X hits on this document

47 views

0 shares

0 downloads

0 comments

10 / 16

Burdekin et al. / NBA FANS INDIFFERENT TO RACE?

153

Black players. In contrast, during the 1998-1999 season, TWHITE had a positive impact on revenue (the implicit coefficient is 0.584, i.e., –0.416 + 8*0.125). As dis- cussed above, the positive coefficient supports both hypotheses of customer dis- crimination: that fan demand for White players exceeds the supply, and that teams with higher concentrations of White players also have higher caliber White players.

The second specification uses an alternative measure of racial mix: the interac- tion between the percentage of the team that is White and the percentage of the SMSA population that is White (TWHITE*POPWHITE). We also include POPWHITE on its own as well as interacting the matching variable with time. The results are generally in line with those of our first specification: The statistical and economic importance of racial matching becomes more pronounced throughout time and gives rise to similar-sized incremental revenues when evaluated at the White population mean. The driving factor is clearly the interaction with the time trend.13 As shown in Table 4, this compound variable is significant at the 95% level, and the positive sign supports the hypothesis that higher quality White players go to teams with larger White SMSA populations. Furthermore, the 1998-1999 results imply that customers discriminate in conjunction with White-player undersupply—higher quality White players go either to the whiter teams or to cities with higher White populations.

“SORTING” OF WHITE AND BLACK PLAYERS AND PLAYER MOVEMENT

The preceding results certainly suggest that the player selection process has resulted in the better White players (the “stars”) locating in cities with larger White populations. Using a composite of the five performance measures considered ear- lier in the article, we examine the cross-sectional differences in these performance measures. The five performance statistics (assists per minute, blocks per minute, field-goal percentage, points per minute, and rebounds per minute) are first stan- dardized by the mean of the sample for that statistic. We then equally weight and sum the standardized statistics to arrive at the composite performance index for a player.14 Regressing the overall performance measure (PERFORM) on the player’s race (RACE is defined as 1 if the player is White and 0 otherwise), the percentage of the population in the team’s market area that is White (POPWHITE), and the inter- action between RACE and POPWHITE yields the following result for the 1996-1997 through 1998-1999 seasons: 15

PERFORM = 3.526 – 2.622** RACE – 0.730 POPWHITE + 2.743* RACE*POPWHITE

(1.136)

(0.687)

(1.485)

where robust standard errors are in parentheses, and * and ** denote significance at the .10 and .05 levels, respectively.

Document info
Document views47
Page views47
Page last viewedSat Dec 10 05:37:51 UTC 2016
Pages16
Paragraphs357
Words7200

Comments