Race and baseball memorabilia revisited

Abstract

This paper extends previous research on customer racial discrimination in the market for baseball memorabilia. Specifically, it investigates if there is evidence that differential prices for baseball cards can be explained at least in part by race. First, it analyzes 2008 prices for cards of recently retired players and compares these results to a previously published study. Then it re-examines the 1994 price differentials found in the previous study by analyzing the 2008 prices for those cards. In particular, changes in the effect of race on card prices motivate a discussion about how attitudes toward race have varied over time.

Keywords:

• Baseball memorabilia
• Customer racial discrimination

Notes

1 See Gabriel, Johnson, and Stanton (1995) for a fuller discussion of this point.

2 For instance, there is on-base percentage, slugging percentage, on-base plus slugging, among others.

3 What performance statistics collectors consider might be debatable, but those of our specification in Eq. (2(2) $\begin{array}{l}\ln P-\ln P_c=\hfill \\ {\beta }_0+B_{1}AB+B_2\text{HITS}+B_3\text{DOUBLES}+B_4\text{TRIPLES}+B_5\text{HOMERUNS}+B_6\text{WALKS}+B_{7}SB+\hfill \\ B_8\text{POSITION}+B_9\text{POST}+B_{10}\text{ACTIVE}+B_{11}\text{NONWHITE}+B_{12}\text{AGE}+B_{13}\text{NWHTAGE}+\epsilon \hfill \end{array}$(2) ) have been widely used in the literature. Further, given that the focus of our article is to compare the earlier GJS results to results with a current data set (see Table 2), we need to use the same independent variables.

4 All data are available upon request.

5 The result of no significance for the NONWHITE variable was robust across different specifications. In particular, this result holds with removal of the nonwhite age interaction, which seems a somewhat problematic variable. Results of our additional specifications are available upon request.

6 The robustness of this result was not as conclusive as was the case for Table 1. For some specifications, NONWHITE was not statistically significant; however, some specification did result in statistical significance at the 5% level for the coefficient. Results are available upon request.

7 While there were a small number of players who were not retired in 1994, the changes in their career performance statistics were negligible. Of course, the ages of the players change over time.