M««‘~~ ma. "1“ AN‘ ANALYSIS OF THE FACTORS CONTRIBUTING TO THE SUCCESSFUL A LOCATION OF MAIORLEACUE ” BASEBALLFRANCHISES- _ Dissertation for 'the Degree Of-Ph. D- I MICHIGAN STATE'UNIVERSITY * * CHARLES LOUISALOINI - ‘ ’ ‘ 1977 ' .v 7-;- _ ..,. «AV! ... W .... 1.“, H u. We H- .....~ m. .. a. ~-I\ V‘ , Cw . I... «9i. ., ... -; . TO. "we? .5”. w. . :1, - aw?! I «a... “ '31...’ ‘ 3 155; ., r . «'13 «v. 3;;1 . III III III III IIIIII II I III 08080 80446 « LIBRARY Michigan State University This is to certify that the thesis entitled AN ANALYSIS OF THE FACTORS CONTRIBUTING TO THE SUCCESSFUL LOCATION OF MAJOR LEAGUE BASEBALL presented by CHARLES L . ALDINI has been accepted towards fulfillment of the requirements for Doctor of Philosophy degree in Resource Development imw III Major professor Date May 19, I977 0-7639 ABSTRACT The primary objective of this study is to delineate those var- iables which influence the location of professional baseball franchises and to match those variables with elements of particular clubs to be located. A further objective is to rank and measure the relationship of each variable to the presence or absence of major league teams in a particular area. Another objective is to determine prime potential Standard MetrOpolitan Statistical Areas (SMSA's) for future league ex- pansion or franchise relocation. This study concentrates on paid attendance. Since profit maxi- mization is the objective function of most major league baseball fran- chise owners, this study focuses upon the only available and repre- sentative component of the economic infrastructure of these franchises-- paid attendance. A model was built to predict this paid attendance. The model identified five variables which significantly contributed to an explanation of paid attendance. The resultant regression equa- tion accounted for approximately 74 percent of the variation in paid attendance, and a consequently much more unified and systematic over- view of the role of paid attendance within the infrastructure of major league baseball franchises. The final task of this study was to construct attendance projec— tions for non-franchised SMSA's within the parameters of the study. Upon completion of this task, l0 SMSA's emerged as potentially strong major league baseball franchise locations. From this total only three SMSA's were deemed ideal future locations in that only they suggested the potential to compete successfully in the market place. These three SMSA's were Tampa-St. Petersburg, Newark and Miami. A basic outgrowth of the model is that the general market place for major league baseball franchises in the United States is extremely limited. Thus, given such a saturated market, there is only one way for future franchise location to turn in our mass society. It must develop critical mass by expanding major league baseball to inter- national markets. The recent awarding of a major league baseball franchise to Toronto, Ontario, Canada was a reaffirmation of this fact. This was a conscious and much considered decision based upon the initial location of the Montreal Expos in Canada in l969, and should be seen as a harbinger of future international expansion to such areas as Mexico, Central America and Japan. AN ANALYSIS OF THE FACTORS CONTRIBUTING TO THE SUCCESSFUL LOCATION OF MAJOR LEAGUE BASEBALL FRANCHISES by Charles Louis Aldini A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Resource Development 1977 ACKNOWLEDGMENTS Projects such as this are seldom completed without a tremendous amount of cooperation and support from others. This was no exception. I would like to thank my dissertation committee for their guidance and consideration; my family for their confidence and concern; my friends for their gentle prodding and selfless assistance. This project is as much theirs as mine, for it would never have reached completion without their cooperation and support. ii TABLE OF CONTENTS LIST OF TABLES ......................... v LIST OF FIGURES ........................ vi CHAPTER 'I. INSTITUTIONAL AND HISTORICAL BACKGROUND .......... 1 Major League Framework and Structure .......... 1 Operation of Team Sports ................ 2 Franchise Failures: Limits to Growth .......... 5 Football ....................... 6 Basketball ...................... 6 Hockey ........................ 6 Baseball ....................... 7 Franchise Stability ................... 7 Baseball ....................... 7 Football ....................... 9 Basketball ...................... 9 Hockey ........................ l3 Geographical Distribution of Major League Franchises . . l6 Problem Statement .................... l8 II. ELEMENTS OF FRANCHISE SUCCESS ............... 20 Sources of Revenue ................... 2l Team Quality and Profits ................ 22 Broadcast Rights .................... 23 Drawing Potential and Team Quality ........... 25 Elements of the Demand for Professional Baseball . . . . 27 Defining the Variables ................ 27 Demographic Factors ................. 27 Economic Factors ................... 28 Employment Characteristics .............. 30 Entertainment Competition .............. 3l Factors Inherent in Major League Baseball ...... 3l Hypothesis ....................... 32 III. THE MODEL: METHODOLOGICAL AND CONCEPTUAL CONSIDERATIONS . 33 Method ......................... 35 Factor Analysis ..................... 35 Correlation Regression Analysis ............. 38 Interpretation of the Factor Analysis .......... 41 Summary ......................... 47 IV. THE MODEL: ANALYSIS AND RESULTS ............. 48 Interpretation of Predicting Variables ......... 52 Beta Weights ...................... 54 Residuals From Regression ................ 55 V. THE MODEL: IMPLICATIONS AND FUTURE DIRECTIONS ...... 57 Viability of Non-Franchised SMSA's ........... 58 The Seattle Situation .................. 6l Gate Sharing Arrangements ................ 62 SMSA's Motives for Franchise Location .......... 62 Areas for Future Research ................ 64 General Summary ..................... 65 General Conclusions ................... 66 APPENDICES ........................... 69 BIBLIOGRAPHY .......................... 76 iv Table —J crooowosu-I-wa—I LIST OF TABLES Major League Franchise Expansion l950-l975 ........ 4 Geographical Districution of Major League Franchises . . . l7 Varimax Factor Loading Matrix ............... 4l Percent Variance Per Factor ................ 42 Economic Strength and Population ............. 44 Simple Coefficients of Correlation ............ 50 Change in Coefficient of Multiple Determination ...... 52 Beta Weights ....................... 54 Prime SMSA's Estimated Attendance ............. 58 Summer Mean Rainfall ................... 59 LIST OF FIGURES Figure l. boo Major League Baseball Expansion and Relocation l950-l977 ....................... National Football League Franchise Expansion and Relocation 1950-1976 ................ American Football League Franchise Expansion and Relocation l959-l966 ................ Professional Basketball Expansion and Relocation 1950-1977 .................. National Hockey League Franchise Expansion and Relocation 1950-1976 .............. .. . . . World Hockey Association Franchise Expansion and Relocation l950-1976 ................ vi CHAPTER ONE INSTITUTIONAL AND HISTORICAL BACKGROUND During the last quarter-century the professional sports industry in the United States has risen to the status of "big business." Striking players, multimillion dollar contracts, and lawsuits are as much a part of professional sports as championship play-Offs. Basic decisions about franchises are being made for reasons of profit maximization. Winning is no longer the objective function of the owner-manager. Of the nation's 100 major league teams, no more ‘than 12 are owned and managed by individuals whose principal business interest is that particular sport franchise (Durso, 1971). Alan Cohen, operating head of Madison Square Garden and managing director of the New York Knickerbockers of the National Basketball Association (NBA) and New York Rangers of the National Hockey League (NHL) recently stated, " . . . if it's a choice between the Knicks winning and the corporation making money, I want the corporation to make money," (Koppett, 1974). The basic decisions about franchise location, league composition and television policies are being made for reasons consistent with profit maximization, but not necessarily consistent with the provis- ion of providing a better product. Major League Framework and Structure Major league classification is confined to those clubs which field teams of "major league" quality in any of four sports: baseball, 2 football, basketball and hockey. The generally accepted criterion for major league status in each of the above sports is membership respect- ively in either the National or American League of Professional Clubs (NL and AL), the National Football League (NFL), the National Basket- ball Association (NBA), or the National Hockey League, (NHL) or World Hockey Association (NHA). The two leagues in major league baseball are united by the Major League Agreement and for the purposes of analysis will be considered a single league. In hockey, however, the two leagues are economic competitors and considered separate leagues (Demmert, 1973). Operation of Team Sports As Noll points out in his text, Government and the Sports Business, a professional sport league is essentially a cartel with the purpose of restricting competition and dividing markets among firms in the industry. Each league has three types of restrictions, one dealing with interteam competition for players, another with the location of team franchises and a third with the sale of broadcasting rights. Just as league rules limit competition in the acquisition of players so, too, do they limit competition in selling the product of the indus- try. Teams have three important sources of income: admissions, broad- casting rights, and concessions. In all three areas, teams are essenti- ally monopolistic: for each sport only one team in any city has the right to sell tickets to major league professional contests, to offer broadcasts and sell food, beverages and souvenirs to those in attend- ance at its games. Each professional sport has rules governing the location of teams in the league. Although the rules vary among the leagues, the general 3 effect is to prohibit a member team from locating in a city that another team has already designated as its home, unless the latter gives its approval. Thus each team has exclusive rights to sell admissions to major professional sports contests in its home terri- tory. Leagues also control the movement of existing franchises to cities without teams. A team wishing to relocate its franchise must obtain the approval of most other teams in its league. Established teams control the addition of new teams to the league by requiring not only that new teams pay multi-million dollar fees to join the league, but that they also locate in areas approved by the established teams. The exclusivity of franchise rights is threatened only by the emergence of a new league and in time by the existence of inter-league competition, and even this threat is minimal. Only the New York Jets of the now defunct American Football League survived the competition in the home city of an established team. Nevertheless, commercialized professional team sports are a thriving enterprise and one of the most successful and expanding industries in the United States. The following table illustrates the tremendous growth achieved by the four major professional sports over the 25-year period between 1950-1975. 4 TABLE 1 MAJOR LEAGUE FRANCHISE EXPANSION 1950 — 1975 Major League Franchises Year Number of Teams Football 1950 10 1975 26 Basketball 1950 8 1975 28 Baseball 1950 16 1975 23 Hockey 1950 6 1975 24 As Table 1 indicates,since 1950 the number of major league football franchises has increased from 10 to 38, but with the failure of the World Football League (WFL) in 1974, the number of franchises stabilized at 28 for the 1976 season. Professional basketball expanded from 10 franchises to 34, however, this number also decreased to 28, with the failure of the American Basketball Association at the conclusion of the 1975—1976 season. Professional baseball's expansion has been the most conservative, only 10 franchises have been awarded in the 25 year period, all of which are still operating, though several have relocated. Professional hockey has had the most dramatic increase of all professional sports since 1950. With National Hockey League (NHL) expansion coupled with the . creation of the World Hockey Association (WHA) the number of teams rose to 32 before finally reaching equilibrium at 24. The number of major league sport franchises existing in the United States in 1976 was 100. 5 Franchise Failures: Limits to Growth Is there a limit to the number of major league team spOrt franchises our society is willing and able to support? It would seem so. The inflationary rush of new teams and leagues combined with the generally poor economic conditions of the early 1970's has left many teams and several leagues steeped in debt. No fewer than 29 major league franchises have failed since 1970. Professional football has lost 12, Basketball 7, and hockey 8. A recent Louis Harris survey showed a 2 percent decline in fan interest in both basketball and football since 1973. The same survey indicated that baseball dropped 4 percent and hockey 6 percent. Of the 12 sports investigated only tennis and horse racing showed an increase in fan interest (Durso, 1974). Not all major league franchises are successful, as evidenced by the 29 failures previously mentioned. A primary reason for this is that the cost to acquire and subsequently maintain a professional team has increased dramatically over the years. These costs continue to constant- ly rise. The chief cost increase has occurred in the area of player's salaries. Coaching, administrative and scouting salaries have also risen. The largest non-payroll expense is the amount paid visiting teams. Generally speaking major league franchises usually forward 30 percent of their total attendance revenues to league administration offices to pay for the salary and expenses of the Commissioner's Office. The stadium owner in turn collects a percentage of each gate and this percentage varies according to individual contractual arrangements. Football An American Football League (AFL) franchise cost $25,000 in l960. Both Tampa and Seattle are paying $l6,000,000 for their franchises in the NFL. Furthermore, twelve NFL teams lost money during the l975 season, including Washington and Philadelphia which each lost approxi- mately $500,000, while selling every ticket for each home game. Basketball 0f the four major professional sports, basketball is the least profitable. In the National Basketball Association (NBA) the franchise price has increased from the $2 million paid by the Kansas City owners to $6.l5 million remitted by the New Orleans organization. In l967, five NBA teams had salaries below $200,000, in l974, l7 players earned amounts greater than $200,000. Average player salaries of $90,000 are a major reason that in l974 only Milwaukee, New York, and Los Angeles earned a profit in the l7 team NBA (Noll, 1974). Hockey The enormous increase in the number of professional hockey teams began late in the l960's with NHL expansion. It continued with the formation of the WHA, and seems to have a much sounder economic justi- fication than the growth of professional basketball which preceeded it. Although the cost of an NHL franchise has risen to over $6 million al. most all of the sixteen teams are generating a profit. The California Golden Seals appear to be the only non-profitable franchise in the NHL. Owner Charles Finley stated that the Seals lost $l million in l972 and have never generated a profit (Rothenberg, 1975). email Baseball franchise costs have also risen dramatically. Montreal, and San Diego, the league's most recent expansion teams, paid $l0 million for the privilege of National League affiliation. The two most profitable baseball franchises are the New York Mets and the Los Angeles Dodgers. Both teams consistently attract over two million fans in paid attendance. An interesting aspect of the financial situa- tion of baseball is the overwhelming dominance of the National League: in the early l970's the six presumably most profitable were National League team members. Franchise Stability Baseball The best example of franchise stability occured in baseball between l903-l953, when the league consisted of 16 teams in fixed locations. In l953 the league experienced its first franchise shifts when the Boston Braves moved to Milwaukee and the St. Louis Browns relocated in Balti- more (Figure l). In I954 the Philadelphia Athletics transferred to Kansas City. The most devastating franchise relocation took place in l958 when two New York franchises, the Giants and Dodgers, migrated to San Francisco and Los Angeles respectively. This was followed in l96l by the departure of the Washington Senators from the nation's capital to Minneapolis-St. Paul, Minnesota. The league's first expansion took place in l96l with the addition of two teams to the American League, Washington and California. The National League added two additional franchises in l962, one each in New York and Houston. In l966 the nomadic Braves moved from Milwaukee to Atlanta, followed in l968 by Kansas City's transfer to Oakland. Further expansion took place in $3553 3.2.2.2qu ommp Lo mmumuopmm mmwsucmgu mcwmmm omwgucmcu ”ozmwmu smo— Ommp oom~ Nom— mmmp cmmp nmm— I Omm_ onh zmz mm—wmc< mop :z—xoocm oumwucmcu :mm xLo> 3mz mammmb Pacowumz o»:ogoe wpupmmm mpommCsz coumcwzmmz ucmpxmoi- xpwo mmmcmx, mwcq—muwpwza, mcosvp—mm mwzou .um owcgomwpmo p_ocpmo\ coumcwzmmz mmxmhx ucmpm>mPQI zuwu mmmcmx ommowcu wrupmmm mmxsmzymz.x coumom. xLo> 3oz. «ammo; :wowgme< 9 l969 with franchises being located in Seattle, Kansas City, San Diego and Montreal. In l97O Seattle moved to Milwaukee and in l972 Washington relocated in Arlington, Texas. In l977 franchises were awarded to Seattle and Toronto bringing to an end an era of expansion and franchise relocation. Football In l950 the NFL was comprised of 12 teams. Franchise relocation began in l95l when the New York Yanks moved to Dallas (Figure 2). The following year that same team moved to Baltimore. In l960 the Chicago Cardinals migrated to St. Louis and Dallas was granted an expansion franchise. A year later Minnesota also received rights to an NFL expansion team. In l966 with the AFL merger (Figure 3) and expansion in Atlanta the number of NFL teams increased to 24. In l967 an expansion franchise was awarded to New Orleans and in l968 another to Cincinnati. The league remained stable until both Seattle and Tampa were brought into the NFL for the l976 season. Basketball The NBA was comprised of ten clubs in l950 most of which were confined to the nation's northeast quadrant. In l953 the Tri-Cities franchise shifted to Milwaukee (Figure 4). With the failure of both Indianapolis in 1953 and Baltimore in l954 the league decreased to eight clubs. Milwaukee transferred to St. Louis in l955 where they re- mained for 13 years before finally settling in Atlanta. A move to larger cities began in 1957 when Fort Worth shifted to Detroit. Rochester moved to Cincinnati in 1958 and Minneapolis to Los Angeles in l959. The league's first expansion franchise was awarded to Chicago in l96l, the club remained there only two seasons before moving to Balti- more in l963, finally settling in Washington D.C. in 1974. The IO 33: .223 mm .225: Jl mmHMUOwa mecucmLu 32 Lo mcwmwm mmASUCMLm ”QZmeA l 352. l 238m wumccvocwu Tamwz coumom ormeezm I. Lw>zmo M copmzo: H xco> zmz n 95:8 W omwTo cam m xpwu mamcmx mm mcmwpgo 3mz m 3:22 mpommccvz mGFFmo IIIJ mcoswprem .IIIIA mm__mo I] 382:; .>.z II4 mwzou .pm mFMCWULmu ommowcu “Forums oomrucmru cam ommowcu mw_wmc< mo; :opmcwcmmz gmeznmupwa mvga_wum_w:a macewo .>.z u:m_m>w_u Amm :wwgw okay «Na— FNmF moa_ mom~ Nom_ mmmp omm— mmafi omm_ wmm— I omm— onk 3oz copmzoz Smotwmo: Lm>cmo o—mwwsm :opmom mmmfi mwm_ comF mmmp Nomp meF omm_ omm~ mom— I mmmp.onk<004mm oz< onmzcwo xxo> 3oz mcmm—Lo zmz ucapw>m_u use—“Log ofimmwzm mwxzmz_wz chmocm wpupmmm coumzo: Hi 03.5 cum ommomcu .u.o .couchzmmz IIIHI g wLoEprmm l 38.20 IIIIIIr “vogumo zzucoz ugom “To mmmcmx IHIIII I4, wpmccwucwo qummzuom IIIIIIA mwamc< m04 mwyoammccwz IIIL eucu_g< II4 mwzog .Hm IIIIII4 mwxzmzfiwz IIIIII mwruwu wee mv—oamcmwucfi cepmom mcoewu_mm A oomwucmru cam a_;q_mea_w;a xgo> zmz IIIIIIA , e4;a_memr_;a wmaongm onmp wmmp FumF momp mom— Nomp mmmp mmmr mmmp 0mm— mnmp I ommp onh 3oz apcme< omemzm m_304 .pm vCMFxmo mmemc< web cmeznmpuwa Fae; .pm I mwponmm::wz mega.me0_w;a xLo> zwz “Cognac ommumcu mnmp enm— cepmom me— wom— mmmF Nom— mmm— ommp mmme 0mm— onmp I omm— onH<004mm oz< onmzm¥ooz 4cmo chwoca mw_oamcmwucH wpwccwocwu mzmppo mmgwccwz cow:OEum use—m>m_u I; etoepeez copmom Fawn .pm ownmzo ommwo com I4 AFFAI xgemcov ammgww zwz xgo> 3oz _ L0>300cm> megaFmee_w;a oLoEPmem Sweepao mmemc< meg scumso: omm— mnmp vxm— will coaxmo ommuwco mxmfi NNmF omm— I omap onHmxuo: ogmoz o mmstm It 16 immediately with the Dayton club shifting to Houston. The Los Angeles franchise moved to Detroit in 1974 before folding in Baltimore during the 1975-1976 season. The Philadelphia organization relocated in Vancouver in 1973. New York shifted for one season to Cherry Hill, New Jersey, before settling in San Diego. New England, which originally chose Boston as its home, relocated in Hartford in 1973. The WHA began an era of expansion in 1973 adding three clubs: Cincinnati, Indianapolis and Phoenix. In 1975, Denver, the league's most recent club, was grant- ed a franchise. It failed shortly thereafter as did the Minnesota fran- chise midway through the 1976-1977 season. Geogrgphical Distribution of Major League Franchises Table 2 depicts the geographical distribution of professional sport franchises in the United States. Nearly all of the 35 most populous metropolitan areas in the United States have at least one professional club, and 75 percent of the first 20 have at least three. The geographical diversity apparent from Table 8 has not always been the case. During the past 20 years, the industry has expanded and simultaneously shifted its geographical center westward and to the south. Prior to 1958, there were a total of only 41 major professional team sport franchises in 10 cities stretched along the population belt from the Great Lakes to the northeastern seaboard. By 1973 the industry had expanded to 106 clubs distributed throughout most of the United States and Canada, and the number of scheduled professional athletic contests had risen from fewer than 2,000 to more than 4,000 per year. 17 TABLE 2 GEOGRAPHICAL DISTRIBUTION OF MAJOR LEAGUE FRANCHISES Baseball Basketball Football Hockey Total New York 2 1 2 2 8 Chicago 2 1 1 1 5 Los Angeles 1 1 1 1 4 Philadelphia 1 1 1 1 4 Detroit 1 1 1 1 4 San Francisco - 2 1 2 1 6 Oakland Washington, D.C. x 1 1 1 3 Boston 1 1 I 1 4 Dallas - Ft. Worth 1 x 1 x 2 St. Louis 1 1 1 1 4 Pittsburgh 1 x 1 1 3 Houston 1 1 1 1 4 Baltimore 1 x 1 x 2 Cleveland 1 1 1 1 4 Minneapolis - St. Paul 1 x 1 1 3 Atlanta 1 1 I 1 4 Anaheim 1 x x x 1 San Diego 1 x 1 1 3 Milwaukee 1 1 I x 3 Seattle 1 1 1 x 3 Cincinnati 1 x 1 1 3 Denver x 1 1 1 3 Miami x x 1 x 1 Buffalo x 1 1 1 3 Kansas City 1 1 1 1 4 Tampa - St. Peters- x x 1 x 1 burg Indianapolis x 1 x 1 2 Phoenix x 1 x 1 2 New Orleans x 1 1 x 2 Portland x 1 x x 1 San Antonio x I x x 1 Louisville x 1 x x 1 Green Bay x x 1 x 1 Columbus, Ohio x x x 1 1 Hartford x x x 1 1 - _ 18 Problem Statement As previously stated there are 100 professional major league sport franchises located within the United States. The NFL and NBA have recently completed expansion with the addition of two and four franchises respectively for the 1976 season. Hockey expresses plans for expansion in the near future. Baseball, though at present not contemplating expansion, is continually faced with relocation decisions. As the number of major league franchises increases, competition for market space will also increase. Future sites, be they by expansion or relocation, will have significant impact upon the eventual economic success or failure of a particular franchise. When this observation is coupled with the fact that an eventual ceiling on television revenues is approaching, it becomes evident that the average amount of empty stadium seats must be reduced. Thus, the question of franchise location becomes critical. Due to the enormity and obvious logistical difficulties inherent in addressing the location question for all major league sports, the scope of this paper will embrace only major league baseball. The loca- tion problem for baseball is especially critical, in that the sport no longer enjoys the lofty position of being the leading spectator sport. Since there are more teams, playing more games than ever before, total attendance is at a record level. However, average game attendance for major league baseball has declined through the years. Relocation occurs frequently in major league baseball. And each shift in location has had an adverse effect upon the game, i.e., loss of fan support and increased financial burdens for management. With this in mind, the purpose of this paper is to develop a predictive model that will indicate within some degree of probability-limits the 19 potential of a prospective Standard Metropolitan Statistical Area (SMSA) to successfully support a major league baseball franchise. One of the initial and basic assumptions made in this study is that there is a relationship between the spatial variation in the location of professional baseball franchises and socio-economic conditions which also vary spatially. The primary objective of this study is to delineate those variables which influence the location of professional baseball franchises and to match those variables with elements of particular clubs to be located. A further objective is to rank and measure the relation- ship of each variable to the presence or absence of major league teams in a particular area. Another objective is to determine the location prime potential SMSA's for future league expansion of franchise relocation. However, the specific location site within the SMSA is not a concern of this study. Further limitations focus upon the lack of comparable data therefore attention will focus only upon the United States even though major league baseball franchises exist in Canada. Also, attendance will serve as a proxy for success since franchise revenue information is not readily available. CHAPTER TWO ELEMENTS 0F FRANCHISE SUCCESS There are many factors which account for the potential success or failure of a given major league franchise in a particular metropolitan area. Although published research ignores the locational aspects of professional sports, several economists, Neal, 1964, Jones, 1969, El Hodiri and Quirk, 1969 and 1971, and more recently Demmert, 1973, and Noll, 1974, have examined various economic aspects of the professional sports industry. Demmert attenpts to integrate economic analysis with professional sports technological and institutional irregularities, while Noll analyzes the operation of professional sport clubs with particular attention to the effects of government policy on the financial performance of teams. Noll's major purpose is to consider whether the peculiar requirement of sports—~that teams must not differ too greatly in playing strength if popular interest in any team is to be maintained——demands the special status sports enjoy as compared with government policy towards other forms of industry. The pronounced increase in the overall number of franchises would seem to indicate the presence of high rates of return to management. Recent expansion clubs have generally been obtained at much higher prices than expansion teams of a decade ago, in spite of the fact that most recent expansion franchises are located in market areas which. generally would be less attractive than those acquired in earlier 20 21 periods of expansion (see Tables 2, 3, 4, 5, and 6). In 1962, for example, New York and Houston interests paid $3.75 million and $5 million respec- tively, for the right to field professional baseball teams in the National League. In 1969 investors in Montreal, Canada, and San Diego, California, paid $10 million each for that same right (Forbes, April 1, 1971). Sources of Revenue Professional baseball, as well as professional sports in general, is first and foremost a promoter of business and secondarily of sport. If they did not operate on a profit maximization principle, and continuous- ly find total revenues to exceed total costs, all professional sports would follow in the footsteps of the defunct World Football League. A franchise is a firm and its goal is to produce revenue which will exceed costs and theoretically attain a point of profit maximization in a com- petitive market, where marginal revenue equals marginal costs. The largest source of revenue for any major league club arises from the sale, either via tickets or broadcasting rights, of its primary prod- uct, the game. The stream of utility of the game can be divided into three parts: uncertainty of outcome, entertainment value, and the vicar- ious pleasure a consumer attains from relating to a “winner." Historic- ally the potential audience has paid more to identify with a winning club (Demmert, 1973). Other key sources of revenue include broadcasting rights to radio and television, concessions, stadium rental, marketing rights to club emblems, and the sale of player contracts. The value of each of these components increase as the stream of utility increases. The cost function for profit maximization can be divided into two categories, costs associated with specific game items such as arena 22 operation, vendors, attendants, and those incurred in hiring, training and maintaining athletic talent. Davenport (1969) attempted to calculate rates of return for the Baltimore baseball club, a moderately successful member of the American League. He arrived at a rate of return fluctuating from 10 or 11 percent in I'bad years" to greater than 25 percent in "good years." It is inter- esting to note that over the four year period investigated by Davenport, Baltimore's yearly attendance was 969,000, only 35,000 more than the average for the entire American League. Furthermore, Baltimore's other major revenue source, broadcasting rights, has typically been below the league's average. Team Quality and Profits There are two plausible specifications of the individual franchises' objective function. It may act either as a profit maximizer within a given institutional and technological framework, or it may endeavor to maximize an objective function whose elements do include both profit and team quality. Jones (1969) has profit maximizing as entirely consistent with the behavior of the clubs in the NHL, and Koppett (1967) commenting on major league baseball points out that,”When a clear cut choice arises between more victories and more profit, the path toward more profit is chosen." Demmert suggests a strong positive relationship between relative team quality and club profits. Since potential consumers derive utility from association with a winning team, then one would logically expect that demand for attendance would increase with increases in relative team quality. Furthermore, Demmert points out that since the potential audience for game broadcasts is also larger for a winning team, it 23 follows that the media's demand for radio and television broadcasting rights would also shift upward as team quality increased. Broadcast Rights As alluded to earlier the potential success or failure of a franchise hinges greatly on its ability to sell broadcasting rights. With the spread of television to 96 percent of United States households, televised sports has assumed an important role in American life. Corres- pondingly, radio and television broadcasts of professional sports have become a central part of the corporate planning of major league clubs. Presently over 1,000 hours of sports are being broadcast annually. This constitutes over 11 percent of the total network time, and repre- sents an increase of 100 percent in a 10 year period (Durso, 1971). Revenues from television contracts have insured owners against finan— cial losses regardless of a team's success in competition. Reacting to the impact of television revenue, Walter O'Malley, president of the Los Angeles Dodgers, stated ”. . .My goodness, without the National Game of the Week program on N.B.C. many of our teams would be running in the red” (Johnson, 1970). Network payments have risen from $3,250,000 to 18.8 million within the last decade. Also, live sport telecasts have created such interest in professional sports that attendance at sporting events has actually increased for each of the sports viewed regularly (Durso, 1971). Horowitz identifies several impacts the broadcasting of major league events have upon major league franchises, two of which have particular significance with regard to franchise location and success: (1) Franchise location decisions are influenced by potential broadcast revenues, to the detriment of areas with a small broadcast audience. 24 (2) The framework of sports broadcasting increases prices all along the line, including prices paid by sponsors, and to the extent that sponsors do not completely absorb higher adver- tising costs, by consumers. Horowitz completed a statistical analysis of the relationship between the number of telecasts and baseball attendance. The results demonstrate that telecasts do not erode attendance at major league games, but in fact contribute to attendance by generating greater fan interest. In l970, a typical year with respect to home, away and total major league baseball telecasts, Horowitz identified a slight positive relation- ship between home attendance and the number of games televised by each of the 23 United States based clubs. This positive relationship reflects the fact that clubs in more populous areas televise more games and attract larger crowds. Although the magnitude of the effects of fan interest and team per- formance on broadcasting revenues cannot be measured, they affect the size of the broadcast audience, which is of primary concern to sponsors. In view of the importance of broadcasting revenues to the clubs, the potential total audience can be a key influence in managerial decision making (Horowitz, l974). Several recent cases illustrate this influence. The Milwaukee Braves reached a high point of $.5 million for their 1963 local broadcasting rights. This fell to $.2 million in l965 when the Braves had publicized their relocation to Atlanta. An important consideration in the move was the vast southeast broadcasting audience that could be reached by the Brave's sponsors. Indeed, when the Braves moved in l966, they were guaran- teed $l.2 million in annual rights fees, published reports in l964 had alledged a five year, $7.5 million contract awaited. The l96l shift of the original Washington franchise to Minnesota brought a three-year 25 contract that tripled revenues to $600,000. Similarly in 1967 the Kansas City Athletics received only $98,000 for television rights and a total of $165,000 for broadcasting rights. Oakland interests promised the Athletics a television contract worth $705,000 annually, and the Athletics subse- quently moved to California. Broadcast potential has also played a role in major league baseball expansion. The New York Mets entered the league in 1962 with a $1 million contract, and the Kansas City Royals began operation with a $650,000 contract that surpassed by $250,000 the best contract the Athletics had received in Kansas City. Similarly, San Diego began operation in 1969 with a three-year, $700,000 per year contract. The Houston Astro's began in 1962 with a $500,000 contract, but by 1965 it had increased to $1.8 million, the highest in baseball (Johnson, 1970). Drawing Potential and Team Quality The formation, relocation and abandonment of franchises reflects dynamic elements in the evolution of a viable league structure and must be considered if any analysis is to be valid. Quirk and El Hodiri identi- fy several prominent elements, which reflect the dynamic nature of fran- chise success or failure. These include: changes in the size and dis— tribution of population, technological advances, competition from existing or potential rival leagues, and of central importance, drawing potential. They demonstrate that the early history of most major league fran- chises has been characterized by the movement of franchises away from smaller towns as public acceptance of the league grows. Much of the changes in franchise location over the past 30 years represents the expansion of leagues from regional to national organizations made possible by dependable air transportation and by the growth of population centers 26 on the west coast and in the southeastern sections of the United States. In contrast, potential competition accounts for the movement of profess- ional baseball into Minneapolis and New York City. The constant movement of small city franchises during the past decade (i.e. Kansas City, Seattle, and Milwaukee) indicated that in a league's maturity, drawing potential is a primary factor in the movement of franchises. Quirk and El Hodiri investigated the hypothesis that high drawing potential franchises have strong teams while low drawing potential fran- chises have weak teams. Key variables examined to determine the validity of this proposition included: population of metropolitan area percent of games won franchises abandoned franchises relocated championships won 0‘! h (UNI—a Vvvvv The results of their study indicate that the nature of the franchise relocation problem is in “supporting a losing team,” from 1903 to 1970 every American League team with a losing record relocated. Using the championship—won variable as a measure of success, the record points out that in the American League the four largest cities won 49 of 68 champion— ships or 72 percent; and in the National League the four largest cities won 41 of 68 or 60 percent. This information supplemented by the history of abandonment of small town franchises and the records of championships won, offers some evidence of the relationship between drawing potential and playing strength. A fair appraisal would indicate clearly that small city teams have had little, if any, chance of surviving in professional sports. Furthermore, the highest drawing potential areas get more of their share of championships; and the championships-won population comparisons, though not conclusive tend to support the Quirk and El Hodiri hypothesis. 27 Elements of the Demand for Professional Basebajj It can be safely stated that no single factor could be considered as a total explanation of demand for professional baseball contests and subsequently franchise location. Rather, it would appear that a combina- tion cn=elements are relevant in accounting for the spatial distribution of major league baseball frachises and would weigh heavily upon the poten- tial success or failure of a particular location. Defining_the Variables Determination of factors that ought to influence attendance at a major league baseball contest is a relatively easy task. The difficulty lies in grouping the elements into meaningful categories. For the purposes of this study,elements influencing the demand for major league baseball are to be divided in two specific groups: those extraneous to the sport, which stem from the general socio-economic and cultural climate prevail- ing; and those inherent in the business, over which the owner-managers working individually and collectively, can exert a direct measure of con- trol or influence. Demographic Factors Of the outside forces that impinge upon major league baseball attendance, population of the home market merits considerable attention. Population is a primary factor in understanding demand. ‘Therefore, it is anticipated that paid attendance is directly related to the population of the Standard Metropolitan Statistical Area (S M S A ) as well as other demographic characteristics of its market. Demmert (1974) points out that an additional one million residents within an S M S A is equivalent to an increase of attraction by about 28 40,000 per season. Although raw population figures are an important consideration, knowledge of specific socio-economic characteristics of the group at large would provide more insight into the general character- istics of those most likely to attend a major league baseball contest. With this in mind, the racial composition of the metropolitan area is included as an element of professional baseball consumption. Profes- sional team owner-managers have long regarded racial composition as an important factor in determining paid attendance. Industry sources believe that a smaller number of blacks are fans and some have argued that Caucasians attend less frequently if more blacks are in attendance (Scully, 1974). A 1972 Harris survey indicated that the greatest direct consumers are males under age 35 and over age 50. This would seem to indicate that those involved with families are least likely to attend professional sporting events. In keeping with this line of thought it is apparent some identi- fication of the age composition of a metropolitan area is needed to further explain potential attendance at a given site. Percent population over age 65 as an element of demand may not prove to be a powerful element of consumption as those previously identified. Nevertheless, it is included as this researcher believes some measure of age composition of metropolitan areas is desireable. Economic Factors In 1974 the average fan attending a professional sport contest spent $7.10 for a ticket, $.50 for the priviledge of parking and $.40 for concessions while at the event, for a total of $8.00 per person for ap- proximately two hours of entertainment (Ball, 1975). 29 With both ticket prices and total attendance costs rising, one would have to expect an average fan within the market area to have a relatively high income. Their are numerous ways in which income might partially explain attendance. Inner city variations in per capita income reflect among other things, differences in industrial and employment structure, educational attainment and age and racial composition of the population. Per capita income tends to be higher in a city that is in the north or that has a large percentage of its workers in white-collar level occupations, more of its population of prime working age, or more college graduates. It tends to be lower in the south or in a city that has an abnormally large percentage of residents who have not finished high school, who are young, very old, or who work in blue-collar occupations. In order to gain a complete picture of the income level of a particu- lar area several measures will be examined. Among those measures will be the figure of Effective Buying Income developed by Sales Management. This figure is derived by subtracting federal, state and local taxes from personal income. It is generally equivalent to the federal govern- ment's "disposable income” and indicates the general ability to buy and is essential in comparing, selecting and grouping markets. It would seem that the amount of capital available would have a positive relationship to paid attendance. By examination of retail sales, one gains a valid impression of the amounts of goods and services which are traded off within a market area. The greater the amount of total retail sales the greater the indication that residents are more inclined to spend their income in that location for all types of goods and services. These goods and services include the demand for major league baseball. Statistics demonstrate that more 30 individuals are willing to spend a higher amount to attend a sporting event in an area where the amount of retail sales is proportionately greater. Retail sales, therefore, must be considered a primary factor in determining whether an area has profit potential as an active major league baseball market (Sales Management, 1974). Other specific elements considered with regards to income and alluded to earlier include median household income, family income greater than $15,000 and less than $5,000 as well as average family effective buying incomes. Employment Characteristics Another factor that should affect attendance is the metropolitan area's employment structure. The interaction of two factors is herein considered. First, baseball tickets are relatively inexpensive, compared to other sports and forms of entertainment. Most baseball teams have some seats priced at or below $1.50, while very few teams in other sports have any seats priced below $3.00. Second, people in less physically exerting occupations may find the leisurely pace of baseball less attractive. If so, cities having relatively fewer high-paying white collar jobs, such as those associated with government would have signifi— cantly lower attendance than areas with a greater percentage of employees engaged in manufacturing or services. With this in mind, three elements of employment structure will be examined: percent employed in government service, manufacturing and wholesale retail. It is felt that these measures allow one the broad coverage needed to adequately represent the employment structure of contemporary urban areas. 31 Entertainment Competition Another factor which should impact attendance is the amount of competition the franchise is exposed to. This is manifest in two forms. First, the number of major league baseball franchises in the metropolitan area, a factor which can be measured and will be included in this study. Second, the geographical location of the metropolitan area and its environs, which may provide a form of non-sport but defin- itely leisure time competition. However, this is very difficult to cap- ture quantitatively. Population of area would seem to serve as a valid surrogate when one views non-sport entertainment competition as a func- tion of size. Factors Inherent in Major League Baseball As previously stated there exist factors over which management has direct influence. Paramount among these would be a measure of a team's quality relative to other teams in its league. Several indices of team quality are available. Demmert used "games behind the leader" as a quality measurement. The number of star players is another, but a rather dubious method. Noll in his research has experimented with sev- eral methods including: playing success, championships won, percent of games won and games behind the leader. For purposes of this study, and consistent with the Quirk and El Hodiri prOposition that success on the field leads to success at the gate, percent of games won will be the index used to capture the team quality element. The stadium in which a team plays its home games can also affect total paid attendance. Two elements are involved here. First, a newer facility may attract fans because it provides a better view or affords greater comfort. Secondly, the capacity of the stadium imposes an 32 upper limit to attendance. It would be expected that the impact of this variable would be significantly greater in metropolitan areas with more than one franchise. Elements of each can be measured quantitatively, and the age of stadium will serve in the former instance, while stadium capacity will be applied in the latter. Hypothesis In their text McCarty and Lindberg (l966) state that: "A hypothe- sis should be considered simply as an analytical device, a purported explanation of the place to place variation which appears in a problem, stated in the form in which its validity can be tested and the extent of its applicability measures." This investigation is concerned with the distribution of professional baseball franchises. Thus an ideal hypoth- esis would assume the form: There is a significant relationship between the variation of paid attendance to major league baseball franchises within the S.M.S.A. and the variation in the following phenomena outside the control of the owner-manager: Population Effective buying income Total retail sales Percent White population Percent population over age sixty-five Percent population employed in wholesale/retail service Percent population employed in manufacturing Percent population employed in government Percent of population with family income greater than $l5,000 Percent of population with family income less than $5,000 Median household income Average household effective buying income as well as elements directly under the control of the owner-manager in— cluding: Club won-lost percentage Age of stadium Capacity of stadium Number of franchises located within the S.M.S.A. Appendix A lists the sources of information for each variable. CHAPTER THREE THE MODEL: METHODOLOGICAL AND CONCEPTUAL CONSIDERATIONS This study attempts to bring more unified and systematic thought to bear upon the problem of professional sport franchise location. Obviously, the basis for the distribution of franchises is the disposi- tion of resources. Franchise owners seek to locate where earnings will be large and assured. There is justification in approaching the problem of franchise location by investigating the factors, as described in Chaptar Two, that determine the location of the franchises. Hoover points out in his text, The Location of Economic Activity, that although some new locations may offer greater income potential than sites presently occupied, firms will not necessarily relocate. Hoover cites several obstacles which must be overcome if a firm is to success- fully relocate or expand to a near site. All of these have significance for the professional sports industry. A major obstacle is the lack of sufficient knowledge upon which to make a rational decision. Specifically, the client may be unaware of the superior prospects of another location. Secondly, he may be under some artificial constraint, i.e., stadium rental or contract. Also, many times both private and social interests are best sensed by continued operation at a location that may be “obsolete” on a total cost basis as long as no extensive renewal of capital facilities is needed. 33 34 The location of economic activity is by no means a haphazard occurrence, and of course this holds true for the professional sports industry as well. The element of competition from both within and without professional sports will reward and encourage well located clubs and eliminate poorly located franchises, as witnessed by the 27 failures previously cited. Even if new franchises were to be located purely by whim, and if they were never relocated, some semblance of a reasonable pattern would emerge on the basis of competition alone. Therefore, in the location of professional sports franchises as well as economic activity in general, it is not necessary to have both competition and wise business planning to exhibit rational locational patterns. However, each method has shortcomings and a combination of each seems desireable. Competition among poorly informed firms implies that many new clubs are destined to fail, entailing a vast misallocation of resources. On the other hand, an attempt to plan and administer the geography of the professional sports industry without the stimulus of competition shows little promise in the absence of a degree of informa- tion and foresight far beyond our present experience (Hoover, 1948). As Nourse points out, geographical patterns of economic activity are not static, but change with time. In order to analyze the changes, one must first be able to measure them. In order to measure activity in a place, the site must be bounded and hence defined. For the purpose of this study, measurement and subsequent change will be based upon the Bureau of Census' Standard Metropolitan Statistical Areas (SMSA). All of this ordering is necessary if there is to be an effecting of a rational decision—making mechanism for future franchise locations. 35 fleshed The sample consists of 91 SMSA's all measured on 12 socio-economic, externally related parameters. The sample is divided into two groups. Group one consists of those SMSA's (20) measured over a five year period between 197091974 which are the home market for a major league profes- sional baseball club, termed franchised SMSA's. The remaining 71 SMSA's, arbitrarily selected on the basis of population, do not posess a major league baseball franchise and are termed non-franchised SMSA's. As well as the 12 external parameters, four internal or baseball related variables will also be analyzed, however, data for the internal variables only exists for the 20 franchised SMSA's. The goal is to build projections for the 71 non-franchised SMSA's. This is to be accomplished by a factor analyzing of the 12 internal variables to discern patterns which will be good predictors for atten- dance. Factor scores will be written for each case and those determined significant will be utilized with the internal variables in a regression equation based upon the franchised SMSA's with attendance serving as the dependent variable. The slopes and constants derived from the regression analysis are combined to build a prediction equation from which attendance figures will be generated for the 71 non-franchised SMSA's. Factor Analysis Given the nature of the data utilized in this study, factor analysis seemed a logical choice of the techniques available. With 12 variables and 171 observations, it was necessary that the techniques chosen be capable of handling large amounts of data. Factor analysis has that capability. As Nie, Bent, and Hull (1975) state: The single most distinctive characteristic of factor analysis is the data reduction capability. Given an array of correlation coefficients for a set of variables, factor analysis techniques 36 enable us to see whether some underlying pattern of relationship exists such that may be 'rearranged' or 'reduced' to a smaller set of factors. With data as highly correlated as socio—economic variables, it is necessary to employ a technique in which the assumption of inter- dependence is unnecessary. Factor analysis requires no such assumption. In factor analysis, one may begin by considering the data matrix. The initial step of factor analysis involves calculation of a correlation matrix from the data. In this step each variable is correlated to every other variable to determine their relationship. Coefficients of correlation are derived which express the linear relationship between row and column variables of the matrix. It is possible that some dis- cernible pattern or regularity can be derived from a perusal of this correlation matrix. Normally, however, the principal means of analysis will be the principal components or varimax solutions of the next phase in the factor analysis sequence. Principal components analysis is per- formed on the correlation matrix to determine major patterns of data. If this is viewed geometrically, each observation would be defined as a coordinate axis of geometric space, and each characteristic of the observation would be considered as a point located according to its value for each SMSA. These observations and characteristics are defined on the basis of a cartesian coordinate system. By laying out lines from the origin to each characteristic vector representations of the data are derived. The angle between the vectors so obtained represents the relationship between the characteristics and the observations. The con- figurations of these vectors will be representative of the interrelation- ships of the characteristics. Characteristics highly interrelated will cluster together and these clusters will be indicative of the patterns 37 that exist. Each cluster will be defined by a factor, which is located in vector space by weighing each vector equally and fitting the factor in a gravity like situation. By determining the relatiOnship of each characteristic to each factor, a value known as a factor loading is derived. These loadings can be used to determine the strength of the relationship between the characteristics and the factor. In the principal axis solution, factors maintain a 90 degree separ- ation and each variable loads on each factor; hence, this solution may not precisely delimit the patterns that exist. In order to achieve the maximum clustering of selections the orthogonal factors are normally rotated. By subjecting the orthogonal factors to varimax rotations, it is possible to obtain separation of variables so that a limited number of variables are loading on each factor. Having utilized varimax ro- tation, the eigen values can be examined to obtain a notion of which clusters of characteristics are accounting for the greatest proportion of the variation in the number of professional teams per SMSA. In order to obtain some concept of how each individual observation is relating to the factors derived, the factor scores are analyzed. Factor scores are weighted measures of the way an observation rates with a factor. Thus, factor scores can be used in scaling observations and in mapping them according to each individual factor derived in the analysis. The factors that are obtained are composites or clusters of elections. By mapping the factor scores, it is possible to differentiate within each factor as to the strength of the relationship of individual SMSA's to each factor or cluster of selections. Factor analysis can be employed in a deductive approach involving the hypothesis of the existence of particular dimensions and factor analysizing the data to see whether these dimensions emerge. It has 38 been hypothesized that the ability of an SMSA to support a major league baseball franchise is dependent upon its population and economic viability. Therefore, both population and economic factors should emerge in which variables measuring aspects and functions of both demographic and economic indicators cluster. Furthermore, it will be interesting to note whether major league baseball franchise location is associated with more than one cluster of variables. Although results from the factor analysis could be considered as final output, the factors will be utilized as inputs into the final analysis. This will entail usage of conceptual socio—economic "factors" as independent variables and regressed against the dependent variable paid attendance. Correlation-Reg:e§sion Analysis A regression model is appropriate for situations wherein a function- al relationship is postulated. When using regression analysis, one is faced with selecting independent variables which may serve as a surrogate for a set of highly related variables. Regression analysis does partially account for interrelationships of variables though partial correlation coefficient, but assumes that “independent“ variables are in fact independent. Adelman and Morris in their text, Society, Politics and Economic Development: A Quantitative Approach, indicate concisely the difference between regression analysis and factor analysis. Technique of factor analysis shares certain characteristics with both non—quantitative comparative studies and statistical regression analyses. In essence, it is equivalent to a systematic application of comparative studies which simultaneously test a large number of ceteris paribus propositions. 39 As in regression analysis, factor analysis breaks down the original variance of a variable into variance components associated with the variation of a set of other quantities. In regression analysis, the variable whose variations are decomposed in this manner is known as the dependent variable, and the variables that account for different portions of its variation are the independent variables. In factor analysis, all variables are dependent and independent in turn. Thus, by contrast with regression analysis which is a study of dependence, factor analysis is a study of mutual interdependence (Adelman and Morris, 1967). The ultimate goal of science is that of prediction. A primary function of regression analysis is to find the best linear prediction equation and evaluate its prediction accuracy. This technique is an appropriate test of a hypothesis which states that variations in the magnitude of the problem (dependent) variable are related mathematically to variations in the magnitude of several other, explaining (independent) variables. It represents a test of spatial association. The multiple regression equation is of the form: Y = + + + .,,+ c a b1 x1 b2 x2 b3 x3 bN xN Since the hypothesis of this paper is concerned with spatial assoc— iation and is stated in similar form, the regression technique is applicable. From the regression analysis, three types of coefficients will be obtained. Robinson, Lindberg, and Brinkman describe these three as 1) a coefficient of correlation (r) which describes the degree of association between two variables; 2) a coefficient of multi— ple correlation (R) which describes the degree to which three of more variables are associated, and 3) the coefficient of partial correlation, again describes the degree of association between two variables, but accomplishes this by holding other variables which may be interrelated, constant. All of these represent summary measures for the entire study and are used to describe the degree of a real association among selected 40 variables to test the validity of hypothesis formed for the research project. As Robinson, Lindberg and Brinkman state, the value of R is a "numerical description of the linear association between the dependent variable and all the independent variables included in its computation." Implied in this statement is the basic assumption that a linear relation- ship exists between variables. 41 Interpretation of the Factor AnaLysis The results of the factor analysis are summarized in the matrix of common factor coefficients presented below. 10. 11. 12. TABLE3 VARIMAX FACTOR LOADING MATRIX Variables . Population . Effective Buying Income . Total Retail Sales Percent White . Percent Over Age Sixty-Five . Percent Wholesale/ Retail Employment . Percent Government Employment . Percent Manufacturing Employment . Income Greater Than $15,000 Income Less Than $5,000 Average Effective Buying Income Median Household Income Factors I II III IV .9878 .9834 .9843 .8167 .7319 .9478 .9317 ~.8750 .9734 .9749 .9568 -.7888 42 The factoring of the correlation matrix produced five factors (Tableii) that accounted for 91 percent of the total variance of the data. TABLE4 PERCENT VARIANCE PER FACTOR 35 3O 25 20 15 IO I II III IV V FACTOR As Table 4 indicates, factor I accounts for 30 percent of the variability in the data. The characteristics having their highest loadings in factor I are personal income greater than 15,000 dollars, average effective buying income, medium household income and personal income less than 5,000 dollars. More specifically, factor I may be interpreted to represent both high individual and family income levels. The low negative loading of family income below 5,000 dollars helps reinforce the high income nature of this factor. High positive factor scores were considered characteristic of sufficient income levels to support professional major league baseball, while low negative values were indicative of areas lacking necessary 43 individual and family income to warrant a major league franchise. It is interesting to note, however, that 44 percent of the observa- tions of franchised SMSA's recorded factor scores with negative values. Further, of the 20 franchised SMSA's, only two, Detroit and Chicago, recorded positive values for each of the five years examined. Clearly, it would appear that the income factor can be discounted as a significant force in attracting franchises and positively impacting attendance. Factor 11 accounts for 26 percent of the variability in the data. The socio-economic indicators with their highest loadings on factor II are SMSA, population, effective buying income, and total retail sales. In particular, the pattern of associations incorporated in factor II is strongly suggestive of size and economic viability. This factor is termed "economic strength." Clearly suggesting strong relationship between the presence of a franchise, or at least the potential to support a franchise with the population of the SMSA, related buying power index and spending rates. An examination of both franchised and non-franchised SMSA factor scores provides further insight into the relationship between factor II (economic strength) and franchise location. 44 TABLE 5 ECONOMIC STRENGTH AND POPULATION SMSA POPULATION FACTOR SCORE ECONOMIC NUMBER OF RANK RANKING STRENGTH BASEBALL FRANCHISES New York 1 1 4.01484 2 Chicago 2 2 2.06416 2 Los Angeles 3 3 2.16346 1 Philadelphia 4 4 1.06654 1 Detroit 5 5 .7449 1 San Francisco 6 6 .53812 2 Boston 8 7 .43736 1 Dallas 10 13 —.2547 1 St. Louis 11 8 .08826 1 Pittsburgh 12 9 .06356 1 Houston 13 12 -.14958 1 Baltimore 14 10 -.O312 1 Cleveland 16 11 -.14334 1 Minneapolis 17 17 -.36202 1 Atlanta 18 18 —.38614 1 Anaheim 19 19 -.55216 1 San Diego 20 15 -.35282 1 Milwaukee 21 16 -.35840 1 Cincinnati 23 13 -.33088 1 Kansas City 27 14 -.33942 1 Table 5. depicts the relationship of population ranking to factor score generation. As would be expected, there is a direct correlation. 45 However, factor II values take a definite decline for the 7 least popu- lated SMSA's. It is interesting to note that 9 non-franchised SMSA's record higher factor II values than Cincinnati's .33088, also Seattle, the new site for the 1977 season, is not among them. Of the seven franchise cities in question, Minnesota, Milwaukee and Kansas City have previously lost franchises while Anaheim, San Diego and Atlanta have publicly discussed relocation several times. Only Cincinnati qualifies as a deeply entrenched, highly successful operation. Of the non-franchised cities scoring higher than the seven lowest franchised SMSA's, five (New Orleans, West Palm Beach, Miami, and Birmingham) are located in geographical areas where climate is a limiting factor. However, technology in the form of domed stadiums makes these areas prime possibilities for future consideration. Nassau-Suffolk, New York, Riverside, San Bernadino—Ontario, and Newark also qualify as strong candidates. Washington, D.C. completes the list of high ”economic strength" scoring non—franchised SMSA's. In reference to economic strength, it appears that Washington, D.C. is an anomaly. Its factor score of .4659 is higher than all but the six largest franchised SMSA's yet Washington, D.C. has lost more fran- chises than any other SMSA. Clearly there is more at work than economic strength. It seems that Washington, D.C.‘s unique position as our nation's capital demands that the national pastime be located within the SMSA, but by its very nature, eliminates or at least restricts opportun- ity for success. Generally, it would appear that a high factor 11 score points to potential in terms of socio-economic criteria needed to support a profes- sional baseball franchise. Factor III is unique in that only one variable (percent wholesale/retail) recorded a value of (.9478) above 46 the .70 cut-off level. Factors with more than one variable loading highly are termed common factors. Common factors account for the variables' intercorrelations, whereas unique factors represent that portion of a variable not accounted for by its correlations with other variables in the set. Factor III termed "services” accounted for approximately 12 percent of the variability in the data. Generally, franchised SMSA's recorded negative values on this factor, which reflects the high percentage of employment involved in manufacturing common to many of the franchised SMSA's. Positive scores on factor III were restricted almost exclusively to the south and west, areas of more recent and expanding urban develop- ment. This regional shift in population and employment structure is re- flective of the everchanging nature of our urban areas. It would seem at this point that employment structure would be a significant measure of a community's ability to support a professional sport franchise. Furthermore, relatively large population percentages employed in manufac- turing and services would be preferable to similar percentages employed in governmental service or high percentages of retired individuals. Factors IV and V, termed Retirement and Governmental Service, con— tributed 23 percent of the total variability of the data. The percent of population over age 65 was perceived as an important element with regards to the presence or absence of professional baseball franchises. However, upon examination of the factor scores this does not appear to be the case. Only four franchised SMSA's recorded positive values for this factor and it is evident that population percent over age 65, as a causal factor is of little consequence. The emergence of this factor appears to be a function of the type of data analyzed, as opposed to its 47 relationship to successful franchise location. Factor IV,termed "WhiteCollar/Government Service} though contributing over 13 percent of the variability does not significantly add to the analysis. At this point, it appears that high percentages of government employment would work against successful franchise location. The primary example being the case of Washington, D.C. The nation's capital ranks eighth in total population and seventh in economic strength, yet it has lost two baseball franchises in the last 25 years. Conceivably, this may be attributed to the unusually high percentage of residents employed in governmental service. Both San Francisco and San Diego also recorded high values on factor V. Although they have not yet lost a professional baseball franchise, they rank well below the league average paid attend- ance mark and have discussed relocation openly on several occasions. Summary In summary, the original 12 socio—economic variables analyzed were reduced by approximately 60 percent to five conceptual variables. These conceptual variables retained over 91 percent of the original information found in the total matrix. Although results from the factor analysis could be considered as final output, the conceptual variables will be utilized as inputs into the final analysis. The conceptual variables determined in this analysis and the factor scores generated provide one of the basic inputs into the analysis of socio—economic structure and professional baseball franchise location. CHAPTER FOUR THE MODEL: ANALYSIS AND RESULTS Implementation of factor analysis and the subsequent delineation of conceptual variables solidifies the hypothesis, making it necessary to restate it at this point. Thus the revised form appears as: There is a significant relationship between the variation in the attendance of major league baseball (Y) and the variation in: X1 Economic Strength (Factor II) X2 Government/WhiteCollar Employment (Factor III) X3 Percent of Games Won (PCT) Number of Baseball Franchises in SMSA (MULT) X5 Stadium Capacity (CAP) Age of Stadium (AGE) The null hypothesis becomes: H0 R = O = b1 + b2 + ...b6 Testing for significance was accomplished at the .05 level of confidence 48 49 The systematic evaluation of the hypothesis is an integral part of the problem solving process. As McCarty and Lindberg have recognized: "The test of any hypothesis, whether simple or complex involves measuring the degree to which the hypothesized distribution matches the problem distribution," (McCarty, Lindberg, 1966). The primary activity is to compare two sets of values Y and Yc. The former represents the actual magnitudes of paid attendance for major league baseball franchises. It is the problem distribution as well as the dependent variable. These are the values that are computed for “Y locations” on the basis of the dependent variables. If one formulates a completely accurate hypothesis, the magnitude of all occurences of Y are exactly equal to the magnitudes of Yc at those same locations. Before a comparison of this nature can be made the Yc values must be calculated. The method of calculation used in this study is a com- bination of standard multiple regression and stepwise procedures. Through the use of this routine a candidate for inclusion into the equation is selected from among all those independent variables not yet included in the equation. The variable selected is that independent variable which will reduce unexplained variance around the mean of the dep6ndent variable the most. Conversely, the selected independent variable is that one which will raise the coefficient of determination (r2) the greatest. The regression analysis was completed with the franchised SMSA's over the five year period from 1970—1974. The stepwise procedure was utilized with a minimum F at .01 and the tolerance to enter at .001. Both stadium capacity and age of stadium were eliminated at this point, as they failed to meet minimum levels for inclusion. However, a conceptual variable was created which did meet minimum levels for inclusion. 50 The interaction between percent of games won and factor III (PCT/FS III) was calculated and included in the final regression equation. Simple coefficients of correlation (r's) were computed between each variable and every other variable. The result appears in Table 6. TABLE 6 SIMPLE COEFFICIENTS OF CORRELATION Attendance 1.00000 Economic Strength .67443 1.0000 Government/White-. —.20127 .04054 1.0000 Collar Employment Won-Lost Percent .53987 .19490 - 05315 1.0000 Number of Franchises .29917 .66752 .16529 .22077 1.00000 FS III/PCT -.22146 .04749 .98672 —.O3982 .19621 1.00000 The values of (r) describe the degree of correlation between any two variables. This model also measures the strength of the univariate relationship between the dependent variable and each independent variable. Additional utility is derived from these coefficients when they are used to obtain estimates of the necessary regression coefficients. Substitu— tion of these coefficients in the regression model results in the follow— ing equation: Attendance = -316614.28 + 330360.25 F511 + 2843.17 PCT — 339865.62 MULT - 1192.33 FSIII/PCT + 465389.70 FSIII The preceeding equation is a generalized description of the manner in which the five hypothesized independent variables are related to the distribution of attendance for major league baseball franchises when considered simultaneously. Since it is a generalization, its value in description and prediction depends upon how accurately it "fits” the 51 actual distribution of attendance. In order to obtain a measure of its "fit", it is necessary to consider another correlation coefficient, the coefficient of multiple correlation (r). As Robinson, Lindberg, and Brinkman have indicated, ". . . the meaning of R is similar to that of Y in that it is a numerical description of the linear association between the dependent variable and all the independent variables included in its computations" (Robinson, Lindberg, and Brinkman, 1961). In this instance R assumes the value of .86157 (significant at the .005 level). The critical level of the correlation coefficient (r) with 90 degrees of freedom is .267 (Fisher, Yates, 1952). Clearly this relationship is not likely to occur by chance and on the basis of the significance of (r) one can confidently reject the null hypothesis presented previously. It is obvious that an association exists between the spatial variation of baseball attendence and the spatial variation of the five independent variables. A more accurate indication of the extent of this association can be gained by squaring R to compute the coefficient of multiple deter- mination (r2). The r2 value of 0.74320 indicates that approximately 74 percent of the actual problem distribution is accounted for by the regression equation (hypothesized) distribution. To more clearly under- stand the effect of the five independent variables upon the problem distribution, each variable is examined individually. Table 7, illus- trates this relationship and Table 3 identifies individual variables grouped together to form the conceptual factors. 52 TABLE 7 CHANGE IN THE COEFFICIENT OF MULTIPLE DETERMINATION VARIABLE R SQUARE CHANGE IN R SQUARE FSII .45486 .45486 PCT .62826 .17340 MULT .69406 .06580 FSIII/PCT .72615 .03029 FSIII .74230 .01615 The r2 for FSII is .45486. The simple squared correlation (r2) between the dependent variable (attendance) and FSII is also .45486. NI. When PCT is added to the equation r2 becomes .62826. Prediction is improved by approximately 17 percent through the addition of PCT. Similarly as MULT is added to the equation, r2 increases to .69406, the addition of MULT to the equation has accounted for an additional .06580 of the variation in attendance, improving prediction accuracy by 6.58 percent. FSIII/PCT increases r2 to .72615, a gain of 3.21 percent. With the addition of FSIII the previous value is increased by .01615, F2 is now equal to .74230. As Table 7 illustrates the addition of new variables to act on the variation left unexplained by those already in the equation add less and less to the prediction accuracy of the equation. Interpretation of Predicting Variables It is not surprising that “economic strength“ (FSII) emerges as the most potent"exp1ainjng variable”. As the factor analysis suggests, franchise success and economic strength variables co—vary. Therefore, one would expect a strong relationship between an SMSA's economic strength and the success potential of a professional baseball franchise 53 located in that region. Further evidence of the impact of economic strength as a factor influencing attendance is evidenced by the fact that in 1971 four major league franchises finished thirty or more games behind the eventual league champion. Only one, Philadelphia, drew more than one million fans. The Philadelphia SMSA contains nearly five million residents and registered a factor 11 score of 1.2387. All the other franchises are located in SMSA's with populations less than three million and signifi- cantly smaller scores on factor II. Given the negative correlation between attendance and government/ white—collar employment, baseball appears to attract the working class fan. Two factors may account for this negative relationship. Baseball tickets are relatively inexpensive compared both to other sports and other forms of entertainment. Also, individuals employed in less physically exerting occupations may find the sedate pace of baseball less attractive. If so, cities having relatively more higher paying white—collar jobs would have lower attendance figures. Both factors are related to the suggested long term decline in the popularity of baseball. Increased government and white-collar employment levels are associated with both a declining sensitivity to price differentions among entertainment options and a smaller relative number of persons in physically exerting occupations (Seymour, 1971). Won—lost percentage apparently has a strong effect upon attendance during the season,but even a greater effect upon the attendance of succeeding years. If a highly successful team can increase season ticket sales by only 2,500 for the following year that will net an increase of over 200,000 in paid attendance for the season. Further, this suggests as Noll (1974) points out that team attendance will 54 be sustantially higher if several teams alternate winning seasons than if one team tends to dominate. Beta Weights Because our attention is focused upon prediction rather than measures of degrees of relationship the Beta weights were selected for analysis and appear in Table 8. Beta weights indicate how much change in the dependent variable is produced by a standardized change in one of the independent variables when the others are controlled. Analysis of Beta Weights enables one to simplify the linear regression equation. Furthermore, when there are two or more independent variables measured on different units (such as economic strength in factor scores and team quality in winning percentage) standardized coefficients may provide the only sensible way to compare the relative impact to the dependent variable of each independent variable (Nie, et al., 1975). TABLE 8 BETA WEIGHTS Variable Beta Weight FSII .77094 PCT .44936 MULT -.25465 FSIII/PCT -.98045 FSIII .80087 Blalock, in discussing the relationship between the unstandardized equation and the Beta weights points out that: 55 The partial correlation is a measure of the amount of variation explaineg_by one independent variable after the others have explained all they could. The beta weights, on the other hand, indicate how much change in the dependent variable is pro- duced by a standardized change in one of the independent variables when the others are controlled. (Blalock, 1972). In this study the standardized equation would take the following form: Attendance = .77FSII + .45PCT —.25MULT -.98FSIII/PCT + .8OFSIII Thus, a change of 1 unit in FSII, with the other independent variables remaining constant would produce a change of .77 standardized units in attendance. For the first time one gains insight into the actual impact each of the independent variables upon attendance. Further it is clear for the owner—manager as to which areas he may allocate resources to maximize attendance. Residuals_from Regressiqp Thus far, none of the measures used in the analysis has indicated the spatial variations in the degree to which the regression equation fits the actual distribution of baseball attendance. As indicated in the initial portion of this chapter, a comparison of the distribution of Y and Yc is a key feature of an analysis of this type. To accomplish this task residuals from regression must be considered. Such a residual may be defined as: ”that part of the magnitude which a phenomenon reaches within a given area which is independent of the areal assoCia- tion between the given phenomenon and the other factors included in the investigation“ (Thomas, 1968). In this analysis residuals were determined by subtracting Yc from Y. It should be noted that negative residuals are cases where the regression equation overestimates the value of Y and the positive 56 residuals are cases where the equation underestimates the value of Y. The residuals from the regression analysis appear in Appendix 3. Only five SMSA's fall outside one standard error and none fall beyond two standard errors. However, in four of the five cases this error may be explained by the completion of a new stadium, the fifth case being Washington, D.C. CHAPTER FIVE THE MODEL APPLIED: IMPLICATIONS AND FUTURE DIRECTIONS The preceeding statistical results have important implications for the expansion and relocation of major league baseball franchises. Within the industry a common rule of thumb is that a franchise needs to attract approximately 850,000 fans to its home games in order to break even. Noll (1974) points out that a franchise needs to attract One million spectators if it expects to earn a modest profit of $100,000 - $200,000. This appears to be the minimum level of profit necessary for a profit oriented franchise to remain in an SMSA. It is assumed that the predictive model can be generalized to non franchised SMSA's, that the regression coefficients remain stable over time and that general ceteris paribus conditions hold. To apply the predictive equation to the non franchised SMSA's, PCT is set to equal .500 and MULT is set to equal zero. The resulting equation takes the following form: Estimated Attendance = -3l66l4.28 + 330364.25 FSII + 465389.70 FSIII + 2843.17 (PCT 500) — 1192.33 FSIII/PCT. The above equation represents the actual relationships between attendance at major baseball contests and the hypothesized independent variables identified in this study. 57 58 Viability of Non-Franchised SMSA's Estimated attendance figures for the 71 non-franchised SMSA's are generated from the prediction equation and appear in Appendix 5, 0f the 71 SMSA's investigated only 10 generated attendance figures larger than the suggested figure of 1,000,000 which represents profit-making. Table 9 lists these SMSA's and provides their estimated attendance figures. TABLE 9 PRIME SMSA's ESTIMATED ATTENDANCE SMSA ESTIMATED ATTENDANCE Tampa - St. Petersburg 1,127,127 Newark 1,113,237 Miami 1,106,545 Tulsa 1,040,913 Greenville — Spartanburg 1,040,411 Nassau - Suffolk 1,038,592 Greensboro - Winston Salem — High Point 1,034,995 Allentown Bethlehem — Easton 1,033,490 Birmingham 1,019,824 Youngstown - Warren 1,013,722 As Table 9 illustrates, only Tampa - St. Petersburg, Newark, and Miami stand out as prime future franchise locations. The Tampa - St. Petersburg SMSA generated the greatest amount of estimated attendance and with the recent placement of an NFL franchise there, they have already attained major league status. However, baseball operates under a different set of constraints than does professional football, and a successful football franchise is no indicator that a major league 59 baseball franchise will also be successful. One of the elements not investigated in this study that definitely impacts the potential success of Tampa - St. Petersburg as well as Miami as a viable location site is summer rainfall. Traditionally baseball is played out—of—doors during the summer months. During this period both Tampa and Miami experience large amounts of rainfall. The combined total of rainy days during the months of May, June, July, August and September for Tampa and Miami equal 82 and 99 days respectively. Table 10 illustrates mean rainfall for both SMSA's during the baseball season. TABLE 10 SUMMER MEAN RAINFALL Miami, Tampa — St. Petersburg SMSA's Miami Tampa - St. Petersburg Mean Rain- Number of Mean Rain- Number of Fall (Inches) Rainy Days Fall (Inches) Rainy Days April 3.60 6 2.00 8 May 6.12 10 2.41 15 June 9.00 15 6.49 18 July 6.91 16 8.43 20 August 6.72 17 8.00 20 September 8.74 18 6.35 18 Total 41.09 82 33.68 99 Source: Statistical Abstract of the United States, 1976. The information presented in Table 'U3clear1y illustrates the poten- tial problems summer rainfall would have upon both Tampa - St. Petersburg and Miami and reduces their attractiveness as a viable location site. It is doubtful that a major league baseball franchise would locate in 60 an area with such high potential for summer rain without the benefit of a domed stadium. When attention is focused upon two prominent facts, the Newark - Patterson area emerges as the "ideal" location for either an expansion or relocation major league baseball franchise. The first fact is that Newark is bordered by two other potentially strong non-franchised SMSA's. Paterson ~ Clifton — Passaic, located directly north 0 f Newark generated an estimated attendance of 980,979. While Jersey City, situated to the east, was estimated at 972,019 by the prediction equation. When these figures are combined with the 1,113,237 predicted attendance figures for the Newark SMSA, the area's potential for a successful franchise location becomes apparent. Secondly, the regression equation discussed in Chapter four underestimated the potential of the New York SMSA to support yet another major league baseball francise indicating even greater potential for that particular region. When combined, these two facts further enhance Newarks position as the ideal site for future major league base- ball franchise location. If one accepts the one million attendance level as a minimum for a profit—seeking franchise, very few franchised SMSA's exhibit the potential to support an additional major league team. As previously mentioned the New York City SMSA could probably support one more franchise as well as the Dallas — Fort Worth SMSA. Certainly, though, the success of these sites would be a function of subsequent league expansion and/or the relocation of other franchises within the league. SMSA's which are often mentioned as potential franchise locations include: New Orleans, Denver, Indianapolis, Memphis, Phoenix, Washington, D.C., Seattle and Buffalo. Results of this study indicate that these 61 SMSA's are for the most part on the borderline of viability and would not be able to support a profit seeking major league baseball franchise (Appendix c). The regression Analysis further indicates that Cleveland, Baltimore, Minneapolis — St. Paul and San Francisco - Oakland, under present operating conditions are not viable franchise locations. This is further evidenced by the fact that all franchises located in those SMSA's record low attendance figures relative to league averages. It is therefore extremely likely that if relocation is to occur, at least one of these SMSA's would be involved. The Seattle Situation As previously stated, Seattle does not appear to be a viable site for the location of a major league baseball franchise. .Nevertheless, it has been awarded a franchise for play during the 1977 season. Seattle is both a present and a former franchise site, and is predicted to be on the borderline of viability with an estimated attendance of 909,671. Under normal circumstances Seattle would be considered deficient in terms of economic strength (FSII) to support a franchise, but it has the advantage of access tOEilarge regional broadcasting audience as well as being located in close proximity to another metropolitan center, Tacoma, which was estimated to have generated an attendance of 662,031 by the prediction equation. Certainly these combined factors could contribute to the successful location of a major league baseball franchise in the Seattle SMSA. A further consideration, which may reduce the area's potential to generate the predicted value, is the fact that as an expansion franchise it is not likely Seattle will register a won-lost percentage of .500. Therefore, it is fairly safe to assume in this instance that the predicted 62 attendance is an inflated figure. The fact that the Seattle franchise will play in a newly constructed domed stadium could offset the negative effect of a poor won—lost percentage. This fact and the novelty of a major league baseball franchise will increase attendance, but again the limiting factor will be the quality of the team as evidenced by the won- lost percentage, and as the novelty wears off, attendance will decline. Gate-Sharing Arrangements Noll (1974) discusses expansion and points out that expansion prospects for major league baseball are quite dim. He attributes this to the gate—sharing arrangement presently in effect in major league baseball. The current arrangement mandates that the home team retains 80 percent of the gate receipts in the American League and 90 percent in the National League. If attendance revenues were divided equally, substantial expansion would be possible. Even if this resulted in smaller home attendance figures and smaller broadcasting revenues for franchises in the smaller SMSA's, they would still be in a better position to register profits due to a reduction in the significance of the economic strength factor (FSII) and an increase in the revenue generated by road games. _§MSA's Motives for Franchise Location As stated in Chapter One, 29 major league sport franchises have failed since 1970. This could lead one to believe that there is a limit to the number of major league franchises the United States can support. This thought is further reinforced when one considers that of the 24 major league baseball franchises in existence in 1976, only nine claimed to have made a profit during the 1976 season. Those claiming a 63 profit include: California (Anaheim — Garden Grove), Cincinnati, Houston, Kansas City, Los Angeles, Milwaukee, Montreal, Philadelphia, and San Diego (Block, 1977). Block acknowledges that more major league baseball franchises are now, or soon will be, placed on sale than at any time in recent memory. He attributes this to both increases in players' salaries and operational costs. Nevertheless, large metropolitan areas continually attempt to attract a major league franchise. In a recent study completed by the University of Pittsburgh and the Pittsburgh Chamber of Commerce, Civic pride and community self interest were identified as dominant factors in explaining an SMSA's desire to attract a major league baseball franchise. The study indicated that major league baseball contributes $21 million to the economy of the Pittsburgh SMSA. Further, the Pittsburgh franchise attracts 560,000 people from outside Allegheny County to the central city for the purpose of attending major league baseball games. Each visitor spends an average of $5.77 in downtown Pittsburgh after each of the 81 home games. Finally, visiting teams contribute $400,000 to the areas restaurants and hotels, (Mulloy, 1977). It is clear that the benefits which accrue to an SMSA as a function of the location of a major league baseball franchise are significant and many. It is also obvious the owner-manager of a major league franchise risks large sums of money with each location decision. If for no other reason it is critical that the owner~manager have complete and reliable information before such a decision is reached. 64 Areas for Future Research A major purpose of this study is to bring more unified and systematic thought unbear upon the problem of major league baseball franchise location. This study increases the amount of available knowledge from which rational decisions may be made, but due to the dearth of such previous studies can only be viewed as a beginning. In the course of examining the major league baseball location problem, several questions not addressed in this study need to be confronted in future research. The first, and probably most important question would be that of the relationship of ticket prices to attendance. In a management situation that features both an inflated dollar and rapidly rising costs the current I or future major league baseball owner is obviously going to have to face the question of increasing ticket prices. This fact makes the variable of ticket pricing a most important element for future consideration. The influence of this variable is very difficult to capture in that there is always present an extremely wide range of ticket prices at any one given event. This problem is further compounded by the fact that a plethora of promotional inducements tend to widely distort both attendance figures and the effect of price upon the marketplace. A second question worthy of future consideration would be that of the quest for the ideal stadium location. This problem has long haunted major league baseball owners, and yet at present there appear to be only two viable alternatives to this problem, either a downtown or a suburban location. Remarkably, virtually no research has been published in this problem area of stadium location, although it is commonly agreed that stadium location can be a most important component in the decision-making process of a potential spectator. 65 Another question area for future consideration would be that of season ticket holders. Season ticket holders are a most important element in the potential success or failure of a major league franchise. Although they may be a small figure when compared to stadium capacity they become an important element when received as guaranteed spectators for all 81 home games. But beyond this consideration is the fact that little re- search has been published delineating, in-depth, the socio-economic charac- teristics of major league baseball season ticket holders. Such a series of profiles would provide a much needed comparative basis that would be invaluable in the location and marketing of fugure major league baseball franchises. General Summapy This study has investigated the potential of SMSA's to successfully support major league baseball franchises. This study concentrated on paid attendance. Since profit maximization is the objective function of most major league baseball franchise owners this study focused upon the only available and representative component of the economic infra- structure of these franchises - paid attendance. A model was built to predict this paid attendance. The model identified five variables which significantly contributed to an explanation of paid attendance. The variables that were identified by the model include two conceptual variables as defined by factor analysis. These two variables described elements of economic strength and the employment structure of the specific SMSA's. Two variables internal to the nature of baseball were also identified by the model to be significant. These variables include both the percent of games won and the absence or presence of another major league baseball 66 franchise within the same SMSA. A third conceptual variable was created by the interaction of the “employment structure factor with the percentage of games won. The resultant regression equation accounted for approximate- ly 74 percent of the variation in paid attendance, and a consequently much more unified and systematic overview of the role of paid attendance within the infrastructure of major.league baseball franthises. The final task of this study was to construct attendance projections for non-franchised SMSA's within the parameters of the study. Upon completion of this task 10 SMSA's emerged as potentially strong major league baseball franchise locations. From this total only three SMSA's were deemed ideal future locations in that only they suggested the potential to compete successfully in the market place._ These three SMSA's were Tampa - St. Petersburg, Newark and Miami. They emerged as stronger sites than both the SMSA's of New York and Dallas — Fort Worth -— the only SMSA's that had been identified by the model as potentially able to support an additional major league franchise. General Conclusions There have been three phases in the evolution of major league base~ ball. The first was the Developmental Phase. This was that period from 1903 to 1961 (Table 2). This phase was characterized by a stability in the number of teams at, sixteen, and a consequent lack of expansion. In its last eight years however, this phase was characterized by six relo- cations of franchises. These relocations ushered in the Expansion Phase in 1961. This phase lasted until 1977 and saw eight expansion franchises addedin during its first eleven years. In its last four years, however, this phase, too, had several franchise relocations -- three in a period of four years. As during the Developmental Phase these relocations signaled 67 the beginning of a third, or Neo-Expansion Phase, in 1977. This Neo-Expansion Phase is characterized by "newness" in that it is a "new" (another) expansion that is going to also have to occur in primarily "new" markets. As such, this entire phase is a direct outgrowth of the model's identifying of the fact that the general market place for major league baseball franchises in the United States is extremely limited. The model could identify only five SMSA's of the total of 276 in the United States in 1976 as possessing the potential to support a major league baseball franchise. The model, furthermore, identified only three SMSA'S as having the potential to support an initial franchise. Thus, given such a saturated market, there is only one way for future franchise loca- tion to turn in our mass society. It must develop critical mass by expanding major league baseball to international markets. The recent awarding of a major league baseball franchise to Toronto, Ontario, Canada was a reaffirmation of this fact. This was a conscious and much considered decision based upon the initial location of the Montreal Exposition in Canada in 1969, and should be seen as a harbinger of further international expansion to such areas as Mexico, Central America and Japan. The identi- fication of markets and the location of franchises outside the United States during this Neo-Expansion Phase is a must. Because of the contin- uing and dynamic advancements occuring in the mass media this seemingly futuristic possibility is upon us now. With this in mind, and the large number of franchises losing money in the United States, it would appear that a stabilization through reduction and relocation of major league baseball franchises in the United States will occur before any large-scale international expansion. When viewed from an historical perspective it seems reasonable to predict that this reduction and relocation will take 68 place over no more than a five year time frame and will be followed by a short period of stabilization before engaging in the major concern of the Neo—Expansion Phase -- vigorous and extensive international expansion. -——-_ ‘-..- -—-—-—. . APPENDICES APPENDIX A SOURCES OF DATA FOR VARIABLES IDENTIFIED IN STUDY VARIABLE Attendance Population Effective Buying Income Total Retail Sales Percent White Population Percent Population Over Age 65 Percent Population Employed in Wholesale/Retail Service Percent Population Employed in Manufacturing Percent Population Employed in Government Percent of Population with Family Income Greater Than $15,000 Percent of Population with Family Income Less Than $5,000 Median Household Income Average Household Effective Buying Income Club Won-Lost Percentage Age of Stadium Capacity of Stadium Number of Franchises Located Within SMSA 69 SOURCE OF INFORMATION The World Almanac and Book of Facts Statistical Abstract of the United States Sales Management Sales Management Statistical Abstract of the United States Sales Management Employment and Earnings, States and Areas, 1939-1974 Employment and Earnings, States and Areas, 1939-1974 Employment and Earnings, States and Areas, 1939-1974 Sales Management Sales Management Sales Management Sales Management The Encyclopedia of Baseball The World Almanac and Book of Facts The World Almanac and Book of Facts The World Almanac and Book of Facts APPENDIX B RESIDUALS FROM REGRESSION FRANCHISED SMSA'S 1970-74 SMSA YEAR Y VALUE Y ESTIMATE RESIDUALS Los Angeles 1970 1,784,527 1,944,195 -159,667 1971 1,697,122 1,988,774 -291,661 1972 2,064,594 2,004,657 59,936 1973 1,860,858 2,036,670 -175,811 1974 2,136,192 2,232,744 -96,552 New York 1970 2,175,373 2,319,664 -144,291 1971 2,697,479 2,021,097 676,382 1972 2,266,680 2,087,493 179,187 1973 2,134,185 2,258,432 -124,247 1974 1,912,390 2,015,727 —103,336 Houston 1970 1,412,995 1,154,519 258,475 1971 1,253,444 1,123,828 129,616 1972 1,261,589 1,111,812 149,776 1973 1,469,247 1,308,868 160,378 1974 1,394,004 1,176,447 217,556 Chicago 1970 1,674,993 1,772,342 -97,348 1971 1,642,705 1,591,737 50,967 1972 1,653,007 1,562,438 90,569 1973 1,299,163 1,718,568 -419,404 1974 1,351,705 1,544,676 -192,971 Atlanta 1970 1,458,320 1,289,157 169.163 1971 1,078,848 917,010 161,837 1972 1,006,320 1,025,469 -19,148 1973 752,973 855,862 ~102.889 1974 800,655 971,642 -170,987 Baltimore 1970 1,058,168 1,454,693 -396,525 1971 1,057,069 1,321,484 -264,414 1972 1,023,037 1,320,241 -297,204 1973 899,950 1,049,182 ~149,232 1974 958,667 1,258,463 -299,796 Seattle 1970 677,944 704,425 -26,481 Milwaukee 1971 933,690 713,555 220,174 1972 731,531 793,000 -61,469 1973 600,400 719,149 -118,709 1974 1,092,158 891,493 200,664 70 SMSA Minneapolis Washington D.C. Arlington St. Louis San Diego Cincinnati San Francisco Pittsburgh Philadelphia Boston Anaheim YEAR 1970 1971 1972 1973 1974 1970 1971 1972 1973 1974 1970 1971 1972 1973 1974 1970 1971 1972 1973 1974 1970 1971 1972 1973 1974 1970 1971 1972 1973 1974 1970 1971 1972 1973 1974 1970 1971 1972 1973 1974 1970 1971 1972 1973 1974 1970 1971 1972 1973 1974 71 Y VALUE 1.349.328 1.261.887 940.858 797.901 907.499 918.106 824.789 655.156 662.974 686.085 1.574.046 1.604.671 1.682.783 1.629.736 1.196.894 512.970 643.679 557.513 644.273 611.826 987.991 1.803.568 1.501.122 1.611.459 2.017.601 773.232 778.355 1.106.043 921,323 1.000.763 ' 769.369 1.341.947 1.501.132 1.427.460 1.319.913 519.414 708.248 1,511.223 1.343.329 1.475.934 1.833.246 1.595.278 1.678.732 1.441.718 1.481.002 758.388 1.077.741 926.373 744.190 1.058.206 Y ESTIMATE 1.460.118 1.382.589 950.765 1.024.013 1.076.024 ~618.537 773.126 734.387 559.676 701.017 1.188.365 1,376,635 1.275.941 1.048.985 1,123,729 658.686 715.091 703.220 701.851 702.826 1.294.778 1.598.941 1.019.738 1.432.248 1.448.413 850.223 913.140 856.153 929.433 946.271 1.380.277 1.403.421 1.530.671 1,572,264 1,175,098 1.068.576 1.299.112 1.171.285 1.075.271 1.317.857 1.377.182 1.328.314 1.335.327 1.429.921 1.488.128 841.178 1.160.309 903.971 943.890 972.413 £82019. -110.790 -120.702 -9.907 -226.111 '1683524 299.568 51.662 -79.231 103.294 -14.932 385.680 228.036 406.842 580.750 73,164 “1459716 ~71.412 ‘1459707 -57.578 91.000 -306.787 204.626 481.383 179.210 569.187 -76.991 -134,785 249.889 -8.110 54.491 -610.908 -61.473 -29.539 -144.804 144.814 -549,162 -590.863 339.938 268.058 158.077 456.063 266.963 343.405 11.797 -7,126 -82,790 -82.568 22.401 -199.700 85.792 SMSA Cleveland Detroit Kansas City YEAR 1970 1971 1972 1973 1974 1970 1971 1972 1973 1974 1970 1971 1972 1973 1974 72 Y VALUE 619.970 729.752 591.361 626.354 615.107 1.577.481 1.501.305 1.591.073 1.892.386 1.724.146 902.414 693.047 910.784 707.656 1.345.341 Y ESTIMATE 736.582 1.039.275 646.932 966.964 874.395 1.595.598 1.382.732 1.611.111 1.593.904 1.583.113 855.947 734.524 1.122.648 981.355 1.153.508 RESIDUALS -116.612 -309.522 -55.571 -340.610 -259.288 '189116 118.573 -20.037 298.482 141.033 46.466 -419477 -211.863 -273.699 191.833 APPENDIX C NON-FRANCHISED SMSA's ESTIMATED ATTENDANCE §M§A_ Tampa - St. Petersburg, Florida Newark, New Jersey Miami, Florida Tulsa, Oklahoma Greenville - Spartanburg, South Carolina Nassau - Suffolk, New York Greensboro - Winston Salem - High Point, North Carolina Allentown - Bethlehem — Easton, Pennsylvania Birmingham, Alabama Youngstown - Warren, Ohio Grand Rapids, Michigan Gary - Hammond - East Chicago, Indiana Louisville, Kentucky - Indiana Providence - Warwick - Pawtucket, Rhode Island Fort Lauderdale - Hollywood, Florida Paterson - Clifton - Passaic, New Jersey Jersey City, New Jersey Norfolk - Virginia Beach - Portsmouth, Virginia North Carolina 73 ESTIMATED ATTENDANCE 1,127,127 1,113,237 1,106,545 1,040,913 1,040,411 1,038,592 1,034,995 1,033,490 1,019,824 1,013,722 998,887 995,065 985,333 984,165 982,020 980,979 972,019 970,475 74 SMSA Nashville - Davidson, Tennessee Buffalo, New York Chattanooga, Tennessee - Georgia Springfield - Chicopee - Holyoke, Massachusetts - Connecticut Bridgeport, Connecticut Charleston - North Charleston, South Carolina Portland, Oregon - Washington Worcester, Massachusetts Indianapolis, Indiana Rochester, New York New Orleans, Louisiana Phoenix. Arizona Toledo, Ohio - Michigan Orlando, Florida Denver - Boulder, Colorado Beaumont - Port Arthur - Orange, Texas Wilmington, Delaware - New Jersey - Maryland Mobile, Alabama Flint, Michigan Memphis, Tennessee - Arkansas - Mississippi Jacksonville, Florida West Palm Beach - Boca Raton, Florida Seattle - Everett, Washington Akron, Ohio New Haven - West Haven, Connecticut Shreveport, Louisiana ESTIMATED ATTENDANCE 969,820 968,071 967,463 952,836 965,999 965,373 963,658 948,560 944,870 942,874 941,464 939,544 935,804 931,031 928,968 924,869 924,675 923,815 922,306 915,182 914,239 913,077 909,671 908,432 908,090 899,515 75 SMSA ESTIMATED ATTENDANCE Green Bay, Wisconsin Wichita, Kansas Salt Lake City - Ogden, Utah San Jose, California Dayton, Ohio Knoxville, Tennessee El Paso, Texas Omaha, Nebraska - Iowa Syracuse, New York Spokane, Washington Richmond, Virginia Riverside - San Bernardino - Ontario, California Columbus. Ohio Fresno, California Las Vegas, Nevada Washington D.C. - Maryland - Virginia San Antonio, Texas Oklahoma City, Oklahoma Albany - Schenectady - Troy, New York Raleigh - Durham, North Carolina Albuquerque, New Mexico Hartford, Connecticut Tacoma, Washington Tucson, Arizona Baton Rouge, Louisiana Honolulu, Hawaii Sacramento, California 898.263 895,125 893,733 865,886 853,742 845,143 838,425 829,720 826,644 815,602 811,449 797.393 784,559 755,728 755,711 742,710 728,266 723,285 699,442 693,686 671,897 666,097 662,031 659,340 613,015 563.751 427.819 BIBLIOGRAPHY BIBLIOGRAPHY Adelman, Irma and Morris, Cynthia. Society, Politics, and Economic Deve19pment: A Quantitative Approach. Baltimore: The John Hopkins Press, 1967. Andreano, Ralph. No Joy in Mudville--The Dilemma of Major League Baseball. Cambridge: Schenkman Publishing Company, 1965. Ball, Donald W. Sport and Social Order: Contributions to the Soci- ology of Sport. Reading, Massachusetts: Addison-Wesley, 1975. Blalock, Hubert M. Social Statistics. New York: McGraw-Hill Book Company, 1972. Block, Alex. "50, You Want to Own a Ball Club,” Forbes 119 (April, 1977): 37-40. Davenport, David. ”Collusive Competition in Major League Baseball: Its Theory and Institutional Development." American Economist (Fall, 1969): 6-30. Demmert, Henry, G. The Economics of Professional Team Sports. Lexington: Lexington Books, 1973. Durso, Joseph. The All-American Dollar: The Big Business of Sports. Boston: Houghton Mifflin, 1971. . "The Pro Sports Explosion,” Detroit Free Press, 28 July 1974, Sec. C, p. l. El-Hodiri, Mohamed, and Quirk, James. "An Economic Model of a Profes- sional Sports League.“ Journal of Political Economy. (November/ December, 1971): 1302-19. "The Economic Theory of a Professional Sports League,” Government and the Sports Business. Washington, D.C.: Brookings Institute, 1974. Ferguson, George, A. Statistical Analysis in Psychology and Education. New York: McGraw-Hill Book Company, 1966. Forbes, April, 1971. "Who Says Baseball is Like Ballet?" 76 I I i.ill.‘ ‘IIII'V ‘III‘II II 77 Hoover, Edgar. The Location of Economic Activity. New York: McGraw- Hill. 1963. . Horowitz. Ira. "Sports Broadcasting," Government and the Sports Business. Washington, D.C.: Brookings Institute, 1974. Johnson, William. "Television and Sport," Sports Illustrated. (January, 1970): 87-102. Jones, J.C. "The Economics of the National Hockey League," Canadian Journal of Economics II (February, 1969): 1-19. Koppett, Leonard. A Thinking ManfisGuide to Baseball. New York: Dutton, l967. . "The Sports Explosion," Detroit Free Press, 28 July 1974, Sec. C, p. 2. Malloy, Edward. "The Impact of Baseball on the Pittsburgh Economy." Pittsburgh Chamber of Commerce, 1977. McCarty, Harold and James Lindberg. A Preface to Economic Geography. New York: Prentice Hall, 1966. Nie, Norman; Hull, C. H.; Jenkins, J. G.; Steinbrenner, K.; and Bent, D. H. Statistical Package for the Social Sciences. New York: McGraw- Hill Book Company, 1975. Neale, Walter C. "The Peculiar Economics of Professional Sports," Quarterly Journal of Economics. (February, 1964): 1-14. Noll, Roger 6., ed. Government and the Sports Business. Washington, D.C.: Brookings Institute, 1974. Robinson, Arthur; Lindberg, James B.; and Brinkman, Leonard. "A Corre- lation and Regresiion Analysis Applied to Rural Farm Population Densities in the Great Plains," Annals of the Association of American Geographers 51 (1961): 211-221. Rothenberg, Fred. "Unacceptable Product, Financial Burden Plague Expansion,” Richmond News Leader, 28 March 1975, p. 19. Rummel, R.J. "Understanding Factor Analysis," Journal of Conflict Resolution II (December, 1967): 446-480. ‘ Sales Management Magazine, July, 1970, 1972, 1973, 1974. "Survey of Buying Power.” Scully, Gerald, W. "Economic Discrimination in Professional Sports," Law and Contemporary Problems. (Winter/Spring) 1973. "Discrimination: The Case of Baseball," Government and the Sports Business. Washington, D.C.: Brookings Institute, 1974. 78 Seymour, Paul. Baseball: The Golden Age. New York: Oxford University Press, 1971. Thomas, Edwin. "Maps of Residuals from Regression," in Brian Berry and Duane Marble (ed.). Spatial Analysis: A Reader in Statis- tical Geggraphy. Englewood Cliffs, N.J.: Prentice Hall, 1968. Treet Roger. The Encyclopedia of Baseball. New York: A.S. Barnes, 1974. United States Census Bureau. Characteristics of the Population, United States Summary. Washington, D.C.: U.S. Government Printing Office, 1970. United States Bureau of Labor. Employment and Earnings, States and Areas 1939-1974. Washington, D.C.: U.S. Government Printing Office, 1975. United States Department of Commerce. Statistical Abstract of the United States. Washington, D.C.: U.S. Government Printing Office, 1970, 1971, 1972, 1973, 1974. United States House of Representatives. Inquiry Into Professional Sports. Washington, D.C.: U.S. Government Printing Office, 1977. The World Almanac and Book of Facts, Newspaper Enterprise Association, 1970, 1971. 1972, 1973, 1974. MICHIG IIIIIIIIIIIII 31 29 UNI IIIIIIIIIIII @808 L IIIIIIIIII 446 IES