universi ty of Chicago's Center for Research in Security Prices (CRSP), which is well known as the most comprehensive database of stock price, return, and volume data for the NYSE, AMEX and NASDAQ stock markets. 36

PAGE 37

According to Brown and Warner (1985) who provided a comprehensive outline of event study analysis, great care should be taken to precisely de termine the date of the first trading day following the spons orship announcement. To do so, an extensive search by marking use of electronic data bases such as Lexis-Nexis, Factiva, U.S. newspapers and periodicals was conducted. Tabl e 1 presents lists of the title sponsor, title sponsorship event, sponsorship announcement date, and following event date analyzed in this study (Table 4-1 for PGA, Table 4-2 for LPGA, Table 4-3 for NASCAR Sprint series, and Table 4-4 for NASCAR Nati onwide series). In an effort to enhance the relevance, only those sponsorships init iated between January 1, 2000 and December 31, 2008 were included in the analysis. The to tal number of samples is 84. While small sample size is quite common in the manage ment literature using event study analysis (McWilliams & Siegel, 1997), 84 samples in this study are considered to be a large enough sample. For example, among 12 studies using event study analysis in sportrelated literature, the largest samples we re 53 by Cornwell et al. (2005a), and the average of those 12 studies was 28 samples. Event Date As noted previously, besides the announc ement day of the title sponsorship agreement, this study also examines stock price reactions in the sporting event day whose title is named after the company following the sponsorsh ip agreement. When considering that naming right sponsorship has advantage over traditional advertisings because of the longer duration, high visib ility via nationwide br oadcasting, and larger target audiences (Becker-Olsen, 2003; Stip p, 1998), the actual sporting event day following the new sponsorship agreement seems to have some impacts either positively or negatively on their stock prices Therefore, it would be impo rtant to take into account 37

PAGE 38

the stock price changes around the event day too in or der to fully measure the effectiveness of title sponsorship in terms of stock price changes. For the purpose of this study, the event day was set to the first day market opened following the title sponsorship event day, otherwise the event was held on weekday that market is opened as usual. That is, in the NASCAR whose events are usually held on Saturday or Sunday, the event day was set to the next market available day that is usually Monday (or sometimes Tuesday if the Monday was also market closed day). In the PGA and LPGA, most of the event day was set to Thursday because most tours officially start on Thursday. In such case, +2 day in the event windows indicates Monday, which will be likely impacted on stock prices due to the great number of title sponsorship exposure and recognition by Sundays media coverage. That is, in the NASCAR, some stock market reactions are likely to be observed on the event 0 day if any, and in the PGA and LPGA, those are like ly to be seen on the event +2 day if any. Possible Moderators In addition to the primary analysis on sto ck prices, several variables to be used for exploring the possible moderators influenc ing on the stock prices were collected and categorized as following: Image congruence (H2) To determine the level of Image Congruenc e between title sponsor and sponsored event, the researcher employed questionn aire developed by Speed and Thompson (2000), which comprised of 5 questions asking sponsor-event fit with a 7 Likert (see Appendix A). After careful discussion with an expert, the questionnaire was answered in all 84 samples and the average of 5 questi ons was each employed as the congruence level between title sponsor and spons ored event for all 84 samples. 38

PAGE 39

Corporate vs brand name (H3) and event competition l evel (Sprint/Nationwide)(H6) These were coded as dummy variables for the analysis. For Corporate vs Brand Name, 1 was coded for corporate name title sponsors and 0 was coded for brand name title sponsors in all samples. For Event Co mpetition Level, title sponsors for Sprint series were coded as 1 and those for Nationwide series were coded as 0 in all NASCAR samples. Amount of purse (H4) and traditionality (H5) Specific figures were obtained from W eb search and used for the analysis. For example, the total amount of purse and traditi onality of the SEI Penn sylvania Classic in the PGA 2000 tour were $3,200,000 and 1 (which me ans this was inaugural year of this tour event) respectively. Such specific figures were collected in all samples in the PGA and LPGA. Industry segment (H7) First, based on their business segment, 84 title sponsors were categorized into 6 categories as followings; (1) High-technology, (2) Automobile, (3) Food & Retail, (4) Banking, (5) Investment, and (6) Service. The sample numbers for each category were 19 for High-technology, 13 for Automobile, 23 for Food & Retail, 13 for Banking, 6 for Investment, and 13 for Service. Second, due to small sample size in the first categorization, the researcher further ca tegorized them into (a) High-technology by combining above (1) High-technology and (2 ) Automobile, (b) Food & Retail, and Services by combining (4) Banking, (5) In vestment, and (6) Service. Thus, the final sample number for ANOVA was 32 for Hightechnology, 23 for Retails, and 29 for Services respectively. 39

PAGE 40

Data Analysis Procedures Event Stud y Analysis The basic procedures followed the st andard event study methodology proposed by Brown and Warner (1985), and McWilliams and Siegel (1997). The event study analysis is designed to measur e the impact of a specific ev ent has on the stock price of a selected company by comparing actual st ock returns of the company to estimated stock returns. The actual stock returns were obtained from th e CRSP databases. The estimated stock returns is w hat the stock price would have been in the absence of the event and is determined by the stock price against a market index such as the CRSP value weighted index or the Standard & Poors 500 on the event day (Becker-Olsen, 2003). In this study, the CRSP value weighted index was used as a market index. By calculating the difference bet ween the actual stock retu rns and the estimated stock returns, the researcher is able to obtain the abnormal stock market returns, which are assumed to reflect the stock markets reaction to the event. The first step is estimating a market model for each company. The equation is as follows (McWilliams & Siegel, 1997): The rate of return on the share price of firm i on day t is expressed as Rit = i + i Rmt + it where Rit = the rate of return on the share price of firm i on day t, Rmt = the rate of return on a ma rket portfolio of stocks (such as the CRSP value weighted index) on day t, = the intercept term, = the systematic risk of stock i, and it = the error term, with E( it) = 0. 40

PAGE 41

The estimation of the above equation allows the calculation of the daily abnormal stock market returns (AR) as follows (McWilliams & Siegel, 1997): ARit = Rit ( a i + bi Rmt ) where a i and b i are the estimations obtained with the par ameter estimates obtained from the regression of Rit on Rmt over an estimation period (T). In this study, the researcher and used a maximum of 255 daily return observations for the period around the announcement, starting a day -251 and ending at day +3 relative to the event. That is, the first 248 days (-251 to -4) was desi gnated for the estimation window, and the following 7 days (-3 to +3) was designated fo r the event window. The purpose the event window was set from -3 to +3 is to obtain the full effects of the event. The 3 days period accounts for information leakage pr ior to the announcement, and the + 3 days period accounts for information spread after the announcement. For the market model parameter estimates, while some research use the ordinary least squares (OLS) model, this study employed the Scholes-Williams st andardized cross-sectional market model (Scholes &Williams, 1977) because the Schol es-Williams standardized cross-sectional market model is designed to eliminate t he problems associated with nonsynchronous trading that sometimes occurs in event-bas ed studies with firms of widely varying market values (Cornwell et al., 2005a). Following the standard practice (Cornwe ll et al., 2005a; Pruitt et al., 2004), all statistical calculations were perform ed using the EVENTUS program developed by Cowan Research, L.L.C. The EVENTUS program is specifically designed for the event 41

PAGE 42

42 study analysis using the CRSP databases and the SAS statistics software. Specific procedure was followed the EVENTUS st udy guide manual 8.0 (Cowan, 2007). Multiple Regression Analysis and ANOVA To examine H2 through H7, multiple regression analysis and ANOVA test were executed. In this stages, the cumulative abnormal return (CAR) (i.e., stock price change) recorded by each title sponsor over the event windows t = 0 to t = +1 served as the dependent variable, while selected possible moderators served as the independent variables. For these tests, the SPSS 18.0 program was used.

PAGE 43

Table 4-1. PGA title sponsorship sample Sport Company name Title sponsorship name Announcement date Event date PGA SEI INVESTMENTS SEI Pennsylvania Classic 1/10/2000 9/14/2000 PGA VERIZON COMMUNICATIONS Genuity Championship 11/20/2000 3/1/2001 PGA ELECTRONIC DATA SYSTEM EDS Byron Nelson Championship4/12/2001 5/15/2003 PGA WACHOVIA Wachovia Championship 5/6/2002 5/8/2003 PGA COCA COLA The Tour Championship presented by Coca Cola 5/11/2002 10/31/2002 PGA BANK OF AMERICA Bank of America Colonial 5/20/2002 5/22/2003 PGA FORD MOTOR Ford Championship at Doral 9/16/2002 3/6/2003 PGA DEUTSCHE BANK Deutsche Bank Championship 11/14/2002 8/29/2003 PGA FRIEDMAN BILLINGS RAMSEY GRP FBR Capital Open 3/19/2003 6/5/2003 PGA GENERAL MOTORS Buick Championship 7/29/2003 2/12/2004 PGA FRIEDMAN BILLINGS RAMSEY GRP FBR Open 10/8/2003 1/29/2004 PGA I C O S Cialis Western Open 1/23/2004 7/1/2004 PGA LILLY ELI Cialis Western Open 1/23/2004 7/1/2004 PGA U.S. BANK U.S. Bank in Milwaukee 3/31/2004 7/22/2004 PGA BARCLAYS Barclays Classic 10/12/2004 6/23/2005 PGA ST PAUL TRAVELERS Travelers Championship 4/11/2006 6/21/2007 PGA INTERCONTINENTAL HOTELS GRP Crowne Plaza Invitational at Colonial 7/25/2006 5/24/2007 PGA A T & T AT&T National 3/7/2007 7/5/2007 PGA BANCO POPULAR Puerto Rico Open presented by Banco Popular 4/9/2007 3/20/2008 PGA WAL MART STORES Children's Miracle Network Classic presented by Wal Mart7/24/2007 11/1/2007 PGA NORTHERN TRUST Northern Trust Open 10/15/2007 2/14/2008 PGA ROYAL BANK CANADA RBC Canadian Open 11/1/2007 7/24/2008 PGA HEWLETT PACKARD HP Byron Nelson Championship 10/2/2008 N/A*1*1 Eventusprogramcould retrieve stock price data before 12/31/2008 andthe event was played in March 2009 43

PAGE 44

Table 4-2. LPGA title sponsorship sample Sport Company Name Title Sponsorship Name Announcement Date Event Date LPGA SYBASE Sybase Big Apple Classic 1/9/2001 7/19/2001 LPGA OFFICE DEPOT The Office Depot Hos ted by Amy Alcott3/26/2001 4/12/2001 LPGA TYCO INTERNATIONAL Tyco/ADT Championship 5/9/2001 11/15/2001 LPGA SAFEWAY Safeway Ping 7/2/2003 3/18/2004 LPGA H S B C HOLDINGS HSBC Wome n's World Match Play Championship1/25/2005 6/30/2005 LPGA CANADIAN NATIONAL RAILWAY CN Canadian Wo men's Open 10/19/2005 8/10/2006 LPGA HONDA MOTOR Honda LPGA Thailand 5/17/2006 10/20/2006 LPGA NAVISTAR INTERNATIONAL Navistar LPGA Classic 7/6/2006 9/27/2007 LPGA H S B C HOLDINGS HSBC Wome n's Champions 8/30/2007 2/28/2008 LPGA BELL MICROPRODUCTS Bell Micro LP GA Classic 1/10/2008 9/11/2008 LPGA PROCTER & GAMBLE P&G Beauty NW Arkansas Classic 1/31/2008 7/4/2008 44

PAGE 45

45 Table 4-3. NASCAR Sprint se ries title sponsorship sample Sport Company Name Title Sponsorship Name Announcement Date Event Date NASCAR (Sprint) DAIMLERCHRYSLER UAW Daimler Chrysler 40010/19/2000 3/5/2001 NASCAR (Sprint) HARRAH'S ENTERTAINMENT Harrah's 50011/17/2000 4/2/2001 NASCAR (Sprint) PEPSI Tropicana 4003/7/2001 7/16/2001 NASCAR (Sprint) AARON RENTS Aaron's 4994/8/2002 4/22/2002 NASCAR (Sprint) SIEMENS Sylvania 3001/30/2003 9/15/2003 NASCAR (Sprint) M B N A Bass Pro Shops MBNA 5002/4/2003 3/10/2003 NASCAR (Sprint) FORD MOTOR UAW Ford 5002/22/2005 10/3/2005 NASCAR (Sprint) ALLSTATE Allstate 400 at the Brickyard 4/28/2005 8/8/2005 NASCAR (Sprint) U S G USG Sheetrock 4005/16/2005 7/11/2005 NASCAR (Sprint) BANK OF AMERICA Bank of America 5006/29/2005 10/16/2006 NASCAR (Sprint) SONY Sony HD 5009/1/2005 9/6/2005 NASCAR (Sprint) TOYOTA MOTOR Toyota/Save Mart 3501/23/2006 6/25/2007 NASCAR (Sprint) DIRECTV GROUP DirecTV 5003/7/2006 4/3/2006 NASCAR (Sprint) DIAGEO Crown Royal 4004/17/2006 5/8/2006 NASCAR (Sprint) NEWELL RUBBERMAID Lenox Industrial Tools 3006/22/2006 7/17/2006 NASCAR (Sprint) ADVANCED MICRO DEVICES AMD at the Glen 8/3/2006 8/14/2006 NASCAR (Sprint) 3M 3M Performance 4001/31/2007 8/21/2007 NASCAR (Sprint) LOWES COMPANIES Kobalt Tools 500 3/5/2007 3/19/2007 NASCAR (Sprint) CITIZENS BANKING Citizens Bank 4004/9/2007 6/18/2007 NASCAR (Sprint) COCA COLA Coke Zero 400 presented by Coca Cola 12/7/2007 7/7/2008 NASCAR (Sprint) PEPSI Pepsi 5002/19/2008 9/2/2008 NASCAR (Sprint) SUNOCO Sunoco Red Cross Pennsylvania 500 7/2/2008 8/4/2008

PAGE 46

Table 4-4. NASCAR Nationwide se ries title sponsorship sample Sport Company Name Title Sponsorship Name Announcement Date Event Date NASCAR (Nationwide) KELLOGG Cheez It 250 11/20/2000 3/26/2001 NASCAR (Nationwide) C V S CVS Pharmacy 200 P resented by Bayer 3/31/2001 5/14/2001 NASCAR (Nationwide) SARA LEE Sam's Club P resents Hill Bros. Coffee 3006/13/2001 7/16/2001 NASCAR (Nationwide) KROGER COMPANY Kroger 300 P resented by Oreo 4/16/2002 6/17/2002 NASCAR (Nationwide) KRAFT FOODS Kroger 300 P resented by Oreo 4/16/2002 6/17/2002 NASCAR (Nationwide) PEPSI Tropicana Twist er 3004/18/2002 7/15/2002 NASCAR (Nationwide) UNITED ONLINE NetZero 2506/17/2002 7/29/2002 NASCAR (Nationwide) HERSHEY FOODS Koolerz 3002/12/2003 2/18/2003 NASCAR (Nationwide) KRAFT FOODS Meijer 300 Presented by Oreo 3/17/2003 6/16/2003 NASCAR (Nationwide) I T T INDUSTRIES Goulds Pumps ITT Industries 200 4/16/2003 5/19/2003 NASCAR (Nationwide) WINN DIXIE STORES Winn Dixie 2008/23/2003 9/2/2003 NASCAR (Nationwide) NEWELL RUBBERMAID Sharpie Professional 25011/13/2003 3/29/2004 NASCAR (Nationwide) GLAXOSMITHKLINE Goody's Headache Powder 2002/11/2004 2/23/2004 NASCAR (Nationwide) VIACOM SpongeBob Squa rePants Movie 3005/27/2004 10/15/2004 NASCAR (Nationwide) DOLLAR GENERAL Dollar General 3001/21/2005 10/14/2005 NASCAR (Nationwide) S B C COMMUNICATIONS SBC 2504/7/2005 6/27/2005 NASCAR (Nationwide) U S G USG Durock 300 5/16/2005 7/11/2005 NASCAR (Nationwide) YELLOW ROADWAY United Way 300 Presented by Yellow Transportation 5/20/2005 10/11/2005 NASCAR (Nationwide) DOMINOS PIZZA Domino's Piz za 2508/9/2005 8/22/2005 NASCAR (Nationwide) GLAXOSMITHKLINE Nicorette 3001/25/2006 3/20/2006 NASCAR (Nationwide) GLAXOSMITHKLINE Goody's 2505/14/2006 7/24/2006 NASCAR (Nationwide) GENUINE PARTS NAPA Auto Parts 2004/16/2007 8/6/2007 NASCAR (Nationwide) BANCO SANTANDER RoadLoans.com 2006/1/2007 9/24/2007 NASCAR (Nationwide) COCA COLA Winn Dixie 250 Powe red by Coca Cola 12/7/2007 7/7/2008 NASCAR (Nationwide) COCA COLA Dollar General 300 Powered by Coca Cola 12/7/2007 7/11/2008 NASCAR (Nationwide) DAIMLERCHRYSLER Missouri Illinois Dodge Dealers 250 1/25/2008 7/21/2008 NASCAR (Nationwide) UNILEVER Lipton Tea 2504/17/2008 5/2/2008 NASCAR (Nationwide) KROGER COMPANY Kroger On Track for the Cure 250 8/27/2008 10/27/2008 46

PAGE 47

CHA PTER 5 RESULTS This study is comprised of two parts: (1) event study analysis to examine stock price fluctuations of new title sponsors in both title sponsorship announcement date and the title sponsorship event date ( H1), and (2) exploration of the selected moderator variables that might influence stock price fl uctuations by using the results of (1) ( H2 through H7). Event Study Analysis Announcement Date The first stage of event study analysis verifies whether announcement of new title sponsorship agreement in the PGA, LPGA, and NASCAR brings significance in the title sponsors' stock prices. Table 5-1 reports the results of the mean abnormal returns (MAR) and their associated statisti cs tests for the interval from t = -3 to +3 for the overall 84 samples of title sponsorship announcements. Table 5-1 also reports the number of announcements in the sample (N), the number of companies registering positive and negative abnormal return changes (Positive: Negative), and the associated statistics tests (Patell Z and Generalized Sign Z) for th is fraction for each event day. The Patell test, which usually referred as the Z statistic and known as the standardized abnormal returns test, assumes cross sectional independence among the samples. The null hypothesis for the Patell test is that t he mean abnormal returns for each event day should be approximate zero. The Generalized Si gn Z test is a nonparametric test for the significance of a differing proportion of positiv e or negative returns in the sample period when compared with the estimation period (Pandey, Shanahan, & Hansen, 2005). That is, this test controls for the normal asymmetry of positive and negative abnormal returns 47

PAGE 48

in the estimation period. The null hypothesis under the Generalized Sign Z test is that the fraction of positive returns is the sa me as that in the estimation period. Table 5-2 shows the results of the mean cumulative abno rmal return (MCAR) and their associated statistics tests over th ree different event windows surrounding the title sponsorship. The event windows includ e t = (0, 1), (0, 2), and (0, 3). Becker-Olsen (2003) noted that several event windows need to be assessed in order to obtain the full effects of the event. According to the results obtained, there has not been detected any abnormal stock market returns around the sponsorsh ip announcement dates. This means that stock market has seen title sponsorship announcements in the PGA, LPGA, and NASCAR as neither positive nor negative events (at least statistically significant level) for sponsoring companies. Therefore, H1 was not confirmed. In event study analysis, however, it must be emphasized that no significant reaction on stock price does not necessarily mean that the company has not received enough benefits or suffered a net loss by the title sponsorship. Rather, it should mean that the company is receiving financial benefits from the event exactly same amount as the cost they paid for the event. That is, stock market has seen titl e sponsors in the PGA, LGPA, and NASCAR are likely to receive some benefits that are fi nancially equivalent to the title sponsorship fee investments in the future due to the new relationship with the sporting event. In other words, market assumes title spons orship fees are set for fair prices. The study of Cornwell et al. (2005a) demonstrated that stock price changes following the sponsorship announcements vary among the different sports entities (i.e, NFL, MLB, NHL, NBA, and PGA). Also, pas t research only focused on the NASCAR 48

PAGE 49

sponsors demonstrated that they experi enced significant abnormal stock price increases following the sponsorship anno uncements (P ruitt et al., 2004). These research imply that there is a possibility that insignificant events may conceal some important trends in the data by crossing off some significances in the data with their insignificances when averaging all data of events regardless of the sport entity. Therefore, the researcher conducted event study analysis in the PGA, LPGA, and NASCAR respectively. The each result is as follows. PGA Table 5-3 and Table 5-4 show mean abnormal returns (MAR) and mean cumulative abnormal returns (MCAR) for the title sponsorship announcements in PGA. As shown in the Table 5-3, title sponsor s for PGA experienced negative stock market returns on the whole including statistically significant level at 10% on the title sponsorship announcement date and at 5% le vel on the 2 days after the announcement date. Table 5-4 also shows statistically ne gative significances on the (0, +2) window. These results indicate that market sees ent ering title sponsorship for PGA tour brings financially net negative for the co mpany and that title sponsor ship fee for PGA tour may be too expensive when considering w hat they get by the sponsorship. LPGA Table 5-5 and Table 5-6 show MAR and MCAR for the LPGA title sponsors. Interestingly, as opposed to the results in PGA, title sponsors in the LPGA experienced significant stock price increases on the +1 day after the announcemen t. In fact, 9 of 11 companies received positive abnormal retu rns on the day. While Patell test reported some negative significances at 10% level around the announcement date, they are far more than 2 days. In general, the longer the event window, the more difficult it is to 49

PAGE 50

claim that the signific ance was caused by the event because of the increased possibility of confounding effects (McWilliams & Siegel, 1997).When considering that the positive significance was recorded on the + 1day which is immediately after the announcement, with 5 % statistical significant level, this positive significance would better reflect the market reaction toward title sponsorsh ip in the LPGA, than do those negative significances at 10 % level in far more than 2 days. This result indicates that stock market sees favorably entering title sponsorsh ip for the LPGA and a ssumes that the title sponsors are likely to get financial benefits more than what they paid for the rights fee. NASCAR (Sprint series) Table 5-7 and Table 5-8 show MAR and MCAR for the NASCAR Sp rint series title sponsors. The results present that NASCAR Sprint series sponsors received significant stock market returns on the announcement day. The positive significance of MCAR on the (0, +1) window also indi cates that stock market has favorably seen being title sponsor for the NASCAR Sprint series for increasing future profits of the company. NASCAR (Nationwide series) Table 5-9 and Table 5-10 show MAR and MCAR for the NASCAR Nationwide series title sponsors. Interestingly, un like the Sprint case, MAR and MCAR did not report any significance in the Nationwide seri es with an exception of +1 day negative significance at 10% significance level. Thes e different results between Sprint series sponsors and Nationwide series sponsors indica tes that the level of competition in NASCAR matters in gaining stock market returns and t hat stock market are only favorable for title sponsors in the higher level event (i.e., Sprint series sponsors), while lower level sponsors (i.e., Nationwide se ries sponsors) are seen as neutral. 50

PAGE 51

In summary, LPGA and NASCAR Sprint sponsors received significant stock market returns following t he title sponsorship announcements, while PGA sponsors experienced significantly negative decreas es in the stock prices and NASCAR Nationwide sponsors did not neither positive nor negative. These results confirmed that stock market reactions toward sport spons orship are different among sport entities (Cornwell el al, 2005) and therefore, resear ch should be conducted separately based on the sports in order to fully measure the impact of title sponsorship on stock prices. Event Date The second stage of event study analysis ve rifies whether title sponsorship event following the title sponsorship announcement in the PGA, LPGA, and NASCAR brings significance in the title sponsors' stock pr ices. Table 5-11 reports the results of the mean abnormal returns (MAR) and t heir associated statistics test s for the interval from t = -3 to +3 for the overall 81 samples of title sponsorship announcements. 1 sample from the PGA and 2 samples from the LGPA were not included for this event date analysis because the EVENUS program coul d not retrieve the precise stock price information from the date. Table 5-12 show s the results of the mean cumulative abnormal return (MCAR) and their associated statistics tests ov er three different ev ent windows (0, 1), (0, 2), and (0, 3) same as the announcement date. Table 5-11 shows that title sponsor s in the PGA, LPGA, and NASCAR experienced significant stock ma rket returns at 5% statistica l significant level on 2 days before the title sponsorship event. Please no te that unlike the case of announcement date, stock market's significant reacti on before the event does not mean information leakage and after the event does not mean market's delay of reaction or interpretation because the event date has been made public (u sually long ago before the event) since 51

PAGE 52

the announcement day. That is, this analys is was conducted under the situation that market knows the event day in advance and, in such condition, demonstrated that stocks of title sponsors were significantly bought by market before 2 days of the event. As well as the announcement date, the re searcher conducted further event study analysis focusing on each PGA, LPGA, NASCAR Sprint, and NACAR Nationwide sponsors respectively and the results were obtained as follows. PGA Table 5-13 and Table 5-14 show MAR and MC AR for the PGA title sponsors. As expected, +2 day (in most samp les, it is Monday which is t he first day market available after the PGA and LPGA event) observed significant market reaction which is negative at 5% statistical significant level. In fact 16 of 22 title sponsors in the PGA received negative abnormal returns on the next mark et available day following the title sponsorship event. These results are consistent with those in the announcement date and confirmed that stock market has seen bei ng title sponsor for the PGA tour brings financially net negative for the company. LPGA Table 5-15 and Table 5-16 show MAR and MCAR for the LPGA title sponsors. As expected again, +2 day observed significant ma rket reaction which is positive at 5% level. In fact, 7 of 9 title sponsors in t he LPGA received abnormal stock market returns on Monday and the mean 1.02% is the most st riking increases observed in this study. These results are consistent with those in the announcement date and put an another evidence that title sponsors in the LPGA are favorably seen by stock market and that market believes being title sponsors in the LPGA produces more financial value than the cost of sponsorship. 52

PAGE 53

NAS CAR (Sprint series) Table 5-17 and Table 5-18 show MAR and MC AR for the NASCAR Sprint series title sponsors. As expected, on the event 0 day (in most samples, it is Monday which is the first market available day after the NASCAR event) observed significant abnormal stock market returns at 10% level. Also, a ll MCAR windows report that the title sponsors are receiving significant stock price increases after the event date including statistical significance 5% level on the (0, +2) window. T hese results are consistent with those in the announcement date and confi rmed that stock market has seen being title sponsor for the NASCAR Sprint series brings financially net surplus for the company. NASCAR (Nationwide series) Table 5-19 and Table 5-20 show MAR and MCAR for the NASCAR Nationwide series title sponsors. Table 5-19 reports mixed results that are event 0 day with significant positive stock price increases and +3 day with significant negative stock price decreases. As noted earlier, when consideri ng that event window nearer to t = 0 day would better reflect the market's reaction towa rd the event than do further day from the 0 day because of the controlling problem of the confounding effe cts. (McWilliams & Siegel, 1997), the positive results recor ded on the 0 day should be seen as primary stock market reaction toward the event. Then, this result is interesting because title sponsors in the NASCAR Nationwide series received neither positive nor negative returns from the sponsorship announcement (Tabl e 2-d & Table 3-d). That is, the result is not consistent with those in the announcement date and demonst rating that stock market positively responds only for the ev ent day, not for the announcement day for the NASCAR Nationwide title sponsors. 53

PAGE 54

In summary, LPGA and NASCAR Sprint sponsors received signific ant stock market returns after the title sponsorship event, while PGA sponsors experienced significantly negative decreases after the event date. These results are consistent with those they received from the announcement date. However, the result of NASCAR Nationwide sponsors is not consistent with t hose in the announcement date in that only event date produced significantly positive sto ck market returns. The different results between announcement date and event date in the NASCAR Nationwide sponsors suggest that research focusing on not onl y announcement date but also event date is necessary in order to fully measure the impact of title sponsorship on stock prices. Multiple Regression and ANOVA Test The second purpose of this study is to explore the existence of possible moderators in title sponsors stock price changes following the announcement by employing the results obtained a bove event studies. For the purse of this examination, multiple regression analysis and ANOVA test were carried out as follows. As noted, CAR window (0, +1) in the event studi es for announcement date served as the dependent variable, while hypothesized possibl e moderators served as the independent variables. Multiple Regressions For the examination of whether Image Congruence ( H2), Corporate vs Brand Name ( H3) and prestige of event including Amount of Purse ( H4), Traditionality ( H5), and Event Competition Level (Sprint/Nationwide) (H6) were significant predictors of abnormal stock market returns, multiple regression analysis for such moderators was carried out. As noted earlier, Image Congruence ( H the 2) and Corporate vs Brand Name ( H3) were examined for all PGA, LP GA, and NASCAR title sponsors, Amount of Purse 54

PAGE 55

( H4) and Traditionality ( H5) were done for PGA and LPGA sponsors, and Event Competition Level (Sprint/Nationwide) (H6) was done for NASCAR sponsors. Table 5-21 presents the result s of the regressions. The mo st striking result shown in Table 5-21 is that Image Congruence showed statistically significance at p <0.01 ( F =12.382, R2= 0.232, Adjusted R2= 0.215). The regression analysis for Image Congruence and Corporate vs Brand Name reports that those variables explained a total 23% variance on the abnormal stock market returns ( R2= 0.232). Therefore, H2 was confirmed. This result is consistent with numerous prior studies that highlighted importance of the degr ee of congruence in the relationship between sponsor and sponsored event in gaining better consumers responses toward sponsorship (Cornwell et al., 2005b; Gwinner & Eaton, 1999; McDani el, 1999). Prior studies using event study analysis tool also demonstrated that congr uence level between sponsor and sponsee is the determinant in gaining positive abnormal stock market returns (Cal deron-Martinez et al., 2005; Cornwell et al., 2001, Cornwell et al., 2005a). The result shown in this study demonstrated that in title sponsorship al so, image congruence level between sponsor and sponsee is the determinant for abnormal sto ck market returns. On the other hand, other variables including Corporate vs Brand Name Amount of Purse Traditionality and Event Competition Level (Sprint/Nationwide) did not indicate significance in the analysis ( p >0.05, F = 1.160, R the 2 = 0.074, Adjusted R2 = 0.100 for Amount of Purse and Traditionality regression, and p >0.05, F =0.318, R2= 0.007, Adjusted R2 = -0.140 for Event Competition Level regression). Therefore, H3 ,H4 ,H5 ,and H6 were rejected. Regarding the H6, however, this result is conflict with those obtained from event studies conducted for the NASCAR Sprint series s ponsors and the NASCAR Nationwide series 55

PAGE 56

56 sponsors respectively (See Table 5-7, Table 5-8, Table 5-9, Table 5-10). As noted, the results examined each level indi vidually showed that stock ma rket are only favorable for title sponsors in the higher level event (i.e., Sprint series sponsors), while lower level sponsors (i.e., Nationwide series sponsors) ar e seen as neutral. Therefore, this result reports the level of competit ion in NASCAR matters in gaining positive stock market returns, though statistic test comparing those did not show the significance. ANOVA for Industry Segment For the examination of whether Industry Segment ( H7) in title sponsors is the determinant of the abnormal stock market returns, ANOVA test was carried out among 3 industry segments including High-technology, Retails, and Services. Table 5-22 presents the result of ANOVA test. As shown in the Table, there were not any significant differences am ong the selected business segments ( p >0.05, F =2.667, M = .0046 for High-technology, M = .0031 for Retails, and M = -.0066 for Services). However, since past research found statistical significance for hightechnology company by regression analysi s by coding dummy variable 1 for Hightechnology company and otherwise 0 (Cornw ell et al., 2001; Cornwell et al., 2005a; Pruitt et al., 2004), the res earcher conducted a regre ssion analysis following their method. For the purpose of this analysis, the dummy variable High-tech was coded as 1 ( N =19 which refers to the first categoriz ation before the grouping for the ANOVA), and otherwise 0 (N =65 including all other segments besides high-tech companies in the first grouping). Table 7-a shows the result of th e regression analysis. The test found the significance at 5% level in the relati onship between high-technology companies and their positive abnormal stock market returns ( p <0.05, F =5.135, R2= 0.059, Adjusted R2= 0.047). Therefore, H7 was partially confirmed.

PAGE 57

Table 5-1. Mean abnormal returns (MAR) for t he full 84 announc ements in PGA, LPGA, and NASCAR Mean Abnormal Positive: Patell Generalized Day N Return Negative Z Sign Z -3 84 0.14% 46:38 0.809 0.969 -2 84 0.22% 36:48 0.135 -1.213 -1 84 -0.06% 40:44 -0.793 -0.340 0 84 0.24% 42:42 0.467 0.096 +1 84 -0.06% 40:44 -0.854 -0.340 +2 84 -0.11% 36:48 -0.844 -1.213 +3 84 -0.03% 45:39 -0.469 0.751 The symbols $,*,**, and *** denote statistical significance at the 0.10, 0.05, 0.01 and 0.001 levels, respectively, using a generic one-tail test. The symbols (,< or ),> etc. correspond to $,* and show the direction and generic one-tail significance of the generalized sign test. Table 5-2. Mean cumulative abnormal return (MCAR) for the full 84 announcements in PGA, LPGA, and NASCAR Mean Cumulative Precision Abnormal Weighted Positive: Patell Generalized Days N Return CAAR Negative Z Sign Z (0,+1) 84 0.18% -0.08% 41:43 -0.273 -0.122 (0,+2) 84 0.07% -0.24% 43:41 -0.711 0.314 (0,+3) 84 0.04% -0.34% 37:47 -0.850 -0.995 The symbols $,*,**, and *** denote statistical significance at the 0.10, 0.05, 0.01 and 0.001 levels, respectively, using a generic one-tail test. The symbols (,< or ),> etc. correspond to $,* and show the direction and generic one-tail significance of the generalized sign test. 57

PAGE 58

Table 5-3. Mean abnormal returns (MAR) for announcement date i n PGA Mean Abnormal Positive: Patell Generalized Day N Return Negative Z Sign Z -3 23 0.25% 16:7> 1.123 1.985* -2 23 0.04% 9:14 -0.343 -0.935 -1 23 -0.19% 10:13 -0.352 -0.518 0 23 0.00% 8:15( -0.577 -1.353$ +1 23 0.06% 12:11 0.844 0.316 +2 23 -0.48% 8:15( -2.697** -1.353$ +3 23 0.18% 14:9 0.058 1.150 The symbols $,*,**, and *** denote statistical significance at the 0.10, 0.05, 0.01 and 0.001 levels, respectively, using a generic one-tail test. The symbols (,< or ),> etc. correspond to $,* and show the direction and generic one-tail significance of the generalized sign test. Table 5-4. Mean cumulative abnormal return (MCAR) for announcement date in PGA Mean Cumulative Precision Abnormal Weighted Positive: Patell Generalized Days N Return CAAR Negative Z Sign Z (0,+1) 23 0.06% 0.08% 13:10 0.188 0.733 (0,+2) 23 -0.42% -0.72% 8:15( -1.403$ -1.353$ (0,+3) 23 -0.24% -0.71% 9:14 -1.186 -0.935 The symbols $,*,**, and *** denote statistical significance at the 0.10, 0.05, 0.01 and 0.001 levels, respectively, using a generic one-tail test. The symbols (,< or ),> etc. correspond to $,* and show the direction and generic one-tail significance of the generalized sign test. 58

PAGE 59

Table 5-5. Mean abnormal returns (MAR) for announcement date i n LPGA Mean Abnormal Positive: Patell Generalized Day N Return Negative Z Sign Z -3 11 0.08% 4:7 -0.432 -0.882 -2 11 -0.23% 5:6 -1.401$ -0.279 -1 11 -0.08% 5:6 0.336 -0.279 0 11 0.10% 4:7 -0.455 -0.882 +1 11 0.24% 9:2> 0.830 2.133* +2 11 -0.81% 4:7 -1.605$ -0.882 +3 11 -0.31% 5:6 -1.297$ -0.279 The symbols $,*,**, and *** denote statistical significance at the 0.10, 0.05, 0.01 and 0.001 levels, respectively, using a generic one-tail test. The symbols (,< or ),> etc. correspond to $,* and show the direction and generic one-tail significance of the generalized sign test. Table 5-6. Mean cumulative abnormal re turn (MCAR) for announcement date in LPGA Mean Cumulative Precision Abnormal Weighted Positive: Patell Generalized Days N Return CAAR Negative Z Sign Z (0,+1) 11 0.34% 0.15% 6:5 0.265 0.324 (0,+2) 11 -0.47% -0.49% 4:7 -0.710 -0.882 (0,+3) 11 -0.78% -1.00% 4:7 -1.263 -0.882 The symbols $,*,**, and *** denote statistical significance at the 0.10, 0.05, 0.01 and 0.001 levels, respectively, using a generic one-tail test. The symbols (,< or ),> etc. correspond to $,* and show the direction and generic one-tail significance of the generalized sign test. 59

PAGE 60

Table 5-7. Mean abnormal returns (MAR) for announcement date i n NASCAR (Sprint) Mean Abnormal Positive: Patell Generalized Day N Return Negative Z Sign Z -3 22 -0.04% 10:12 -0.593 -0.246 -2 22 0.41% 11:11 1.212 0.181 -1 22 -0.12% 12:10 -1.133 0.607 0 22 0.53% 15:7> 1.435$ 1.888* +1 22 -0.21% 11:11 -0.692 0.181 +2 22 -0.27% 9:13 -0.412 -0.673 +3 22 -0.15% 11:11 -0.198 0.181 The symbols $,*,**, and *** denote statistical significance at the 0.10, 0.05, 0.01 and 0.001 levels, respectively, using a generic one-tail test. The symbols (,< or ),> etc. correspond to $,* and show the direction and generic one-tail significance of the generalized sign test. Table 5-8. Mean cumulative abnormal return (MCAR) for announcement date in NASCAR (Sprint) Mean Cumulative Precision Abnormal Weighted Positive: Patell Generalized Days N Return CAAR Negative Z Sign Z (0,+1) 22 0.32% 0.19% 15:7> 0.525 1.888* (0,+2) 22 0.05% 0.09% 12:10 0.191 0.607 (0,+3) 22 -0.10% 0.03% 9:13 0.066 -0.673 The symbols $,*,**, and *** denote statistical significance at the 0.10, 0.05, 0.01 and 0.001 levels, respectively, using a generic one-tail test. The symbols (,< or ),> etc. correspond to $,* and show the direction and generic one-tail significance of the generalized sign test. 60

PAGE 61

Table 5-9. Mean abnormal returns (MAR) fo r announcement date i n N ASCAR (Nationwide) Mean Abnormal Positive: Patell Generalized Day N Return Negative Z Sign Z -3 28 -0.08% 14:14 0.130 0.162 -2 28 0.25% 12:16 0.132 -0.594 -1 28 0.12% 13:15 0.430 -0.216 0 28 0.14% 14:14 0.254 0.162 +1 28 -0.28% 12:16 -1.339$ -0.594 +2 28 0.03% 13:15 0.657 -0.216 +3 28 -0.21% 12:16 -0.853 -0.594 The symbols $,*,**, and *** denote statistical significance at the 0.10, 0.05, 0.01 and 0.001 levels, respectively, using a generic one-tail test. The symbols (,< or ),> etc. correspond to $,* and show the direction and generic one-tail significance of the generalized sign test. Table 5-10. Mean cumulative abnormal return (MCAR) for announcement date in NASCAR (Nationwide) Mean Cumulative Precision Abnormal Weighted Positive: Patell Generalized Days N Return CAAR Negative Z Sign Z (0,+1) 28 -0.13% -0.28% 14:14 -0.767 0.162 (0,+2) 28 -0.10% -0.11% 15:13 -0.247 0.540 (0,+3) 28 -0.31% -0.33% 13:15 -0.640 -0.216 The symbols $,*,**, and *** denote statistical significance at the 0.10, 0.05, 0.01 and 0.001 levels, respectively, using a generic one-tail test. The symbols (,< or ),> etc. correspond to $,* and show the direction and generic one-tail significance of the generalized sign test. 61

PAGE 62

Table 5-11. Mean abnormal returns (MAR) for the full 81 event date in PGA, LPGA, and NASCAR Mean Abnormal Positive: Patell Generalized Day N Return Negative Z Sign Z -3 81 0.19% 39:42 0.262 -0.365 -2 81 0.04% 48:33) 1.783* 1.635$ -1 81 0.01% 35:46 0.460 -1.254 0 81 0.17% 43:38 0.589 0.524 +1 81 0.07% 40:41 0.259 -0.143 +2 81 0.08% 41:40 0.693 0.079 +3 81 -0.08% 36:45 -1.139 -1.032 The symbols $,*,**, and *** denote statistical significance at the 0.10, 0.05, 0.01 and 0.001 levels, respectively, using a generic one-tail test. The symbols (,< or ),> etc. correspond to $,* and show the direction and generic one-tail significance of the generalized sign test. Table 5-12. Mean cumulative abno rmal return (MCAR) for the full 81 event date in PGA, LPGA, and NASCAR Mean Cumulative Precision Abnormal Weighted Positive: Patell Generalized Days N Return CAAR Negative Z Sign Z (0,+1) 81 0.24% 0.17% 40:41 0.600 -0.143 (0,+2) 81 0.32% 0.31% 42:39 0.890 0.302 (0,+3) 81 0.24% 0.08% 40:41 0.201 -0.143 The symbols $,*,**, and *** denote statistical significance at the 0.10, 0.05, 0.01 and 0.001 levels, respectively, using a generic one-tail test. The symbols (,< or ),> etc. correspond to $,* and show the direction and generic one-tail significance of the generalized sign test. 62

PAGE 63

Table 5-13. Mean abnormal returns (MAR) for event date in PGA Mean Abnormal Positive: Patell Generalized Day N Return Negative Z Sign Z -3 22 0.23% 10:12 0.026 -0.293 -2 22 0.02% 13:9 0.474 0.987 -1 22 0.28% 9:13 0.544 -0.719 0 22 -0.61% 9:13 -1.455$ -0.719 +1 22 -0.03% 11:11 -0.108 0.134 +2 22 -0.70% 6:16< -1.785* -1.999* +3 22 0.09% 10:12 0.303 -0.293 The symbols $,*,**, and *** denote statistical significance at the 0.10, 0.05, 0.01 and 0.001 levels, respectively, using a generic one-tail test. The symbols (,< or ),> etc. correspond to $,* and show the direction and generic one-tail significance of the generalized sign test. Table 5-14. Mean cumulative abnormal return (MCAR) for ev ent date in PGA Mean Cumulative Precision Abnormal Weighted Positive: Patell Generalized Days N Return CAAR Negative Z Sign Z (0,+1) 22 -0.65% -0.49% 8:14 -1.105 -1.146 (0,+2) 22 -1.35% -1.05% 7:15( -1.933* -1.572$ (0,+3) 22 -1.26% -0.96% 8:14 -1.523$ -1.146 The symbols $,*,**, and *** denote statistical significance at the 0.10, 0.05, 0.01 and 0.001 levels, respectively, using a generic one-tail test. The symbols (,< or ),> etc. correspond to $,* and show the direction and generic one-tail significance of the generalized sign test. 63

PAGE 64

Table 5-15. Mean abnormal returns (MAR) for event date in LPGA Mean Abnormal Positive: Patell Generalized Day N Return Negative Z Sign Z -3 9 -0.33% 4:5 -0.403 -0.293 -2 9 0.51% 4:5 2.117* -0.293 -1 9 0.29% 4:5 0.535 -0.293 0 9 0.15% 5:4 -0.658 0.374 +1 9 -0.30% 3:6 -0.567 -0.960 +2 9 1.02% 7:2> 2.469** 1.707* +3 9 -0.26% 4:5 -0.991 -0.293 The symbols $,*,**, and *** denote statistical significance at the 0.10, 0.05, 0.01 and 0.001 levels, respectively, using a generic one-tail test. The symbols (,< or ),> etc. correspond to $,* and show the direction and generic one-tail significance of the generalized sign test. Table 5-16. Mean cumulative abnormal re turn (MCAR) for event date in LPGA Mean Cumulative Precision Abnormal Weighted Positive: Patell Generalized Days N Return CAAR Negative Z Sign Z (0,+1) 9 -0.16% -0.50% 5:4 -0.866 0.374 (0,+2) 9 0.86% 0.51% 7:2> 0.718 1.707* (0,+3) 9 0.60% 0.10% 6:3 0.126 1.041 The symbols $,*,**, and *** denote statistical significance at the 0.10, 0.05, 0.01 and 0.001 levels, respectively, using a generic one-tail test. The symbols (,< or ),> etc. correspond to $,* and show the direction and generic one-tail significance of the generalized sign test. 64

PAGE 65

Table 5-17. Mean abnormal returns (MAR) for event date in NASCAR (Sprint) Mean Abnormal Positive: Patell Generalized Day N Return Negative Z Sign Z -3 22 0.47% 11:11 0.547 0.192 -2 22 -0.05% 13:9 -0.031 1.046 -1 22 -0.01% 10:12 0.061 -0.234 0 22 0.34% 15:7> 1.181 1.900* +1 22 0.46% 15:7> 1.216 1.900* +2 22 0.36% 12:10 0.231 0.619 +3 22 0.27% 11:11 0.325 0.192 The symbols $,*,**, and *** denote statistical significance at the 0.10, 0.05, 0.01 and 0.001 levels, respectively, using a generic one-tail test. The symbols (,< or ),> etc. correspond to $,* and show the direction and generic one-tail significance of the generalized sign test. Table 5-18. Mean cumulative abnormal return (MCAR ) for event date in NASCAR (Sprint) Mean Cumulative Precision Abnormal Weighted Positive: Patell Generalized Days N Return CAAR Negative Z Sign Z (0,+1) 22 0.80% 0.66% 14:8) 1.695* 1.473$ (0,+2) 22 1.15% 0.73% 16:6> 1.517$ 2.326** (0,+3) 22 1.42% 0.82% 14:8) 1.477$ 1.473$ The symbols $,*,**, and *** denote statistical significance at the 0.10, 0.05, 0.01 and 0.001 levels, respectively, using a generic one-tail test. The symbols (,< or ),> etc. correspond to $,* and show the direction and generic one-tail significance of the generalized sign test. 65

PAGE 66

Table 5-19. Mean abnormal returns (MAR) fo r event dates in NASCAR (Nationwide) Mean Abnormal Positive: Patell Generalized Day N Return Negative Z Sign Z -3 28 -0.36% 13:15 -0.629 -0.255 -2 28 -0.24% 16:12 0.728 0.880 -1 28 -0.45% 11:17 -0.789 -1.011 0 28 0.32% 18:10) 1.233 1.636$ +1 28 -0.07% 11:17 -0.322 -1.011 +2 28 -0.10% 14:14 0.294 0.124 +3 28 -0.44% 12:16 -2.801** -0.633 The symbols $,*,**, and *** denote statistical significance at the 0.10, 0.05, 0.01 and 0.001 levels, respectively, using a generic one-tail test. The symbols (,< or ),> etc. correspond to $,* and show the direction and generic one-tail significance of the generalized sign test. Table 5-20. Mean cumulative abnormal return (MCAR) for event dates in NASCAR (Nationwide) Mean Cumulative Precision Abnormal Weighted Positive: Patell Generalized Days N Return CAAR Negative Z Sign Z (0,+1) 28 0.25% 0.24% 14:14 0.644 0.124 (0,+2) 28 0.15% 0.32% 16:12 0.696 0.880 (0,+3) 28 -0.29% -0.42% 13:15 -0.798 -0.255 The symbols $,*,**, and *** denote statistical significance at the 0.10, 0.05, 0.01 and 0.001 levels, respectively, using a generic one-tail test. The symbols (,< or ),> etc. correspond to $,* and show the direction and generic one-tail significance of the generalized sign test. 66

PAGE 67

Table 5-21. Multiple Regression Analysis ALL PGA & LPGA NASCAR Variable Coefficient t statistics Significance Congruence *4.863 ---------0.000 Corp/Brand 0.509 --------0.612 Purse ----1.084 ----0.287 Tradition ----0.227 ----0.822 Competition Level --------0.564 0.757 N 84 32 50 FStatistic *12.382 1.160 0.318 Significance 0.000 0.327 0.575 R2 0.232 0.074 0.007 Adjusted R2 0.215 0.100 0.140 *shows statistical significance at less than 1% level 67

PAGE 68

68 Table 5-22. ANOVA for industry segment Cumulative Abnormal Returns (CAR) Industry Segment N Minimum Mean Maximum Std. Deviation High technology 32 .047 .0046 .038 .0177 Retails 23 .050 .0031 .061 .0204 Services 29 .054 .0066 .053 .0223 Fstatistic 2.667 Total 84 .054 .0003 .061 .0205 Significance .076 Table 5-23. Regression for high-technology company N 84(high tech 19, other 65) Coefficient t 2.266 FStatistic 5.135 Significance *0.026 R2 0.059 Adjusted R2 0.047 *shows statistical significance at 5% level

PAGE 69

CHA PTER 6 DISCUSSION AND LIMITATION Discussion Title sponsorship has been increasingly employed as a marketing communication tool, but its effectiveness and re turns are extremely difficult to measure. Likewise, the literature has not examined the impact of title sponsorship on the companys performance regardless of the market size. This study focused on the performance of title sponsorship in the PGA, LGPA, and NASCAR, by analyzing stock investors response on corporate sponsor's title sponsorship announcement. Also, this study investigated the company performance of ti tle sponsorship in sporting event date following the new title sponsorship announcement as well as the announcement date, for the first time in the s ports related literature. The main conclusion of the study is t hat title sponsorship s generally appear to neither positively nor negatively affect on the company's stock prices. However, in event study analysis, it must be emphasized that no significant reaction on stock price does not mean that the company has not rece ived enough benefit or suffered a net loss by the title sponsorship. Rather, it should mean that market assumes that the company has received financial benefits fr om the relationship with the event exactly for the same amount as the cost they paid for the sponsorship fee. In this respect, as long as the results in event study are neutral, the title sponsorship should be accepted and the decision whether entering titl e sponsorship or other alternative communication tools would be left to the extent of marketing preference of the company. Also, as demonstrated by the event study conducted fo r each sport entity respectively, not all title sponsorships are equal. As past studies have demonstrated, market has favorably 69

PAGE 70

seen NASCAR spons ors (Cornwell et al ., 2001; Pruitt et al., 2004). This study demonstrated it from a perspective of titl e sponsors stock price increases and added additional evidence that NASCAR can be a s pecial marketing platform for businesses. Also, in this study, it was demonstrated that the difference of NASCAR event competition level (i.e., Sprint series or Nationwide series) pl ays an important role in the stock price changes, and that more prestigious events (i.e., Sprint series) title sponsors are likely to be favored by market. In the ca se of golf event, titl e sponsors for the PGA experienced significant negative stock ma rket returns from both announcement date and following title sponsorship event date. Thes e results are consistent with opinions by a sport critique (Harig, 2009) and one of the PGAs title sponsor (Mell, 2009) noting that title sponsorship fee in the PGA are just too expensive. Annual 5% increase of title sponsorship fees in PGA may be one of the reasons that caused this negative significances. Or it may be just indicating a conflict of title sponsorship objectives between title sponsor who regards the PGA tour event as more hospitality opportunity to strengthen relationship with key customers a nd stock market who only seeks tangible and direct financial returns from the company' s investment. Interestingly, as opposed to the result of PGA, t he stock prices of title sponsors for LPGA reacted significantly positive following the new title sponsorship announcement. This finding implies that title sponsors for LGPA are seen by stock investor s to be likely to increase their company's performance including sales thanks to the titl e sponsorship, rather than what they paid for the title sponsorship fee. In the other words, the st ock market sees that title sponsorships for PGA are overvalued and those of LPGA are undervalued. For a more practical suggestion, the results of this stud y imply that title sponsorship fee for PGA 70

PAGE 71

should be lowered and for LPGA it could be ra ised when considering what title sponsors paid for and what they are likely to benefit by the new established relationship with the sport entity. Otherwise, PGA should make up the sponsors negative results in stock market returns by demonstrati ng additional financial benefit t hat stock market could not take into account as stock price increases but that title sponsor could get due to the sponsorship. By doing so, current title s ponsorship fee for PGA could keep sounding "reasonable". Emphasizing on non financial or not soon reflected as figures but surely fruitful later benefits would be another strat egy for PGA keeping the current sponsorship fee. Such examples include emphasizi ng on hospitality opportunity with key and potential customers, enhancement of empl oyees loyalty to the company, and enhancement of corporate image thr ough the partnership with PGA. Consistent with past findings, congruence was an important determinant of the stock price increases for title sponsors. This result confirmed that the already established value of high image congr uence between sponsor and sponsored event relationship in title sponsorship too. As we ll as the congruence level, this study also confirmed the past finding that high-technology companies are likely to get higher stock market returns from the announcement of title sponsorship agreement. As a new attempt in event study analysis this study also focused on actual sporting event date following the new titl e sponsorship announcement. As a result, some significance level of abnormal stock returns have been observed in title sponsorship event date as well as the announcement date. By these findings, it is suggested that the importance of taking into account the stock price changes around the event date besides the announcement date in order to fully measure the 71

PAGE 72

effectiveness of title sponsorship in terms of stock price changes. When considering that title sponsorship has advantage over traditional sport sponsorship and other advertising due to the longer duration, high visibility via nationwide broadcasting, and larger target audiences (Becker-Olsen, 2003; Stipp, 1998), it is reasonable that the actual title sponsorship event day fo llowing the new sponsor ship agreement had a positive impact on their stock prices. Specif ically, the results of NASCAR Nationwide series sponsors in the title sponsorship event date were not consistent with those in the announcement date in that only the event date produced significantly positive stock market returns. The different results between announcement date and event date in the NASCAR Nationwide sponsors suggest that research should focus not only on announcement date but also event date in order to fully m easure the impact of title sponsorship on stock prices. It is a common sense in both practice and research that measuring effectiveness of sponsorship is a really challenging i ssue. In addressing this subject, this study attempted to measure the effectiveness of sport sponsorship by regarding sponsors stock price change as represent ing one of the final outcomes of the sponsorship, which cannot be obtained from more traditional approaches such as exposure value conversion and consumer surveys. The researc her hopes that the resu lts of this study contribute to the literature in this regard and will be used to further development of sport sponsorship in the future. Limitations of the Study The main limitations of this study are the following three: (1) event study analysis, (2) the selection of the companies (i.e., title sponsors), and (3) the determination of the congruence level between title sponsor and sport event. 72

PAGE 73

First, employing event study analysis is both a strong and weak point in this study. While the literature and prevalence of this method itself have demonstrated the validity and reliability of this methodology, the researcher who uses this method needs to rely on some strong assumptions under the study. The existence of confounding effects around the event date is one of the most controversial assumptions in event study analysis. While the researcher carefully tried to choose the event that had no confounding effects likely to impact the stock prices around t he date, it seems impossible to affirm that there were no confounding effects around the date, and therefore, that the significance of stock price changes obtained from this study were caused 100% by the title sponsorship. Also, the researcher needs to mention that one of the assumptions that the event should be unexpected to public conflicts the second event study series in this study focusi ng on the event date fo llowing the announcement date. As noted, unlike the announ cement date, the event date is available for everyone and this fact is incompatible with the assu mption that the market is supposed to know the information from the announcem ent for the first time and event date in event study analysis should be unexpected to ma rket before the event date. Second, the selection of title sponsors c an lead to the existence of bias in the results. For example, in t he PGA 2007 tour, there were 14 new title sponsorships. Out of 14, four companies are private and their stocks are unav ailable to public, or they are exchanged in foreign markets. Five companies were not included because specific announcement dates were not obtained from t he Web search. As a result, remaining five companies of 14 from this years PGA tour were used for the analysis. This fact did 73

PAGE 74

not allow the researcher to generate the results obtaine d from this research and conclude the results represent all title sponsorship in the PGA, LPGA, and NASCAR. Finally, the determination of the leve l of congruence between title sponsor and sponsee can lead to the existence of bias in the results. In th is study, the image congruence level was determined by the resear cher after a careful discussion with an expert. In the future research, the level of image congruence should be determined by an extensive survey of the general public and it would enhance the validity of perceived image congruence level, and the conclusion that image congruence level between title sponsor and sponsored event plays an import ant role in stock price changes. Also, in the future research, other sport entities that have title sponsorships such as professional tennis tournament (ATP & WTP) and college football (NCAA) should be chosen for the topic. Title sponsorship in other countri es sport and markets would be an interesting topic as well. Repetition of this study in the future is also meaningful. Title sponsorship fee is generally decided by the supply and demand relationship between sport entity and potential corporate sponsors. T hus, when the equilibrium occurs in the title sponsorship supply and demand curve, it is the time that the title sponsorship fee equals the benefits the title sponsor receive. In this respec t, any stock price change significances recorded in this study show discrepancy between demand and supply in title sponsorship, and the gap would be fixed based on the market prin ciple in due course. Therefore, once the title sponsorship fee and its trend have been changed by the sport entity, the results in event study analysis would be different from t hose in this study. The results gained in this study only reflect the current situation between title sponsors and their stock price 74

PAGE 75

75 performances. In this regard, repetition of this study woul d be beneficial to evaluate the current title sponsorship fee in the PGA, LPGA, and NASCAR, and the results could be used as a guidance to accommodate the cu rrent sponsorship fee or to decide appropriate title sponsorship fee in the following year.

PAGE 76

APPENDIX SPONSOR-EVENT FI T SCALE Speed & Thompsons (2000) spons or-event fit scale items 5 Questions used to determine the level of congruence are; Q1. There is a logical connection between the event and the sponsor. Q2. The image of the event and the im age of the sponsor are similar. Q3. The sponsor and the event fit together well. Q4. The company and the event stand for similar things. Q5. It makes sense to me that th is company sponsors this event. 7 point Liker type scale Strongly Disagree Strongly Agree 1 2 3 4 5 6 7 As noted in the data collection section, the answer of the 5 questions was averaged and then the average was used as the congruenc e level between sponsor and event. This procedure was taken in all 84 title sponsorship samples. 76

PAGE 77

LIST OF REFERENCES Becker-Olsen, K. (2003). Questioning the name game: An event study analysis of stadium naming rights sponsorship announcements. International Journal of Sports Marketing & Sponsorship, 5 (3), 181-192. Brown, S., & Warner, J. (1985) Using daily stock returns: The case of event studies. Journal of Financial Economics, 14, 3-31. Calderon-Martinez, A., Mas-Ruiz, F. J., & Nicolau-Gonzalbez, J. L. (2005). Commercial and philanthropic sponsorship: Direct and interaction effects on company performance. International Journal of Market Research 47(1), 75. Clark, J. M., Cornwell, T. B., & Pruitt, S. W. (2002). Corporate stadium sponsorship, signaling theory, agency conflic ts and shareholder wealth. Journal of Advertising Research, 42 (6), 16 32. Copeland, R., Frisby, W., & McCarville, R. ( 1996). Understanding the sport sponsorship process from a corporate perspective. Journal of Sport Management, 10(1), 32-48. Cornwell, T. B., & Maigan, I. (1998). An inte rnational review of sponsorship research, Journal of Advertising. 27(1), 1-22. Cornwell, T. B., Pruitt, S. W. & Van Ness, R. (2001). An exploratory analysis of the value of winning in motorsports: Sponsorship-linked market ing and shareholder wealth. Journal of Advertising Research 41, 17. Cornwell, T. B., Pruitt, S. W., & Clark, J. M. (2005a). The Relationship between majorleague sports official sponsorship announcements and the stock prices of sponsoring firms. Journal of the Academy of Marketing Science 33 (4), 401-412. Cornwell, T. B., Weeks, C., & Roy, D. P. (2005b). Sponsorsh ip-linked marketing: opening the blackbox. Journal of Advertising 34(2), 23-45. Cowan, A. R. (2007). Eventus 8.0 Users Guide Iowa State Universi ty Press, Ames IA: Cowan Research. Crompton, J. (2004) Concept ualization and alternate oper ationalizations of the measurement of sponsorship effectiveness in sport. Leisure Studies 23(3), 267281. Farell, K. A., & Frame, W. S. (1997). T he value of Olympic sponsorships: Who is capturing the gold? Journal of Market Focused Management 2 (2), 171. Ferguson, D. (2009). Buick ends title sponsor ships with all PGA Tour events, including Buick Invitational. Retrieved October 15, 2009 from Sports News: http://blog.taragana.com/sports/2009/08/04/ buick-ends -title-sponsorships-with-allpga-tour-events-includi ng-buick-invitational-17579/ 77

PAGE 78

General Accounting Office report. (1993). 1992 Olympic sponsorship-profit or loss is unknown. Retrieved November 17, 2008, from LEGISTORM web site: http://www.legistorm.co m/score_gao/show/id/23325.html Giannoulak is, C. & Stotlar, D. (2006). Evolution of Olympic sponsorship and its impact on the Olympic movement. Retr ieved October 20, 2009 from http://findarticles.com/p/articles/ mi_7091/is_2006_Oct/a i_n2850 3594/pg_2/?tag=c ontent;col1 Giannoulak is, C., Stotlar, D., & Chatzief stathiou, D. (2008). Olympic sponsorship: evolution, challenges and impact of the Olympic Movement. International Journal of Sport Marketing & Sponsorship, 9 (4), 256-270. Grunert, K. G. (1996) Automatic and strat egic processes in advertising effects. Journal of Marketing 60(4), 88. Gwinner, K. (1997). A m odel of image creation and image transfer in event sponsorship. International Marketing Review 14(3), 145. Gwinner, K., & Eaton, J. (1999). Building brand image through event sponsorship: The role of image transfer. Journal of Advertising, 28(4), 47-57. Harig, B. (2009). Tours sponsorship woe creating uncertainty. Retrieved October 28, 2009 from ESPN.com: http://sports.espn.go.com/golf/notebook?page=not ebook/bb-090805 IEG Sponsorship Report. (2000). Thatlook.c om sets sights on more sponsorship. Sponsorship Report 19 (5), 3. IEG Sponsorship Report. (2006). Projection: sponsorship growth to increase for fifth straight year. retrieved November 17, 2008, from the IE G Sponsorship Report Sample Issue: http://www.sponsorship.com/documents/SR_Promo_Issue_0107.pdf IEG Press Release. (2009). Sponsorship sp ending to rise 2.2% in 2009. Retrieved September 25, 2009 from http://www.sponsorship.com/About-IEG/PressRoom/Sponsorship-Spending-ToRise-2.2-Percent-in-2009.aspx Ko, Y. J., K im, K., Claussen, C. L., & Kim T. H. (2008). The effects of sport involvement, sponsor awareness and corporate image on intention to purchase sponsors products. International Journal of Spor ts Marketing and Sponsorship, 9 (2), 79-94. MacKinlay, A. C. (1997). Event studies in economics and finance. Journal of Economics Literature, 35 (1), 13-39. McWilliams, A., & Siegel, D. (1997). Event st udies in management research: Theoretical and empirical issues. Academy of Management Journal 40(3), 626-657. 78

PAGE 79

Meenaghan, J. (1991). The role of sponsor ship in the marketing communications mix International Journal of Advertising 10, 35-47. Mell, R. (2009). Deutsche bank CEO scrutin ized PGA Tour purses. Retrieved October 25, 2009 from: http://www.thegolfchannel.com/tour -insider/deutsche-bank-ceoscrutinizes-pga-t our-purses-31959/ Miyazaki, A.D., & Morg an, A.G. (2001). Assessing the market value of event sponsorship. Journal of Advertising Research 41(1), 9-16 Mullin, B., Hardy, S., & Sutton, W. (2000). Sport Marketing Champaign. Illinois / Leeds, UK: Human Kinetics Oztur, M.A., Kozub, F. M., & Kocak, S. (2004). Impact of sponsorship on companies that supported the 2002 Salt Lake City winter Paralympics. International Journal of Sports Marketing & Sponsorship, 5 (4), 282-295. Pandey, V. K., Shanahan, K. J., & Hansen, S. W. (2005). The relationship between shareholder wealth effects, diversity, and public ity as a marketing strategy. Journal of the Academy of Marketing Science 33(4), 423-431. Pham, M. (1991). Evaluation of sponsor ship effectiveness; a model and some methodological cons iderations. Gestion 2000 47-64. Poknywczynski, J. (2000). Sport sponsors know what they want. Sports Business Journal, 5 (11), 2 Pruitt, S. W., Cornwell, T. B., & Clark, J.M. (2004). The NASCAR phenomenon: auto racing sponsorship and shareholder wealth. Journal of Advertising Research 44(3), 281-296. Roy, D. P., & Cornwell, T. B. (1999). Managers use of spons orship in building brands: service and product firms contrasted. International Journal of Sports Marketing and Sponsorship 1 (6), 345-360. Russell, M., Mahar, J., & Drewniak, B. (2005). Examination of stock market responses to publicity surrounding athletic endorsers. The Marketing Management Journal 15(2), 67-79. Samitas, A., Kenourgios, D., & Zounis, P. (2008). Athens Olympic games 2004 impact of sponsors stock returns. applied Financial Economics 18(19),1569-1580. Sandler, D.M., & Shani, D. (1993). Sponsor ship and the Olympic games: the consumer perspective. Sport Marketing Quarterly, 2 (3), 38-43. Show, J. (2009). Prices steady for golf title sponsors. Street & Smiths Sportsbusiness Journal, 12 (6), 4. 79

PAGE 80

80 Spais, G. S., & Filis, G. N. (2006). Stock market reaction on Olympic sponsorship announcement using event-study method. Journal of Korean Academy of Marketing Science 16 (1), 95-108. Spais, G. S., & Filis, G. N. (2008). Measuring stock market reac tion to sponsorship announcements: The case of Fiat and Juventus. Journal of Targeting, Measurement and Analysis for Marketing 16(3), 169-180. Speed, R., & Thompson, P. (2000). Determi nants of sports sponsorship response. Journal of the Academy of Marketing Science 28(2), 226-238. Sponsor Map. (2008). Strong gr owth for the global sponso rship industry in 2008. Retrieved November 17, 2008, fr om the Sponsor Map Web site: http://www.sponsormap.com/?p=72 Stipp, H. (1998). The impact of Olym pic sponsorship on corporate image. International Journal of Advertising 17(1), 75-87. Stipp, H., & Schiavone, N. P. (1996). M odeling the impact of Olympic sponsorship on corporate image. Journal of Advertising Research, 36 (4), 22-28. Tripodi, J. (2001). Sponsorship a confirmed weapon in the promotional armoury. International Journal of Spor ts Marketing and Sponsorship, 20(1), 1-20.

PAGE 81

BIOGRAPHICAL SKETCH Masaki Kudo received his Bachelor of Arts in sports science from Waseda University, Japan in 2007. He then attended the University of Florida where he completed his Master of Science in sport management in 2010. His major research interests are measuring effectiveness of spor t sponsorship, cultural difference in sport sponsorship, and sport sponsorship as corporate social responsibility (CSR). 81