A REINVESTIGATION OF RACIAL DISCRIMINATION AND BASEBALL CARDS

INTRODUCTION

Since the late 1970 an extensive secondary market has existed for Major League Baseball (MLB) cards, with prices of cards published in monthly and annual magazines and volumes Since performance measures for all players can be rest from various publications, the baseball card market is ideally suited for empirical proofs of racial discrimination on the part of card buyer The cards themselves provide information regarding a player's race - a photograph appears upon the front of each card - and place of birth. Another characteristic of the market is that all cards barter for no less than the "common player price." This bound refers to a minimum price for any card in a series; all cards have a certain number of market value for dealers and collectors regardless of the productivity of individual players. a certain quantity of players' cards have values that exce for the use of all player prices. Most of these are cards of star players (Hank Aaron, Sammy Sosa, CaI Ripken, Jr etc) with above average ability and career performance statistics. Other players' cards command premiums because of the notoriety they achieved as coaches, managers or media personalities after their playing careers extreme pointed (for example, Joe Torre, Walt Wrniak, Charlie Lau and move with a jerk Ueker).

Nardinelli and Simon (1990) examined the incidence of racial discrimination in the market for the Topps 1970 series of baseball cards. From 1956 to 1980 legal restrictions upon the use of player photographs limited competition in this market; Topps was the alone national marketer of baseball cards. Nardinelli and Simon (NS) chose the 1970 series because greatest in quantity of the players in this series had retired and career performance statistics would accurately differentiate players. Their sources for card prices and career performance data were the Official 1989 Price Guide to Baseball Cards [Beckett, 1989] and The Baseball Encyclopedia [Reichler, 1988] respectively. Since card prices have positive minimum values1 regressions using price as a hanging variable would result in a censored variable vexed question Therefore, NS used tobit regressions with the log of the difference between a card's price and its corresponding public player price2 as a pendent variable and various career player performance statistics as explanatory variables. Separate tobit regressions were performed for hitters and pitchers since career performance measures are different for each collection

The article reported evidence of statistically significant differences between the prices of hitters and pitchers through race, holding constant measures of career performance. N reported originates for two tobit regressions for each player group: individual included two dummy explanatory variables for race (one for black players and single for Hispanic players) and a next to the first regression included a dummy variable for nonwhite players (black or Hispanic). These issues utilized data for only single time period: card prices for 1989 and career performance data from the extremity of the 1988 season. Since baseball card prices were published as early as 1979 and annual career performance data can be comput for all players exhibited in the Topps 1970 card series for many years it is possible to use the archetype employed by Nardinelli and Simon to examine the incidence of customer racial discrimination in years prior to and after 1989

This article not aways a reexamination and extension of Nardinelli and Simon's work. The article compares the descriptive statistics and regression originates NS reported using data as of the extremity of the 1988 season (hereafter referr to as post-1988 season data) with rises for the same variables and similar data sources. The fresh results feature explicit procedures for determining race; N did not report their processs for identifying the race of players from the Topps 1970 card series. single minor differences exist between the accrues of both studies.

Tobit regressions (for hitters only) similar to those reported by the agency of NS are presented for the years 1979-19841986-1997 and 1999-2001 To keep on space, only results for the race variables are neared These regressions reveal a race variable that is significantly negative in single certain years, primarily between 1984 and 1991 From 1992 race variables are not statistically significant.

A possible explanation for the disappearance of the race differential is given. This explanation assumes that racial elections exist among card collectors on the other hand not among card dealers. Card dealers are assumed to be motivated by the agency of making profits from buying from and selling to collectors. Increases in discretionary incomes among collectors in the 1980 and 1990 bid up the prices of cards of white players more than the prices of nonwhite players with comparable career performance statistics. predilections among collectors for cards of players from their have cities and regions created additional arbitrage opportunities for dealers and dealers replyed by buying the relatively depressed priced cards of nonwhite players. This activity may explain the disappearance of statistically significant differences between the prices of white and nonwhite players' cards.