English forum

The Michael Jordan effect Crawford, Anthony J; Niendorf, Bruce . American Business Review ; West Haven  Vol. 17, Iss. 2,  (Jun

1999): 5-10.

ProQuest document link

ABSTRACT  

In the period immediately following the rumors of Michael Jordan's return to basketball, the five companies that

Jordan had major endorsement deals with experienced a nearly $3 billion increase in the market value of their

equity. Jordan was labeled the $2 billion man in the press, referring to the value he created for the shareholders of

the companies he endorses. However, most of these reports failed to cite the simultaneous bull market that lead

the S&P 500 to record highs. The Michael Jordan effect is examined, and it is found that shareholders experienced

negative abnormal returns after the announcement of his retirement and positive excess returns when rumors of a

comeback surfaced. However, it is also shown that the positive excess returns following the rumors of Jordan's

comeback were only temporary and disappeared within weeks of the original rumors. While some evidence is

found in support of a Michael Jordan effect, it appears that the rumors of Jordan's impact have been greatly

exaggerated. FULL TEXT  

On October 6, 1993 Michael Jordan unexpectedly retired from basketball after leading the Chicago Bull's to three

straight NBA championships. The following spring he showed up for spring training with the Chicago White Sox.

Despite his early retirement Jordan maintained his five major endorsement deals from which he is rumored to

receive a total of approximately $30 million annually. These endorsements are highlighted in exhibit 1.

On March 2nd, 1995 Jordan ended his attempt at professional baseball and left the Chicago White Sox spring

training facilities. Shortly after his retirement from baseball, rumors of Jordan's return to basketball surfaced.

These rumors touched off a media frenzy as the popular press tied increases in the stock prices of the companies

Jordan endorses to speculation over his return. The Los Angeles Times reported that five days after the first

reports of his comeback, advertisers experienced a collective $2.3 billion gain in equity value. Time Magazine,

Newsweek, Sports Illustrated and nearly every major newspaper ran similar reports at about the same time. Jordan

was labeled the $2 billion man. The implication was that Jordan's rumored return to basketball increased his value

as an endorser resulting in an over $2 billion dollar gain to shareholders.

Table I illustrates the increase in the market value of the five companies which Jordan endorses. The collective

increase in market value over a nine business day period, from the date of the first rumors until the first trading day

after his comeback announcement, was more than $2.9 billion. The average return over those nine days was

4.59%. The return on the S&P 500 over the same nine day period was just 2.91%.

We examine two questions. First, were the market value gains experienced by these five companies caused by

Jordan's comeback, or were these gains a result of a contemporaneous market upswing?' Second, if Jordan's

return to basketball impacted the companies he endorses, what effect did his original retirement have? We

examine the market adjusted returns for the companies endorsed by Jordan in an attempt to address these

questions.

More specifically, we examine the abnormal returns experienced by shareholders around Jordan's original

retirement and again at his subsequent return to basketball. Jordan announced his retirement from basketball

October 6, 1993 (t=0). We examine the abnormal returns on the stock of his five largest endorsement deals for a

two day interval beginning the day of the announcement and ending one day after the announcement (t=0,+1).

For Jordan's return to basketball, rumors preceeded the actual announcement as anticipation of a comeback

began almost two weeks prior to confirmation. Table II outlines the events surrounding his return. We examine the

daily abnormal returns for a period running from March 8, 1995, (the date of the first rumors), to March 20,1995,

(the first trading day after announcing his comeback). In addition, we examine the cumulative abnormal returns

over the entire nine day event period.

INTRODUCTION

Enlarge this image.

Product endorsements by celebrities are a common form of advertising in the United States. Celebrities from

sports, television, movies and even politics can be seen pitching everything from potato chips to underwear. Multi-

million dollar deals have become common place. What is the impact of celebrity endorsement on consumer

choice? Research by Petty et al. (1983) found that celebrity endorsement aides in brand recognition. Kamins et al.

(1989) found that endorsements make advertisements believable and creates a positive attitude toward brands.

While McCracken (1989) found that celebrities create a distinct personality for the endorsed brand. Ultimately the

value of celebrity endorsements should be found in the creation of brand equity by means of the "secondary

association" of a celebrity with a brand (see Keller (1993)). Did Michael Jordan's retirement and subsequent return

to basketball impact the "secondary association" and brand equity created by his endorsement?

We employ event study methodology to examine the impact of Michael Jordan's retirement and return to

basketball on the firm's he endorsed. Event studies have been used widely to examine the impact of strategic

business decisions on the value of a firm's common stock. The use of event studies comes from the efficient

market hypothesis, EMH, (Fama (1970)). According to the EMH in a competitive market the price of a firm's stock

should change immediately to reflect the market's perceived value of new information resulting from unexpected

events. Therefore the change in a stock price around meaningful events could be used to measure the markets

estimate of the value of that event (see Brown and Warner (1985)). Until recently the use of event study

methodology was concentrated in finance and accounting literature. However, recently marketing researchers

have used event study methodology to examine the impact of marketing-related events on firm performance. In

particular Agrawal and Kamakura (1995) find that, on average, the impact of celebrity endorsements on stock

returns is positive and suggest that celebrity endorsement contracts are generally viewed as worthwhile

investments in advertising.

Using event study methodology we examine the impact of Michael Jordan's retirement and subsequent return to

basketball on the stock returns of the companies he endorses. It has been argued that celebrity endorsements

create brand equity through "secondary associations" between the endorser and the product. This ultimately leads

to higher sales and profits for the firm endorsed. Clearly Michael Jordan is more visible and successful as a

basketball player than as a minor league baseball player. We hypothesize that Jordan's retirement from basketball

reduced brand equity through a diminished "secondary association" between Jordan and the products he

endorsed. This decreased brand equity should result in a lower stock price to reflect the loss of endorsement

value. Conversely Jordan's return to basketball should have the opposite impact increasing brand equity and

improving stock price.2 We label this the Michael Jordan effect.

Enlarge this image.

Enlarge this image.

DATA AND METHODOLOGY

We collect daily returns for the five endorsed firms in our sample, (the Jordan Portfolio), and for the S&P500 from

the files of the Center for Research in Security Prices at the University of Chicago (CRSP)34. Data for 1995 was

collected from The Wall Street Journal.

We use a standard event study methodology to measure the excess returns and cumulative excess returns around

Jordan's original retirement and subsequent comeback.5 The excess returns are calculated using an OLS market

model.6

For the original retirement announcement we define day '0' as October 6, 1993, the date of Jordan's retirement

announcement. The event period for this test is t=0 to +1 relative to the event. The market model estimation period

includes 51 trading days starting 65 days prior to the retirement announcement and ending 15 days prior to the

first event date.

For Jordan's return we define day '0' as March 8, 1995, the date the rumors began. The event period runs from day

'0' to day +8, the first trading day after announcing a return to basketball. This event period is longer than typical

for event studies. Unlike Jordan's retirement Jordan's return does not provide a "clean" event day. Instead rumors

persisted for almost two weeks before Jordan announced his comeback. The efficient market hypothesis posits

that stockmarkets react immediately to unexpected new information. In the case of Jordan's comeback new

information was arriving to the market almost daily until Jordan confirmed his comeback.7 Again the estimation

period is defined as the 51 days starting 65 days prior to the event date and ending 15 days prior to the event date.

RESULTS

Table III contains estimates for the abnormal returns and cumulative abnormal returns around Jordan's original

retirement from basketball. If Jordan's return to basketball generated positive excess returns to shareholders, we

anticipate that his retirement would have had the opposite effect. The results show that the shareholders of the

five firm's in the Jordan portfolio experienced a negative abnormal return of -0.45% which was significant at the

10% level. Negative abnormal returns continued into the day after the announcement but these returns are not

significant at conventional levels. In short, we find evidence that Jordan's retirement announcement may have had

a relatively small negative impact on shareholders the day he announced his original retirement from basketball.

Next we examine the impact of Jordan's return to basketball on the same set of five stocks. Table IV summarizes

the results for the Jordan portfolio. Rumor's of Jordan's return to basketball began on March 8th, 1995. On that

day the Jordan portfolio had a 1.29% positive abnormal return which was significant at the 10% level. The positive

abnormal returns continued for two more days with a return of 1.27% on day t=2, (March 10,1995). Interestingly, on

March 15, 1995 the portfolio experienced a -1.20% abnormal return, which is marginally significant. What is

interesting about this negative return is that it occurred at about the same time articles began to appear linking

Jordan rumors to a $2 billion stock price appreciation, (see Berkowitz (1995)). The cumulative abnormal return for

the entire period up to Jordan's confirmed return on March 18th, 1995 was 2.33% but is not significant at

conventional levels. While we find evidence that rumors of Jordan's return to basketball were followed by a

significant run-up in share price of the five companies he endorses, it appears that the value of the rumors was

temporary, as most of the excess returns were reversed by the time of his actual return.

We also examine the excess returns for each individual stock in the Jordan portfolio to estimate the Jordan effect

individually. The results are contained in Table V. None of the five firms in the sample experienced excess returns

over the event period that are significant at conventional levels. However, four of five companies have positive

excess returns over the event period, running from 0.7 percent to over 6.3 percent for General Mills. Interestingly,

the only firm that did not record positive excess returns over this event period was Nike, which has the closest

relationship with Jordan, marketing a line of basketball shoes with his name.

In support of our previous results we plot the excess returns on the Jordan portfolio beginning March 8th, 1995. In

Table IV we document positive excess returns early in the rumor period preceding Jordan's announcement. It also

appears that much of this run-up in share value is reversed mid-way through the rumor period. Figure 1 illustrates

the pattern of these cumulative excess returns. If Jordan's return to basketball had a significant impact on the

share price of the firms he endorses, the excess returns experienced during the build-up to his return

announcement should be permanent. It can be seen in Figure 1 that although the Jordan portfolio experienced

positive excess returns as the rumors of his return surfaced, these excess returns had disappeared by fifteen days

after the rumors began. It does not appear that Jordan's return had a permanent effect on shareholder wealth.

SUMMARY

In the period immediately following the rumors of Michael Jordan's return to basketball, the five companies that

Jordan has major endorsement deals with experienced a nearly $3 billion increase in the market value of their

equity. Jordan was labeled the $2 billion man in the press, referring to the value he created for the shareholders of

the companies he endorses. However, most of these reports failed to cite the simultaneous bull market that lead

the S&P500 to record highs.

We examine the Michael Jordan effect and found shareholders experienced negative abnormal returns after the

announcement of his retirement and positive excess returns when rumors of a comeback surfaced. However, we

also show that the positive excess returns following the rumors of Jordan's comeback were only temporary and

disappeared within weeks of the original rumors. In short, while we find some evidence in support of a Michael

Jordan effect, it appears that the rumors of Jordan's impact have been greatly exaggerated.

ACKNOWLEDGMENT

Enlarge this image.

Enlarge this image.

Enlarge this image.

Enlarge this image.

The authors wish to thank Professor MaryEllen Campbell for her valuable assistance on this paper.

Footnote

NOTES

Footnote

1 The S&P 500 was up 1.645% over this same period.

2 This is especially obvious for Nike. One of Nike's biggest and most profitable products is the Air Jordan

Basketball shoe worn by Jordan.

3 The sample size represents the full population of companies endorsed by Jordan pre and post retirement. The

small sample size will actually bias our statistical tests against finding significant results.

Footnote

4 We use the S&P500 as our market benchmark because all five of the companies endorsed by Jordan are part of

the S&P 500. Using the CRSP value-weighted index does not materially change the results presented in this paper.

5 Methodology is consistent with Brown and Warner (1985), and Peterson (1990).

6 The nature of event study methodology is to measure returns in excess of market trends. We also examine

simple market adjusted excess returns without any significant change in the results. Both methods address the

bull market problem by removing the markets impact from the total returns for the stocks.

7 We did expand our examination of excess returns beyond Jordan's announcement to examine a longer series of

cumulative abnormal returns. The results are presented later in Figure 1. We do not find significant excess returns

after the announcement date.

References

REFERENCES

References

Agrawal J., and W.A. Kamakura, 1995, The Economic Worth of Celebrity Endorser: An Event Study Analysis, The

Journal of Marketing, 15, 56-62.

Berkowitz, Harry, 1995, They Like Mike: Mr. Jordan's Return Pleases Advertisers, Newsday, The Missoulian, March

15, 1995, Page D1.

Brown S., and J. Warner, 1985, Using Daily Stock Returns: The Case of Event Studies, The Journal of Financial

Economics 14, 3-31.

Dorfman, John, 1995, The Jordan Effect, The Wall Street Journal, April 5, 1995, C2.

Fama, Eugene F. (1970), Efficient Capital Markets: A review of Theory and Empirical Work, Journal of Finance, 25

(2), 383-417.

Heisler M., 1995, Air Jordan's Return Flight Home: "I'm Back," The Los Angeles Times, Sunday March 19, 1995., Al.

References

Kamins, Timothy B., M.J. Brand, S.A. Hoeke, and J.C. Moe, 1989, Two-Sided Versus One-Sided Celebrity

Endorsements: The impact of Advertising Effectiveness and Credibility, Journal of Advertising, 18 (2), 4-10.

Keller, Kevin Lane (1993), Conceptualizing, Measuring and Managing Customer-Based Brand Equity, Journal of

Marketing, 57 January), 1-22.

Leland, John, 1995, Hoop Dreams, Newsweek, March 20, 1995, 48(7).

McCracken, Grant (1989), "Who is the Celebrity Endorser? Cultural Foundations of the Endorsement Process,

Journal of Consumer Research, 16 (December), 310-21.

References

Peterson, Pamela, Anatomy of an Event Study, Quarterly Journal of Business and Economics. Vol. 28 (Summer

1990).

Petty, Richard E., J.T. Cacioppo, and D. Schumann, 1983, Central and Peripheral Routes to advertising

Effectiveness: The Moderate Role of Involvement, Journal of Consumer Research, 10 (September), 135-46.

Wulf, Steve, 1995, More Air Goes Out of Baseball, T/me, March 20, 1995., 82.

AuthorAffiliation

Anthony J. Crawford and Bruce Niendorf

AuthorAffiliation

Dr. Anthony J. Crawford is Assistant Professor of Finance, College of Business Administration, University of

Montana, Missoula.

Dr. Bruce Niendorf is Associate Professor of Finance, College of Business Administration, University of Wisconsin,

Oshkosh. DETAILS

Subject: Professional basketball; Athletes; Economic impact; Endorsements; Stock prices;

Statistical analysis; Studies

Location: US

Classification: 9190: US; 1110: Economic conditions &forecasts; 3400: Investment analysis; 8307:

Entertainment industry; 9130: Experimental/theoretical treatment

Publication title: American Business Review; West Haven

Volume: 17

LINKS Check for full text in other resources

Database copyright  2018 ProQuest LLC. All rights reserved. Terms and Conditions Contact ProQuest

Issue: 2

Pages: 5-10

Number of pages: 6

Publication year: 1999

Publication date: Jun 1999

Publisher: University of New Haven

Place of publication: West Haven

Country of publication: United States, West Haven

Publication subject: Business And Economics

ISSN: 07432348

Source type: Scholarly Journals

Language of publication: English

Document type: PERIODICAL

Accession number: 01840982

ProQuest document ID: 216297517

Document URL: https://search.proquest.com/docview/216297517?accountid=8289

Copyright: Copyright University of New Haven Jun 1999

Last updated: 2017-11-10

Database: ProQuest Central