- Email: [email protected]
Racial bias in baseball card collecting revisited
Available online at www.sciencedirect.com
The Social Science Journal 47 (2010) 865–874
Racial bias in baseball card collecting revisited Eric Primm a , Nicole L. Piquero b , Robert M. Regoli c , Alex R. Piquero b,∗ a
b
Department of Sociology, Pikeville College, Pikeville, KY, USA College of Criminology and Criminal Justice, Florida State University, Tallahassee, FL, USA c Department of Sociology, University of Colorado, Boulder, CO, USA Received 10 December 2009; received in revised form 17 May 2010; accepted 8 July 2010
Abstract Although research examining the role of racial bias in the secondary sports card market has been an emerging area of inquiry, empirical knowledge on the question: “Does the race of the player on a sports card affect the value of the card?” remains inconclusive. This paper revisits one of the first studies on this topic. Data were derived for 66 Black, White, and Latino members of the National Baseball Hall of Fame who were elected by a vote of the Baseball Writers’ Association of America. Data for each player’s race, career performance statistics, rookie card price, and card availability were obtained from secondary sources. Findings indicate that card availability and, to a lesser extent, player performance is the most important factor affecting the value of a player’s card, while importantly, a player’s race is not a significant contributor to card value. Suggestions for future research are outlined. © 2010 Western Social Science Association. Published by Elsevier Inc. All rights reserved.
In recent decades, sports card collecting has evolved from its traditional form of swapping cards with friends and stashing those “prized acquisitions” in dusty shoeboxes under the bed (or using clothespins to attach them to the spokes of your Schwinn “Sting Ray”) into a multibilliondollar industry (Henricks, 2008). Perhaps coincidentally, around the same time the hobby of card collecting started to change, scholarly inquiry into collecting began to appear (Belk, Wallendorf, Shery, Holbrook, & Roberts, 1988; Danet & Katriel, 1989; Fine, 1987; McIntosh & Schmeichel, 2004; Robinson, 1987; Spooner, 1988). Social scientists recognized that studying the variety of items people collected and their reasons for collecting could yield insight about human behavior. For example, one reason people ∗
Corresponding author at: Florida State University, College of Criminology and Criminal Justice, 634W. Call Street, Hecht House, Tallahassee, FL 32308-1127, United States. E-mail address: [email protected] (A.R. Piquero). 0362-3319/$ – see front matter © 2010 Western Social Science Association. Published by Elsevier Inc. All rights reserved. doi:10.1016/j.soscij.2010.07.005
866
E. Primm et al. / The Social Science Journal 47 (2010) 865–874
collect sports cards and other memorabilia is to reconnect with their childhood memories, or to satisfy the simple urge to surround themselves with organized assortments of precious objects. For others, it is a love of history and the great players of the past that drive their pursuit. And, of course, others collect sports memorabilia because it may be profitable to do so; in fact, sometimes the appreciation for sports memorabilia is greater than that realized in more conventional investments (i.e., stocks or bonds) perhaps because of the scarcity of these items (e.g., how many really nice, strong signature Babe Ruth or Lou Gehrig baseballs are available for sale?). Regardless of their motives, individuals often develop deep personal connections to the objects they collect.
1. Racial bias and sports card values Sports card prices are affected by a variety of factors including the cards’ age, condition, scarcity, and the status of the players pictured on them. Other, not so obvious, factors may also influence the value of sports cards, such as the race of the player who is pictured. The research thus far into “customer racial discrimination” has generated mixed results. Some scholars have found collector bias in the prices of sports cards of Black and Latino baseball players (Andersen & La Croix, 1991; Burnett & VanScyoc, 2004; Fort & Gill, 2000; Gabriel, Johnson, & Stanton, 1999; Nardinelli & Simon, 1990), while other researchers have reported minimal or no racial bias or a mix of results in baseball card values (Gabriel, Johnson, & Stanton, 1995; Hewitt, Mu˜noz, Oliver, & Regoli, 2005; McGarrity, Palmer, & Poitras, 1999; Messitte & Powell, 1995; Mulligan & Grube, 2006; Regoli, 1991; Scahill, 2005). Scholarly work in this area has quickly expanded beyond baseball cards to other sports cards such as football and basketball. So far, this research has found little evidence that race plays a role in the value of sports cards. For example, Stone and Warren (1999) found no bias in their study of basketball trading cards, and Regoli, Primm, and Hewitt (2007a) found no bias in their examination of football cards, a finding replicated and extended in Primm, Piquero, Regoli, and Piquero (2010). Some researchers have utilized a relatively unique dependent variable, card placement, based on the “Topps Numbering System” (O’Connell, 1993), to assess the possibility of racial bias (Primm, Regoli, & Hewitt, 2008; Regoli, Hewitt, Mu˜noz, & Regoli, 2004; Regoli, Primm, & Hewitt, 2007b). In these studies, Regoli et al. (2004) found evidence of racial differences in the placement of players’ cards with Black baseball players disadvantaged in the early years (1956–1966); however in the latter years (1967–1980), this arrangement is reversed and Black players had more favorable card positions in card sets (compared to White players) than their prior year performances would have indicated. Other research using card placement as the dependent variable has reported no evidence of racial differences in the placement of players’ cards (Primm et al., 2008; Regoli et al., 2007b). What explains the contradictory results reported in these different research studies? They are likely the result of different samples and methodologies: Gabriel et al. (1999) examined only rookie cards of baseball players produced by the Topps Company (Topps) between 1974 and 1982; Burnett and VanScyoc (2004) studied the cards of all non-pitchers from 1960 to 1969; Nardinelli and Simon (1990) examined the Topps’ 1970 baseball card set; Fort and Gill (2000) analyzed the 1987 Topps baseball card set; Stone and Warren (1999) studied the
E. Primm et al. / The Social Science Journal 47 (2010) 865–874
867
cards of all players in the National Basketball Association (NBA) who were active players during the 1976–1977 season and who had retired by 1993; Hewitt et al. (2005) examined the rookie cards of players who had been elected to baseball’s Hall of Fame (HOF) by the Baseball Writers Association of America; and Regoli et al. (2007a) studied the rookie cards of a subset of football HOF members. Further, despite the similar focus of these studies, a variety of independent variables have been utilized. For instance, in some studies composite measures of player performance were used, whereas other studies relied on several measures to represent performance (e.g., an individual’s batting and fielding statistics, position played, World Series appearances, and appearances in the annual All-Star game). Similarly, some studies include card availability measures and others do not. Finally, identifying the race of a player is not as self-evident as one might think since an athlete’s race is not specifically identified on the card on which he appears. In an effort to address this limitation, some scholars have constructed a continuous measure of race for each player (Fort & Gill, 2000, p. 25); however, most studies have used a more “traditional” method of measuring race as a discrete variable obtained from photo identification.
2. Current study The point of departure for the current paper is Regoli’s (1991) analysis, which was one of the first scholarly works on the topic, and also one of the few with a parsimonious approach to studying racial bias in the secondary baseball card market. Specifically, Regoli examined the “rookie” card1 of players who had been elected to baseball’s HOF by the Baseball Writers’ Association of America (BBWAA).2 He also limited the cases in his study to include only HOF members who had been elected in 1962 or later, which was the year the first non-White player (Jackie Robinson) became eligible and elected into the HOF. After excluding several other players’ cards to control for the effects of price outliers, Regoli was left with a very small sample of 29 cases for analysis. Results from his study showed no price difference between the cards of White and non-White HOF members. Since 1990, another 31 players have been elected to baseball’s HOF by the BBWAA. This paper extends Regoli’s database to include these 31 players (plus an additional 6 originally excluded for statistical purposes) in an effort to not only replicate his research more rigorously, but also to assess whether there have been any changes in the relationship between race and rookie card values since 1990 among baseball HOFers. Importantly, this study advances the earlier research by Regoli (1991) in three ways: (1) the sample size is more than doubled permitting the utilization of more advanced statistical techniques (also see point 3), (2) card prices are log transformed to correct for a positive skew providing not only for an analysis based on data that are normally distributed, but allowing for the inclusion of the six players that were previously excluded because of the extremely high and low values of their cards, and (3) regression analysis is conducted, which improves on the means-difference and Point–Biserial correlation tests utilized in the prior study since the effects of other potential contributing or intervening factors in the examination of race and card value can be investigated.
868
E. Primm et al. / The Social Science Journal 47 (2010) 865–874
3. Methods Data were derived from five secondary sources: (1) Beckett’s Sports Card website (2009), (2) the Professional Sports Authenticator (PSA) website (2009), (3) Slocum, Foley, and Berger, Topps Baseball Cards: The Complete Picture Collection (1990), (4) Thorn, Birnbaum, and Deane, Total Baseball, 8th edition (2004), and (5) the National Baseball Hall of Fame website (2009). Four variables were used to assess the effect of race on the value of baseball cards. The dependent variable is card value (price), and the three independent variables are players’ race, card availability, and players’ performance. 3.1. Dependent variable Price. Card value is measured in U.S. dollars. Data were obtained from the prices listed on Beckett’s Sports Cards website (January, 2009) Price Guide section, for the rookie cards in near-mint or higher condition for the 66 players in the study. The dependent variable, price, was log transformed (base e) because of its skewed distribution. The residuals of the regression model described below are approximately normally distributed, indicating that the appropriate transformation was applied. 3.2. Independent variables Race. The race of player was determined by a visual inspection of each player’s photograph as it appears on the HOF website and in Topps Baseball Cards: The Complete Picture Collection (Slocum, Foley, & Berger, 1990). The players were divided into non-White (n = 26) and White (n = 40) groups. While there are drawbacks to this method, a point we return to later, it is the most common method used by card collectors to identify a player’s race, and is the approach typically followed in this area of empirical inquiry. As such, this approach also allows us to directly tap into the decisions and consequences of collectors’ perceptions and behavior. Availability. A card’s value is affected by its relative scarcity; however, card companies do not publish data about how many cards they produce each year, just as few comic book publishers reveal how many copies of a particular issue of The Amazing Spider-Man, for example, they printed. Some researchers have addressed this problem by assuming card value is “driven by demand rather than supply” (Burnett & VanScyoc, 2004, p. 104); others, while acknowledging the role of availability in card prices, have included only dummy variables for fewer cards of certain card numbers a company produced, called “short prints,” in a set (Andersen & La Croix, 1991); still others assume uniform availability across individual card sets effectively ignoring the problem altogether (Fort & Gill, 2000; Gabriel et al., 1995, 1999; Mulligan & Grube, 2006; Nardinelli & Simon, 1990; Scahill, 2005; Stone & Warren, 1999). For this study, we approximated the availability of cards through the population reports published by the PSA and Beckett’s websites by constructing a measure based on the number of rookie cards of each player that were reported to exist in near-mint condition or higher, divided by the total number of each of these specimens submitted for grading (Beckett.com, 2009; PSA, 2009). This measure of availability is considerably more appropriate than relying on the raw numbers of each card submitted. Because it costs roughly $10 or so to have a card graded
E. Primm et al. / The Social Science Journal 47 (2010) 865–874
869
by one of these services, collectors will be much less likely to submit lower value “common” cards for grading unless they are very confident the cards will be graded highly (otherwise, the card would cost more to grade than its value). On the other hand, unusually sought-after cards are often submitted regardless of their condition, such as a 1952 Topps card #311 of Mickey Mantle or a 1949 Bowman card #305 of Willie Mays. Performance. Total Player Wins (TPW) “is an estimate of the number of wins, positive or negative, that a player contributed to his team as compared to an average player” (Thorn, Birnbaum, & Deane, 2004, p. 968). This measure was created with the intent of measuring players’ total performance, thus allowing meaningful comparisons to be made between individual players across positions (for a more detailed discussion of TPW and sabermetrics, see Thorn et al., 2004).
4. Results We begin by exploring a simple, baseline model where we assess the relationship between race and rookie card prices. As seen in column 1 of Table 1, race has no effect on rookie card prices (t = .05, p = .96). Next, we estimate a second model (Column Two), where card price is regressed onto the three independent variables: race, availability, and performance.3 The variance explained in this model is .78. As expected, players’ career performance had a significant impact on their card value (t = 2.34, p = .023), indicating that the more a player contributed to his team as compared to an average player, the higher the price of his rookie card. What was unexpected was the relatively small impact of performance when compared to a card’s availability, which had nearly six times the influence in determining the value of cards (t = −13.39, p < .001). As expected, less available cards exhibited higher card prices. These findings are not too surprising, however, since all the players in this study are elected members of Baseball’s HOF. They have already been deemed the “best of the best;” therefore, the potential effects of individual-level performance on card values would be logically mitigated as the variance in levels of performance within this group is much smaller than would be the case if non-HOF members were included in the analysis. Finally, and perhaps most importantly, a player’s race has no significant effect on the value of his rookie card (t = −.05, p = .96). Table 1 Unstandardized regression coefficients and (standard errors) predicting natural log of 2009 card prices. Model 1 (Constant) Racea Availability TPW (performance) R2 ∗ ∗∗ a
p < .05. p < .001. Whites are reference group.
**
5.140 (.252) .018 (.402)
.000
Model 2 7.598** (.327) −.010 (.194) −5.742** (.429) .012* (.005) .779
870
E. Primm et al. / The Social Science Journal 47 (2010) 865–874
To gain a more intuitive grasp of the effects of the coefficients in this model, next we examined a representative or hypothetical card in terms of real dollar amounts rather than logged dollars. Setting performance and availability at their mean levels, 35.3 and .50, respectively, we are able to show the difference between the price of a White and non-White player’s rookie card. The model predicts that a card of this fictional White player would be worth $172.53 while the value of the same card for a non-White player would be worth $170.81 – a small, but insignificant difference. If, however, we vary the availability (or scarcity), which is the variable with the strongest influence in the model, while holding race and performance constant, a different picture emerges. With average performance levels, the model predicts that if the availability of the card is increased by 10 percentage points above that of the mean (to .6) it reduces the card value of a White player to $97.16 (from $172.53) and the card value of a non-White player is reduced to $96.20 (from $170.81). If we lower the availability levels of this same fictional card to .4 (10 percentage points below the mean) the predicted value increases to $306.37 for a White player and to $303.32 for a non-White player. A further 10 percentage point reduction in availability (less than one standard deviation below the mean) increases the card’s predicted value to an impressive $544.03 for a White player and $538.61 for a non-White player.
5. Supplemental analysis: the case of the “big-market” team Before we conclude our investigation, it is worthwhile to explore the possible effect of market size of the team a player was on and if playing for “big-market” teams has an effect on card value. There is very small knowledge base on this issue, largely from the field of economics, which, though inconclusive, suggests that any effects of teams’ market sizes on the value of players’ cards tend to even out in the end (Andersen & La Croix, 1991; Burnett & VanScyoc, 2004; Fort & Gill, 2000; Nardinelli & Simon, 1990). We begin this supplemental analysis by returning to the HOF website, which identifies a player’s “primary” team. For the most part, this primary team corresponds to the one with which they played most of their career. Next, we classify players’ teams into big- and small-market categories. There are a couple of exceptions to this tendency, but in general we were able to successfully categorize players fairly well – recognizing that there may be different points of view on what is a big-market team. In our conception, it deals primarily with the value of a team’s worth in dollars. Using this conceptualization, yearly estimates of a team’s worth were obtained from Forbes Magazine, which provides this information across the four major sports (baseball, basketball, football, and hockey), international soccer, and auto racing, for the year 2008. According to Forbes (2008), the top five teams, which we defined as “big-market,” were: (1) New York Yankees ($1.3 billion), (2) New York Mets ($824 million), (3) Boston Red Sox ($816 million), (4) Los Angeles (to which we appended Brooklyn) Dodgers ($694 million), and (5) Chicago Cubs ($642 million) (Note: The average major league professional baseball club was worth $472 million).4 Thus, a dichotomous variable was created to reflect big-market teams (=1). Of the 66 players in the database, 22 were from big-market teams, and 8 of those 22 were non-Whites. When we added this variable to the main regression model, it did not change
E. Primm et al. / The Social Science Journal 47 (2010) 865–874
871
any of the substantive findings regarding the other coefficients in terms of size, direction, or significance, nor did its inclusion exhibit a significant relation to card value (t = .684, p = .497). We also created an interaction of race and big-market teams and re-estimated the model. No significant effects were observed here either, though the coefficient for the interaction term (representing non-White, big-market players) was negative while the other two terms (race and big-market) were positive.5
6. Discussion The past 20 years has witnessed considerable change in the world of sports card collecting. The hobby saw a rapidly expanding speculative bubble in card prices peak in the early 1990s fueled by double-digit investment returns and the belief that the sports memorabilia market was “recession-proof” (Bloom, 1997; Williams, 1995; Zillante, 2003). Just as quickly, also in the early 1990s, that speculative bubble burst and most sports card prices, with the exception of highly sought-after high-grade cards such as those of Hank Aaron, Ernie Banks, Roberto Clemente, Joe DiMaggio, Mickey Mantle, Willie Mays, Stan Musial, and Ted Williams, remained stagnant for about 10 years. Today, the card market is starting to reassert itself with certain cards fetching eye-popping sums at auction: a 1952 Topps Mickey Mantle recently sold for over $98,000, passing one previously sold for $96,000 (Worthpoint, 2008), and the “Gretzky T206” Honus Wagner card sold for $2.8 million in September 2007 (O’Keeffe & Thompson, 2007). What has not changed is the underlying finding that, at least among HOF members, a player’s race does not play a significant role in determining the value of his rookie card. The analysis presented here suggests that other factors, like a player’s performance, and especially market factors, like a card’s availability, are the primary determinants of a card’s value in the secondary market. Does this mean that race plays no role in sports card values? Not necessarily. Sports cards have no value in-and-of themselves, as they are only small slices of cardboard with pictures and words printed on them. What gives a card value is a person’s desire to own that card and that is an entirely subjective and social process (Simmel, 2004). There are certainly highly individualistic factors that impact some collectors’ choices about the cards they collect. For example, one collector may buy the cards of a specific athlete because they attended the same high school or he signed a foul ball the collector’s son or daughter caught at a Major League game. However, it is larger forces related to the characteristics of a card (i.e., availability, condition, and if it is a rookie card) and the player pictured on it (i.e., performance and status) that tend to offer the best explanations for the value average card collectors place on sports cards. As noted earlier, some researchers cite race as a significant factor in determining the value of sports cards. If race does affect the value of sports cards, why does it fail to do so in this sample? One reason may be a function of the players included as well as those who were excluded. All the players included for the current analysis were elected to baseball’s HOF by the BBWAA – an honor reserved for the very best to have ever played the game. It is possible that in the process of building a career and body of work sufficient for HOF induction by the BBWAA, these select few may have transcended the effects of race itself.
872
E. Primm et al. / The Social Science Journal 47 (2010) 865–874
When basketball fans think of Michael Jordon, Earvin “Magic” Johnson, Julius Erving (Dr. J), or Larry Bird, their thoughts are of championships and MVP awards. Football fans tend to remember Walter Payton becoming the NFL’s all time leader rusher for the Chicago Bears, and baseball fans who picture Willie Mays will likely be reminded of his patented basket catches exemplified by the one from Game 1 of the 1954 World Series, which was a long drive off the bat of Vic Wertz during the eighth inning, known simply as “The Catch.” The greatness of these players is unquestioned and their race becomes, at most, an afterthought. In essence, once a player achieves a certain stature, their race no longer matters in the eyes of fans and collectors. Because this study is not the final word on this issue, several directions for future research are identified. First, the lack of a race effect for rookie card values does not negate the possibility that there are race differences across players’ cards in subsequent years. One interesting question is whether the (normal) decrease in the value of a player’s rookie card and their 2nd year card is the same across race or whether the two prices are significantly different. There may be something that collectors value about rookie cards regardless of race that they do not value for the same player’s 2nd year card. Second, in order to explore the possibility that superstars transcend race, one avenue for future research is to include players who are not HOF members as well as those elected to the HOF by the veterans’ committee. Another area that deserves further investigation deals with the “unpacking” of race and ethnicity. Since there were only four Latino players in the study they were combined with the Black players to form one category. There may be differences in the value of cards between Black, White, and Latino players that are being masked by their categorization in this study, and this will surely be the case in the future as Latino baseball players become an increasing and dominant presence in the game. In addition, this is likely to open useful inquiries that explore race and ethnicity based on skin tone (see Hunter’s (2005) discussion of “colorism”). These and other questions regarding the role of race in sports card collecting should offer a rich area for researchers for some time to come.
Notes 1. A player’s “rookie” card is the player’s first appearance on a regular issue card from a nationally distributed card set. 2. A player can be elected into the HOF in one of two ways: from a vote of the BBWAA (Baseball Writers, Association of America) or the veterans’ committee. The veterans’ committee selects among baseball’s “old timers:” former players who have been inactive for more than 20 years and are thus no longer considered for the HOF by the BBWAA. The most prestigious way for a player to be selected for the HOF is from the vote of the BBWAA. 3. To test for the possibility that race affected card prices in more indirect ways, two t-tests were conducted. The first examined the possibility of a difference in card availability between Whites and non-Whites (t = −.10, p = .92) and the second determined if there was a difference in performance levels between the two groups (t = −.99, p = .33). There were no differences between Whites and non-Whites across either variable.
E. Primm et al. / The Social Science Journal 47 (2010) 865–874
873
4. There is a close correspondence between team worth and average ticket prices. Data from Team Marketing Report (2009) indicates that the highest average ticket prices for the 2009 season were: (1) New York Yankees ($72.97), (2) Boston Red Sox ($50.24), (3) Chicago Cubs ($47.75), (4) New York Mets ($36.99), and (5) Chicago White Sox ($32.28). The Los Angeles Dodgers had the eighth highest average ticket price ($29.66). The overall league average was $26.64. 5. Additional models were explored to examine the possibility that the year a player was inducted into the HOF may have an effect on his rookie card value. When induction year was included in the model, which had a significant effect on card prices (p = .036), the overall variance explained increased to .795 (from .779) and with the exception of card availability (which showed a slight decrease), the relative size and strength of the other variables in the model were not significantly altered. Since the effects of card availability were slightly altered, we then examined the relationship between this variable and induction year. The correlation between these variables was r = .86 (p < .001) indicating collinearity concerns. As such, another model was estimated replacing card availability with induction year. The variance explained decreased to .706 with the relative size and strength of the other variables unaltered.
References Andersen, T., & La Croix, S. J. (1991). Customer racial discrimination in Major League Baseball. Economic Inquiry, 29, 665–677. Beckett.com. (2009). Beckett sport card website. Retrieved from http://beta.beckett.com/ Belk, R. W., Wallendorf, M., Shery, J., Holbrook, M., & Roberts, S. (1988). Collectors and collecting. Advances in Consumer Research, 15, 548–553. Bloom, J. (1997). House of cards. Minneapolis: University of Minnesota. Burnett, N. J., & VanScyoc, L. (2004). Discriminating buyers of baseball cards: Does race affect value? Journal of Economics and Economic Education Research, 5, 101–109. Danet, B., & Katriel, T. (1989). No two alike: Play and aesthetics in collecting. Play and Culture, 2, 253–277. Fine, G. A. (1987). Community and boundary: Personal experience stories of mushroom collectors. Journal of Folklore Research, 24, 223–240. Forbes Magazine. (2008). Yankees top MLB valuation list. Reuters.com. Retrieved from http://www.reuters.com/article/email/idUSN1636416020080416 Fort, R., & Gill, A. (2000). Race and ethnicity assessment in baseball card markets. Journal of Sports Economics, 1, 21–38. Gabriel, P. E., Johnson, C. D., & Stanton, T. J. (1995). An examination of customer racial discrimination in the market for baseball memorabilia. The Journal of Business, 68, 215–230. Gabriel, P. E., Johnson, C. D., & Stanton, T. J. (1999). Customer racial discrimination for baseball memorabilia. Applied Economics, 31, 1331–1335. Henricks, M. (2008). Home run investments. Celebrated Living. Retrieved from http://www.celebratedliving. com/tabid/3097/tabidext/3710/default.aspx Hewitt, J. D., Mu˜noz, R., Jr., Oliver, W. L., & Regoli, R. M. (2005). Race, performance, and baseball card values. Journal of Sport & Social Issues, 29, 411–425. Hunter, M. (2005). Race, gender, and the politics of skin tone. New York: Routledge. McGarrity, J., Palmer, H., & Poitras, M. (1999). Consumer racial discrimination: A reassessment of the market for baseball cards. Journal of Labor Research, 20, 247–258.
874
E. Primm et al. / The Social Science Journal 47 (2010) 865–874
McIntosh, W. E., & Schmeichel, B. (2004). Collectors and collecting: A social psychological perspective. Leisure Sciences, 26, 85–97. Messitte, J., & Powell, I. (1995). Customer race discrimination in the market for baseball cards. Paper presented at Department of Economics, Grinnell College. Mulligan, R. F., & Grube, A. J. (2006). Modeling markets for sports memorabilia. Journal of Economics and Economic Education Research, 7, 75–102. Nardinelli, C., & Simon, C. (1990). Customer racial discrimination in the market for memorabilia: The case of baseball. Quarterly Journal of Economics, 105, 575–595. National Baseball Hall of Fame. (2009). Hall of famers. Retrieved from http://web.baseballhalloffame.org/index.jsp O’Connell, T. S. (1993). Topps numbering system a subtle record of player status. Sports Collectors Digest, 160. O’Keeffe, M., & Thompson, T. (2007). The card: Collectors, con men, and the true story of history’s most desired baseball card. New York: HarperCollins Publishers. Primm, E., Piquero, N. L., Regoli, R. M., & Piquero, A. R. (2010). The role of race in football card prices. Social Science Quarterly, 91, 129–142. Primm, E., Regoli, R. M., & Hewitt, J. D. (2008). Where are we and how did we get here? Assessing race, performance, and position in Topps football card placement. Journal of African American Studies, 12, 73–84. Professional Sports Authenticator. (2009). Population report. Retrieved from http://www.psacard.com/ Regoli, R. M. (1991). Racism in baseball card collecting: Fact or fiction? Human Relations, 44, 255–264. Regoli, R. M., Hewitt, J. D., Mu˜noz, R., Jr., & Regoli, A. M. (2004). Location, location, location: The transmission of racist ideology in baseball cards. The Negro Educational Review, 55, 75–90. Regoli, R. M., Primm, E., & Hewitt, J. D. (2007a). Men and boys and the price of their toys: Race and the value of football cards. The Social Science Journal, 44, 563–571. Regoli, R. M., Primm, E., & Hewitt, J. D. (2007b). Where o’ where did my baseball cards go? Race, performance, and placement in the Topps era, 1956–1980. The Social Science Journal, 44, 742–750. Robinson, R. (1987). Why this piece? On the choices of collectors in the fine arts. Paper presented at the Annual meeting of the American Sociological Association, Chicago. Scahill, E. M. (2005). A reinvestigation of racial discrimination and baseball cards. Eastern Economic Journal, 31, 537–550. Simmel, G. (2004). The philosophy of money (2nd enlarged ed.). (T. Bottomore & D. Frisby, Trans.). Oxford: Routledge (Original work published in 1900). Slocum, F., Foley, R., & Berger, S. (1990). Topps baseball cards: The complete picture collection. New York: Warner Books. Spooner, B. (1988). Weavers and dealers: The authenticity of an oriental carpet. In A. Appadurai (Ed.), The social life of things (pp. 195–235). Cambridge: Cambridge University Press. Stone, E. W., & Warren, R. S. (1999). Customer discrimination in professional basketball: Evidence from the trading-card market. Applied Economics, 31, 679–685. Team Marketing Report. (2009). 2009 MLB fan cost index. Retrieved from http://teammarketing.com. ismmedia.com/ISM3/std-content/repos/Top/Fan%20Cost%20Index/MLB/MLB%20FCI%2009.pdf Thorn, J., Birnbaum, P., & Deane, B. (2004). Total baseball (8th ed). Wilmington, DE: SPORT Media Publishing, Inc. Williams, P. (1995). Card sharks. New York: Macmillan Publishing Company. Worthpoint. (2008, September). Collecting high-value baseball cards. Retrieved from http://www. worthpoint.com/blog-entry/collecting-high-valued-baseball-cards Zillante, A. (2003). Historical overview of the post World War II baseball card industry. Unpublished paper: Florida State University.
The Social Science Journal 47 (2010) 865–874
Racial bias in baseball card collecting revisited Eric Primm a , Nicole L. Piquero b , Robert M. Regoli c , Alex R. Piquero b,∗ a
b
Department of Sociology, Pikeville College, Pikeville, KY, USA College of Criminology and Criminal Justice, Florida State University, Tallahassee, FL, USA c Department of Sociology, University of Colorado, Boulder, CO, USA Received 10 December 2009; received in revised form 17 May 2010; accepted 8 July 2010
Abstract Although research examining the role of racial bias in the secondary sports card market has been an emerging area of inquiry, empirical knowledge on the question: “Does the race of the player on a sports card affect the value of the card?” remains inconclusive. This paper revisits one of the first studies on this topic. Data were derived for 66 Black, White, and Latino members of the National Baseball Hall of Fame who were elected by a vote of the Baseball Writers’ Association of America. Data for each player’s race, career performance statistics, rookie card price, and card availability were obtained from secondary sources. Findings indicate that card availability and, to a lesser extent, player performance is the most important factor affecting the value of a player’s card, while importantly, a player’s race is not a significant contributor to card value. Suggestions for future research are outlined. © 2010 Western Social Science Association. Published by Elsevier Inc. All rights reserved.
In recent decades, sports card collecting has evolved from its traditional form of swapping cards with friends and stashing those “prized acquisitions” in dusty shoeboxes under the bed (or using clothespins to attach them to the spokes of your Schwinn “Sting Ray”) into a multibilliondollar industry (Henricks, 2008). Perhaps coincidentally, around the same time the hobby of card collecting started to change, scholarly inquiry into collecting began to appear (Belk, Wallendorf, Shery, Holbrook, & Roberts, 1988; Danet & Katriel, 1989; Fine, 1987; McIntosh & Schmeichel, 2004; Robinson, 1987; Spooner, 1988). Social scientists recognized that studying the variety of items people collected and their reasons for collecting could yield insight about human behavior. For example, one reason people ∗
Corresponding author at: Florida State University, College of Criminology and Criminal Justice, 634W. Call Street, Hecht House, Tallahassee, FL 32308-1127, United States. E-mail address: [email protected] (A.R. Piquero). 0362-3319/$ – see front matter © 2010 Western Social Science Association. Published by Elsevier Inc. All rights reserved. doi:10.1016/j.soscij.2010.07.005
866
E. Primm et al. / The Social Science Journal 47 (2010) 865–874
collect sports cards and other memorabilia is to reconnect with their childhood memories, or to satisfy the simple urge to surround themselves with organized assortments of precious objects. For others, it is a love of history and the great players of the past that drive their pursuit. And, of course, others collect sports memorabilia because it may be profitable to do so; in fact, sometimes the appreciation for sports memorabilia is greater than that realized in more conventional investments (i.e., stocks or bonds) perhaps because of the scarcity of these items (e.g., how many really nice, strong signature Babe Ruth or Lou Gehrig baseballs are available for sale?). Regardless of their motives, individuals often develop deep personal connections to the objects they collect.
1. Racial bias and sports card values Sports card prices are affected by a variety of factors including the cards’ age, condition, scarcity, and the status of the players pictured on them. Other, not so obvious, factors may also influence the value of sports cards, such as the race of the player who is pictured. The research thus far into “customer racial discrimination” has generated mixed results. Some scholars have found collector bias in the prices of sports cards of Black and Latino baseball players (Andersen & La Croix, 1991; Burnett & VanScyoc, 2004; Fort & Gill, 2000; Gabriel, Johnson, & Stanton, 1999; Nardinelli & Simon, 1990), while other researchers have reported minimal or no racial bias or a mix of results in baseball card values (Gabriel, Johnson, & Stanton, 1995; Hewitt, Mu˜noz, Oliver, & Regoli, 2005; McGarrity, Palmer, & Poitras, 1999; Messitte & Powell, 1995; Mulligan & Grube, 2006; Regoli, 1991; Scahill, 2005). Scholarly work in this area has quickly expanded beyond baseball cards to other sports cards such as football and basketball. So far, this research has found little evidence that race plays a role in the value of sports cards. For example, Stone and Warren (1999) found no bias in their study of basketball trading cards, and Regoli, Primm, and Hewitt (2007a) found no bias in their examination of football cards, a finding replicated and extended in Primm, Piquero, Regoli, and Piquero (2010). Some researchers have utilized a relatively unique dependent variable, card placement, based on the “Topps Numbering System” (O’Connell, 1993), to assess the possibility of racial bias (Primm, Regoli, & Hewitt, 2008; Regoli, Hewitt, Mu˜noz, & Regoli, 2004; Regoli, Primm, & Hewitt, 2007b). In these studies, Regoli et al. (2004) found evidence of racial differences in the placement of players’ cards with Black baseball players disadvantaged in the early years (1956–1966); however in the latter years (1967–1980), this arrangement is reversed and Black players had more favorable card positions in card sets (compared to White players) than their prior year performances would have indicated. Other research using card placement as the dependent variable has reported no evidence of racial differences in the placement of players’ cards (Primm et al., 2008; Regoli et al., 2007b). What explains the contradictory results reported in these different research studies? They are likely the result of different samples and methodologies: Gabriel et al. (1999) examined only rookie cards of baseball players produced by the Topps Company (Topps) between 1974 and 1982; Burnett and VanScyoc (2004) studied the cards of all non-pitchers from 1960 to 1969; Nardinelli and Simon (1990) examined the Topps’ 1970 baseball card set; Fort and Gill (2000) analyzed the 1987 Topps baseball card set; Stone and Warren (1999) studied the
E. Primm et al. / The Social Science Journal 47 (2010) 865–874
867
cards of all players in the National Basketball Association (NBA) who were active players during the 1976–1977 season and who had retired by 1993; Hewitt et al. (2005) examined the rookie cards of players who had been elected to baseball’s Hall of Fame (HOF) by the Baseball Writers Association of America; and Regoli et al. (2007a) studied the rookie cards of a subset of football HOF members. Further, despite the similar focus of these studies, a variety of independent variables have been utilized. For instance, in some studies composite measures of player performance were used, whereas other studies relied on several measures to represent performance (e.g., an individual’s batting and fielding statistics, position played, World Series appearances, and appearances in the annual All-Star game). Similarly, some studies include card availability measures and others do not. Finally, identifying the race of a player is not as self-evident as one might think since an athlete’s race is not specifically identified on the card on which he appears. In an effort to address this limitation, some scholars have constructed a continuous measure of race for each player (Fort & Gill, 2000, p. 25); however, most studies have used a more “traditional” method of measuring race as a discrete variable obtained from photo identification.
2. Current study The point of departure for the current paper is Regoli’s (1991) analysis, which was one of the first scholarly works on the topic, and also one of the few with a parsimonious approach to studying racial bias in the secondary baseball card market. Specifically, Regoli examined the “rookie” card1 of players who had been elected to baseball’s HOF by the Baseball Writers’ Association of America (BBWAA).2 He also limited the cases in his study to include only HOF members who had been elected in 1962 or later, which was the year the first non-White player (Jackie Robinson) became eligible and elected into the HOF. After excluding several other players’ cards to control for the effects of price outliers, Regoli was left with a very small sample of 29 cases for analysis. Results from his study showed no price difference between the cards of White and non-White HOF members. Since 1990, another 31 players have been elected to baseball’s HOF by the BBWAA. This paper extends Regoli’s database to include these 31 players (plus an additional 6 originally excluded for statistical purposes) in an effort to not only replicate his research more rigorously, but also to assess whether there have been any changes in the relationship between race and rookie card values since 1990 among baseball HOFers. Importantly, this study advances the earlier research by Regoli (1991) in three ways: (1) the sample size is more than doubled permitting the utilization of more advanced statistical techniques (also see point 3), (2) card prices are log transformed to correct for a positive skew providing not only for an analysis based on data that are normally distributed, but allowing for the inclusion of the six players that were previously excluded because of the extremely high and low values of their cards, and (3) regression analysis is conducted, which improves on the means-difference and Point–Biserial correlation tests utilized in the prior study since the effects of other potential contributing or intervening factors in the examination of race and card value can be investigated.
868
E. Primm et al. / The Social Science Journal 47 (2010) 865–874
3. Methods Data were derived from five secondary sources: (1) Beckett’s Sports Card website (2009), (2) the Professional Sports Authenticator (PSA) website (2009), (3) Slocum, Foley, and Berger, Topps Baseball Cards: The Complete Picture Collection (1990), (4) Thorn, Birnbaum, and Deane, Total Baseball, 8th edition (2004), and (5) the National Baseball Hall of Fame website (2009). Four variables were used to assess the effect of race on the value of baseball cards. The dependent variable is card value (price), and the three independent variables are players’ race, card availability, and players’ performance. 3.1. Dependent variable Price. Card value is measured in U.S. dollars. Data were obtained from the prices listed on Beckett’s Sports Cards website (January, 2009) Price Guide section, for the rookie cards in near-mint or higher condition for the 66 players in the study. The dependent variable, price, was log transformed (base e) because of its skewed distribution. The residuals of the regression model described below are approximately normally distributed, indicating that the appropriate transformation was applied. 3.2. Independent variables Race. The race of player was determined by a visual inspection of each player’s photograph as it appears on the HOF website and in Topps Baseball Cards: The Complete Picture Collection (Slocum, Foley, & Berger, 1990). The players were divided into non-White (n = 26) and White (n = 40) groups. While there are drawbacks to this method, a point we return to later, it is the most common method used by card collectors to identify a player’s race, and is the approach typically followed in this area of empirical inquiry. As such, this approach also allows us to directly tap into the decisions and consequences of collectors’ perceptions and behavior. Availability. A card’s value is affected by its relative scarcity; however, card companies do not publish data about how many cards they produce each year, just as few comic book publishers reveal how many copies of a particular issue of The Amazing Spider-Man, for example, they printed. Some researchers have addressed this problem by assuming card value is “driven by demand rather than supply” (Burnett & VanScyoc, 2004, p. 104); others, while acknowledging the role of availability in card prices, have included only dummy variables for fewer cards of certain card numbers a company produced, called “short prints,” in a set (Andersen & La Croix, 1991); still others assume uniform availability across individual card sets effectively ignoring the problem altogether (Fort & Gill, 2000; Gabriel et al., 1995, 1999; Mulligan & Grube, 2006; Nardinelli & Simon, 1990; Scahill, 2005; Stone & Warren, 1999). For this study, we approximated the availability of cards through the population reports published by the PSA and Beckett’s websites by constructing a measure based on the number of rookie cards of each player that were reported to exist in near-mint condition or higher, divided by the total number of each of these specimens submitted for grading (Beckett.com, 2009; PSA, 2009). This measure of availability is considerably more appropriate than relying on the raw numbers of each card submitted. Because it costs roughly $10 or so to have a card graded
E. Primm et al. / The Social Science Journal 47 (2010) 865–874
869
by one of these services, collectors will be much less likely to submit lower value “common” cards for grading unless they are very confident the cards will be graded highly (otherwise, the card would cost more to grade than its value). On the other hand, unusually sought-after cards are often submitted regardless of their condition, such as a 1952 Topps card #311 of Mickey Mantle or a 1949 Bowman card #305 of Willie Mays. Performance. Total Player Wins (TPW) “is an estimate of the number of wins, positive or negative, that a player contributed to his team as compared to an average player” (Thorn, Birnbaum, & Deane, 2004, p. 968). This measure was created with the intent of measuring players’ total performance, thus allowing meaningful comparisons to be made between individual players across positions (for a more detailed discussion of TPW and sabermetrics, see Thorn et al., 2004).
4. Results We begin by exploring a simple, baseline model where we assess the relationship between race and rookie card prices. As seen in column 1 of Table 1, race has no effect on rookie card prices (t = .05, p = .96). Next, we estimate a second model (Column Two), where card price is regressed onto the three independent variables: race, availability, and performance.3 The variance explained in this model is .78. As expected, players’ career performance had a significant impact on their card value (t = 2.34, p = .023), indicating that the more a player contributed to his team as compared to an average player, the higher the price of his rookie card. What was unexpected was the relatively small impact of performance when compared to a card’s availability, which had nearly six times the influence in determining the value of cards (t = −13.39, p < .001). As expected, less available cards exhibited higher card prices. These findings are not too surprising, however, since all the players in this study are elected members of Baseball’s HOF. They have already been deemed the “best of the best;” therefore, the potential effects of individual-level performance on card values would be logically mitigated as the variance in levels of performance within this group is much smaller than would be the case if non-HOF members were included in the analysis. Finally, and perhaps most importantly, a player’s race has no significant effect on the value of his rookie card (t = −.05, p = .96). Table 1 Unstandardized regression coefficients and (standard errors) predicting natural log of 2009 card prices. Model 1 (Constant) Racea Availability TPW (performance) R2 ∗ ∗∗ a
p < .05. p < .001. Whites are reference group.
**
5.140 (.252) .018 (.402)
.000
Model 2 7.598** (.327) −.010 (.194) −5.742** (.429) .012* (.005) .779
870
E. Primm et al. / The Social Science Journal 47 (2010) 865–874
To gain a more intuitive grasp of the effects of the coefficients in this model, next we examined a representative or hypothetical card in terms of real dollar amounts rather than logged dollars. Setting performance and availability at their mean levels, 35.3 and .50, respectively, we are able to show the difference between the price of a White and non-White player’s rookie card. The model predicts that a card of this fictional White player would be worth $172.53 while the value of the same card for a non-White player would be worth $170.81 – a small, but insignificant difference. If, however, we vary the availability (or scarcity), which is the variable with the strongest influence in the model, while holding race and performance constant, a different picture emerges. With average performance levels, the model predicts that if the availability of the card is increased by 10 percentage points above that of the mean (to .6) it reduces the card value of a White player to $97.16 (from $172.53) and the card value of a non-White player is reduced to $96.20 (from $170.81). If we lower the availability levels of this same fictional card to .4 (10 percentage points below the mean) the predicted value increases to $306.37 for a White player and to $303.32 for a non-White player. A further 10 percentage point reduction in availability (less than one standard deviation below the mean) increases the card’s predicted value to an impressive $544.03 for a White player and $538.61 for a non-White player.
5. Supplemental analysis: the case of the “big-market” team Before we conclude our investigation, it is worthwhile to explore the possible effect of market size of the team a player was on and if playing for “big-market” teams has an effect on card value. There is very small knowledge base on this issue, largely from the field of economics, which, though inconclusive, suggests that any effects of teams’ market sizes on the value of players’ cards tend to even out in the end (Andersen & La Croix, 1991; Burnett & VanScyoc, 2004; Fort & Gill, 2000; Nardinelli & Simon, 1990). We begin this supplemental analysis by returning to the HOF website, which identifies a player’s “primary” team. For the most part, this primary team corresponds to the one with which they played most of their career. Next, we classify players’ teams into big- and small-market categories. There are a couple of exceptions to this tendency, but in general we were able to successfully categorize players fairly well – recognizing that there may be different points of view on what is a big-market team. In our conception, it deals primarily with the value of a team’s worth in dollars. Using this conceptualization, yearly estimates of a team’s worth were obtained from Forbes Magazine, which provides this information across the four major sports (baseball, basketball, football, and hockey), international soccer, and auto racing, for the year 2008. According to Forbes (2008), the top five teams, which we defined as “big-market,” were: (1) New York Yankees ($1.3 billion), (2) New York Mets ($824 million), (3) Boston Red Sox ($816 million), (4) Los Angeles (to which we appended Brooklyn) Dodgers ($694 million), and (5) Chicago Cubs ($642 million) (Note: The average major league professional baseball club was worth $472 million).4 Thus, a dichotomous variable was created to reflect big-market teams (=1). Of the 66 players in the database, 22 were from big-market teams, and 8 of those 22 were non-Whites. When we added this variable to the main regression model, it did not change
E. Primm et al. / The Social Science Journal 47 (2010) 865–874
871
any of the substantive findings regarding the other coefficients in terms of size, direction, or significance, nor did its inclusion exhibit a significant relation to card value (t = .684, p = .497). We also created an interaction of race and big-market teams and re-estimated the model. No significant effects were observed here either, though the coefficient for the interaction term (representing non-White, big-market players) was negative while the other two terms (race and big-market) were positive.5
6. Discussion The past 20 years has witnessed considerable change in the world of sports card collecting. The hobby saw a rapidly expanding speculative bubble in card prices peak in the early 1990s fueled by double-digit investment returns and the belief that the sports memorabilia market was “recession-proof” (Bloom, 1997; Williams, 1995; Zillante, 2003). Just as quickly, also in the early 1990s, that speculative bubble burst and most sports card prices, with the exception of highly sought-after high-grade cards such as those of Hank Aaron, Ernie Banks, Roberto Clemente, Joe DiMaggio, Mickey Mantle, Willie Mays, Stan Musial, and Ted Williams, remained stagnant for about 10 years. Today, the card market is starting to reassert itself with certain cards fetching eye-popping sums at auction: a 1952 Topps Mickey Mantle recently sold for over $98,000, passing one previously sold for $96,000 (Worthpoint, 2008), and the “Gretzky T206” Honus Wagner card sold for $2.8 million in September 2007 (O’Keeffe & Thompson, 2007). What has not changed is the underlying finding that, at least among HOF members, a player’s race does not play a significant role in determining the value of his rookie card. The analysis presented here suggests that other factors, like a player’s performance, and especially market factors, like a card’s availability, are the primary determinants of a card’s value in the secondary market. Does this mean that race plays no role in sports card values? Not necessarily. Sports cards have no value in-and-of themselves, as they are only small slices of cardboard with pictures and words printed on them. What gives a card value is a person’s desire to own that card and that is an entirely subjective and social process (Simmel, 2004). There are certainly highly individualistic factors that impact some collectors’ choices about the cards they collect. For example, one collector may buy the cards of a specific athlete because they attended the same high school or he signed a foul ball the collector’s son or daughter caught at a Major League game. However, it is larger forces related to the characteristics of a card (i.e., availability, condition, and if it is a rookie card) and the player pictured on it (i.e., performance and status) that tend to offer the best explanations for the value average card collectors place on sports cards. As noted earlier, some researchers cite race as a significant factor in determining the value of sports cards. If race does affect the value of sports cards, why does it fail to do so in this sample? One reason may be a function of the players included as well as those who were excluded. All the players included for the current analysis were elected to baseball’s HOF by the BBWAA – an honor reserved for the very best to have ever played the game. It is possible that in the process of building a career and body of work sufficient for HOF induction by the BBWAA, these select few may have transcended the effects of race itself.
872
E. Primm et al. / The Social Science Journal 47 (2010) 865–874
When basketball fans think of Michael Jordon, Earvin “Magic” Johnson, Julius Erving (Dr. J), or Larry Bird, their thoughts are of championships and MVP awards. Football fans tend to remember Walter Payton becoming the NFL’s all time leader rusher for the Chicago Bears, and baseball fans who picture Willie Mays will likely be reminded of his patented basket catches exemplified by the one from Game 1 of the 1954 World Series, which was a long drive off the bat of Vic Wertz during the eighth inning, known simply as “The Catch.” The greatness of these players is unquestioned and their race becomes, at most, an afterthought. In essence, once a player achieves a certain stature, their race no longer matters in the eyes of fans and collectors. Because this study is not the final word on this issue, several directions for future research are identified. First, the lack of a race effect for rookie card values does not negate the possibility that there are race differences across players’ cards in subsequent years. One interesting question is whether the (normal) decrease in the value of a player’s rookie card and their 2nd year card is the same across race or whether the two prices are significantly different. There may be something that collectors value about rookie cards regardless of race that they do not value for the same player’s 2nd year card. Second, in order to explore the possibility that superstars transcend race, one avenue for future research is to include players who are not HOF members as well as those elected to the HOF by the veterans’ committee. Another area that deserves further investigation deals with the “unpacking” of race and ethnicity. Since there were only four Latino players in the study they were combined with the Black players to form one category. There may be differences in the value of cards between Black, White, and Latino players that are being masked by their categorization in this study, and this will surely be the case in the future as Latino baseball players become an increasing and dominant presence in the game. In addition, this is likely to open useful inquiries that explore race and ethnicity based on skin tone (see Hunter’s (2005) discussion of “colorism”). These and other questions regarding the role of race in sports card collecting should offer a rich area for researchers for some time to come.
Notes 1. A player’s “rookie” card is the player’s first appearance on a regular issue card from a nationally distributed card set. 2. A player can be elected into the HOF in one of two ways: from a vote of the BBWAA (Baseball Writers, Association of America) or the veterans’ committee. The veterans’ committee selects among baseball’s “old timers:” former players who have been inactive for more than 20 years and are thus no longer considered for the HOF by the BBWAA. The most prestigious way for a player to be selected for the HOF is from the vote of the BBWAA. 3. To test for the possibility that race affected card prices in more indirect ways, two t-tests were conducted. The first examined the possibility of a difference in card availability between Whites and non-Whites (t = −.10, p = .92) and the second determined if there was a difference in performance levels between the two groups (t = −.99, p = .33). There were no differences between Whites and non-Whites across either variable.
E. Primm et al. / The Social Science Journal 47 (2010) 865–874
873
4. There is a close correspondence between team worth and average ticket prices. Data from Team Marketing Report (2009) indicates that the highest average ticket prices for the 2009 season were: (1) New York Yankees ($72.97), (2) Boston Red Sox ($50.24), (3) Chicago Cubs ($47.75), (4) New York Mets ($36.99), and (5) Chicago White Sox ($32.28). The Los Angeles Dodgers had the eighth highest average ticket price ($29.66). The overall league average was $26.64. 5. Additional models were explored to examine the possibility that the year a player was inducted into the HOF may have an effect on his rookie card value. When induction year was included in the model, which had a significant effect on card prices (p = .036), the overall variance explained increased to .795 (from .779) and with the exception of card availability (which showed a slight decrease), the relative size and strength of the other variables in the model were not significantly altered. Since the effects of card availability were slightly altered, we then examined the relationship between this variable and induction year. The correlation between these variables was r = .86 (p < .001) indicating collinearity concerns. As such, another model was estimated replacing card availability with induction year. The variance explained decreased to .706 with the relative size and strength of the other variables unaltered.
References Andersen, T., & La Croix, S. J. (1991). Customer racial discrimination in Major League Baseball. Economic Inquiry, 29, 665–677. Beckett.com. (2009). Beckett sport card website. Retrieved from http://beta.beckett.com/ Belk, R. W., Wallendorf, M., Shery, J., Holbrook, M., & Roberts, S. (1988). Collectors and collecting. Advances in Consumer Research, 15, 548–553. Bloom, J. (1997). House of cards. Minneapolis: University of Minnesota. Burnett, N. J., & VanScyoc, L. (2004). Discriminating buyers of baseball cards: Does race affect value? Journal of Economics and Economic Education Research, 5, 101–109. Danet, B., & Katriel, T. (1989). No two alike: Play and aesthetics in collecting. Play and Culture, 2, 253–277. Fine, G. A. (1987). Community and boundary: Personal experience stories of mushroom collectors. Journal of Folklore Research, 24, 223–240. Forbes Magazine. (2008). Yankees top MLB valuation list. Reuters.com. Retrieved from http://www.reuters.com/article/email/idUSN1636416020080416 Fort, R., & Gill, A. (2000). Race and ethnicity assessment in baseball card markets. Journal of Sports Economics, 1, 21–38. Gabriel, P. E., Johnson, C. D., & Stanton, T. J. (1995). An examination of customer racial discrimination in the market for baseball memorabilia. The Journal of Business, 68, 215–230. Gabriel, P. E., Johnson, C. D., & Stanton, T. J. (1999). Customer racial discrimination for baseball memorabilia. Applied Economics, 31, 1331–1335. Henricks, M. (2008). Home run investments. Celebrated Living. Retrieved from http://www.celebratedliving. com/tabid/3097/tabidext/3710/default.aspx Hewitt, J. D., Mu˜noz, R., Jr., Oliver, W. L., & Regoli, R. M. (2005). Race, performance, and baseball card values. Journal of Sport & Social Issues, 29, 411–425. Hunter, M. (2005). Race, gender, and the politics of skin tone. New York: Routledge. McGarrity, J., Palmer, H., & Poitras, M. (1999). Consumer racial discrimination: A reassessment of the market for baseball cards. Journal of Labor Research, 20, 247–258.
874
E. Primm et al. / The Social Science Journal 47 (2010) 865–874
McIntosh, W. E., & Schmeichel, B. (2004). Collectors and collecting: A social psychological perspective. Leisure Sciences, 26, 85–97. Messitte, J., & Powell, I. (1995). Customer race discrimination in the market for baseball cards. Paper presented at Department of Economics, Grinnell College. Mulligan, R. F., & Grube, A. J. (2006). Modeling markets for sports memorabilia. Journal of Economics and Economic Education Research, 7, 75–102. Nardinelli, C., & Simon, C. (1990). Customer racial discrimination in the market for memorabilia: The case of baseball. Quarterly Journal of Economics, 105, 575–595. National Baseball Hall of Fame. (2009). Hall of famers. Retrieved from http://web.baseballhalloffame.org/index.jsp O’Connell, T. S. (1993). Topps numbering system a subtle record of player status. Sports Collectors Digest, 160. O’Keeffe, M., & Thompson, T. (2007). The card: Collectors, con men, and the true story of history’s most desired baseball card. New York: HarperCollins Publishers. Primm, E., Piquero, N. L., Regoli, R. M., & Piquero, A. R. (2010). The role of race in football card prices. Social Science Quarterly, 91, 129–142. Primm, E., Regoli, R. M., & Hewitt, J. D. (2008). Where are we and how did we get here? Assessing race, performance, and position in Topps football card placement. Journal of African American Studies, 12, 73–84. Professional Sports Authenticator. (2009). Population report. Retrieved from http://www.psacard.com/ Regoli, R. M. (1991). Racism in baseball card collecting: Fact or fiction? Human Relations, 44, 255–264. Regoli, R. M., Hewitt, J. D., Mu˜noz, R., Jr., & Regoli, A. M. (2004). Location, location, location: The transmission of racist ideology in baseball cards. The Negro Educational Review, 55, 75–90. Regoli, R. M., Primm, E., & Hewitt, J. D. (2007a). Men and boys and the price of their toys: Race and the value of football cards. The Social Science Journal, 44, 563–571. Regoli, R. M., Primm, E., & Hewitt, J. D. (2007b). Where o’ where did my baseball cards go? Race, performance, and placement in the Topps era, 1956–1980. The Social Science Journal, 44, 742–750. Robinson, R. (1987). Why this piece? On the choices of collectors in the fine arts. Paper presented at the Annual meeting of the American Sociological Association, Chicago. Scahill, E. M. (2005). A reinvestigation of racial discrimination and baseball cards. Eastern Economic Journal, 31, 537–550. Simmel, G. (2004). The philosophy of money (2nd enlarged ed.). (T. Bottomore & D. Frisby, Trans.). Oxford: Routledge (Original work published in 1900). Slocum, F., Foley, R., & Berger, S. (1990). Topps baseball cards: The complete picture collection. New York: Warner Books. Spooner, B. (1988). Weavers and dealers: The authenticity of an oriental carpet. In A. Appadurai (Ed.), The social life of things (pp. 195–235). Cambridge: Cambridge University Press. Stone, E. W., & Warren, R. S. (1999). Customer discrimination in professional basketball: Evidence from the trading-card market. Applied Economics, 31, 679–685. Team Marketing Report. (2009). 2009 MLB fan cost index. Retrieved from http://teammarketing.com. ismmedia.com/ISM3/std-content/repos/Top/Fan%20Cost%20Index/MLB/MLB%20FCI%2009.pdf Thorn, J., Birnbaum, P., & Deane, B. (2004). Total baseball (8th ed). Wilmington, DE: SPORT Media Publishing, Inc. Williams, P. (1995). Card sharks. New York: Macmillan Publishing Company. Worthpoint. (2008, September). Collecting high-value baseball cards. Retrieved from http://www. worthpoint.com/blog-entry/collecting-high-valued-baseball-cards Zillante, A. (2003). Historical overview of the post World War II baseball card industry. Unpublished paper: Florida State University.