The impact of title event sponsorship announcements on shareholder wealth

Abstract

Title sponsorships are often considered the crown jewels of sports sponsorship programs. Garnering top media coverage, title sponsorships are prized for both generating brand/product awareness and building image for their sponsors. Not surprisingly, the rising cost of title sponsorships has led some managers to question their underlying value. Accordingly, this study presents an analysis of the impact of 114 title sponsorship announcements of professional tennis and golf tournaments (both men’s and women’s), auto racing (NASCAR), and college bowl games on the stock prices of sponsoring firms. Overall, the results of the study suggest that title sponsorships are generally signed at market-clearing prices. Thus, companies undertaking title sponsorships typically receive exactly what they pay for—except in the case of NASCAR races (which show evidence of increases in share prices). Splitting the sample into new and renewing sponsorships generates results which differ dramatically by sport. Finally, a cross-sectional regression finds congruence of sport and sponsor, sponsorship by high tech firms and sponsorships by large firms all correlated with perceived sponsorship success.

Appendix

Known as “event analysis,” the technique used here estimates a time-series of expected stock market returns which are then subtracted from actual security returns over the same period of time to arrive at an estimate of the unexpected or “abnormal” returns associated with a particular event. Slope and intercept coefficients for the market model methodology were estimated using a procedure suggested by Scholes and Williams (1977). Specifically, the Scholes–Williams slope coefficient for stock j was estimated over the period 146 to 21 trading days (event days t = −146 through t = −21) prior to the date of each event sponsorship announcement (event day t = 0) and is defined as:

$$\widehat\beta _j^ * = \frac{{\widehat\beta _j^ - + \widehat\beta _j + \widehat\beta _j^ + }}{{1 + 2\widehat\rho _m }}$$

where $\widehat{\rho }_{m}$ is the estimated first-order autocorrelation coefficient of the market index over the period t = −146 to −21, and the beta terms were the ordinary least squares coefficients estimated from leading, lagging, and coincident regressions between each firm and the CRSP value-weighted index of all stocks included in the CRSP database. As noted above, the Scholes–Williams technique controls for biases associated with less-actively traded stocks.

Abnormal returns associated with sponsorship events were defined as:

$$A_{jt} = R_{jt} - \left( {\widehat\alpha _j + \widehat\beta _j R_{mt} } \right)$$

where R _jt is the actual return on stock j on event day t and the alpha and beta coefficients result from the Scholes–Williams procedure discussed above.

The average abnormal return (AAR _t ) event day t for a sample of event sponsorships is defined as:

$${\text{AAR}}_t = \frac{{\sum\limits_{j = 1}^N {A_{jt} } }}{N}$$

where N is the total number of firms in the respective sponsorship sample.

Finally, the cumulative average abnormal return (CAAR _T1,T2 ) over a specific event window from event days t = T ₁ to T ₂ for a sample of N firms is calculated as:

$${\text{CAAR}}_{T_1 ,T_2 } = \frac{1}{N}\sum\limits_{j = 1}^N {\sum\limits_{t = T_1 }^{T_2 } {A_{jt} } } $$

The actual statistical procedures involved in the evaluation of the average abnormal return (AAR _t ) and cumulative average abnormal return (CAAR _T1,T2 ) levels are described in detail in the Eventus User’s Manual available without charge online at www.eventstudy.com and are not reproduced here due to space limitations.