RACE AS A BASIS FOR MARKET SEGMENTATION: AN EXPLORATORY ANALYSIS Dissertation for the Degree of Ph. D. MICHIGAN STATE UNIVERSITY ANDREW ALLEN BROGOWICZ 1977 This is to certify that the thesis entitled RACE AS A BASIS FOR MARKET SEGMENTATION: AN EXPLORATORY ANALYSIS presented by Andrew Allen Brogowicz has been accepted towards fulfillment of the requirements for Ph.D . degree in Marketing Egg/1W2? ‘ Major professor Date February 25, 1977 0-7639 I Ting 1 9 10°" W b": 071 rm? .1 V Data .3. \ 1“ ”NUS d: {(0594 .‘ .J\ A N Hilts ")N-,:u."- Wen-u mane: wrun ' I I to ‘9 ab"..e hr 7“”, w;l:~ :- p lL A‘inor Mar-IN. .t‘ “.v: T; A‘n Nick um white gunman-t. \ ., fie ~‘. ‘ ‘ 907-:- m "N” ObjeCf1V" 's.’ {L ‘ ~-'.L'..) “1‘ ; P‘. . ‘ v‘m.s . 5.5.» ’3' DH... L a? A}; ~ rch hypothesis was Ur: (ace .9 ,I. v ; -~-'\tf" 1"} r." . . ‘ ‘. ”7"“- 'v ~ 'S's's fur I I I.» ~ “0" "I" Study 6'4 N" ’"L‘U'S naivnwe 3 or ’14 1,5; 1 n‘. yS‘H “3" ;;W the effectiveness {11 race ‘r column“ ' ‘ for market semnutiou tnc'ome. 5mm (that race If . '. _ J'x’ , 'Asf‘al class. race :tra’iflod by nothing.“ p straw. '. ,9 ‘\ ABSTRACT RACE AS A BASIS FOR MARKET SEGMENTATION: AN EXPLORATORY ANALYSIS By Andrew Allen Brogowicz Data published by a variety of sources have indicated that numerous differences exist between the consumption patterns of black and white consumers. Such data have been interpreted by some marketers as conclusive proof that a separate and distinct "black-consumer market" exists. However, in reviewing the marketing literature, this writer found a lack of evidence to show that black consumers comprise a homogeneous market segment that is clearly differentiated from other, presumably white, market segments. 0n the contrary, black consumers appeared to be quite heterogeneous in their needs, preferences, and buying behavior. Moreover, the similarities in consumption patterns between black and white consumers often tended to outweigh the differ— ences. The main objective of this study was to explore the effective- ness of ragg_as a basis for market segmentation. Here, the author's guiding research hypothesis was that race is ngt_an effective basis for market segmentation. The study did not focus exclusively on race but instead analyzed the effectiveness of race in comparison with five alternative bases for market segmentation: income, social class, race stratified by social class, race stratified by motivation to strive, Andrew Allen Brogowicz and benefits sought. The study focused (a) on product-choice behavior involving dif- ferent types of alcoholic beverages and (b) on brand—choice behavior for two types of beverage, beer and Scotch whiskey. The sample used in the study consisted of black and white faculty and administrators at Michigan State University and black and white residents of Lansing's Model Cities Area. A mail questionnaire was used to obtain data from the MSU faculty and administrators, while professional interviewers were employed to conduct personal interviews with the Model Cities residents. Based on the data that were obtained, the respondents were classified into six alternative sets of market segments. Then, a series of two-part statistical hypothesis tests were performed, including both the chi-square test of independence of classification and Goodman and Kruskal's lambda, a measure of predictive association. The cumulative resultSIofthe statistical hypothesis tests were used to determine the effectiveness of each of the six alternative bases for market segmen- tation. To weigh all the evidence objectively, the author employed a set of decision rules which took into accountthenature of the statis- tical hypotheses, the number of significant chi-square values, and the number of lambdas that were significantly greater than zero. The results of the study supported the researcher's guiding hypothesis that race is not an effective basis for market segmentation. Race stood out as being clearly less effective than any of the other five alternatives for segmenting the market for alcoholic beverages. In addition, race was less effective than all but one other alternative fur segmenting the market for beer and malt liquor. Finally, the Andrew Allen Brogowicz segment of the study dealing with Scotch whiskey had to be abandoned due to an inadequate number of Scotch drinkers in the sample. Never- theless, there was some evidence to suggest that race was not as effec- tive as some of the other alternatives for segmenting the market for Scotch whiskey. RACE AS A BASIS FOR MARKET SEGMENTATION: AN EXPLORATORY ANALYSIS By Andrew Allen Brogowicz A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Marketing and Transportation Administration 1977 _» ,‘s- can. at © Copyright by AgDREH ALLEN BROGOWICZ 1 77‘ ‘7 I am Itevuh and wager . I‘llhfr". - (I I If have M" v R. Bacon. -, ion of n :If't' .~ umlerstanm " M I- Eric: prettng M» This dissertation is dedicated "to my parents In mast:.:~ (”Wife “We and encouragement '2‘“ fl Profuse: s.-;..»I‘?.IPR" maketh‘is P955919 I f" couple-,9: c. 4-,“, q. r, - . t, of Florida uh.) 7r 1 'n: ., of the s:.r:,t 'n -*-s um» «v - : . _, ImTfl ,'"‘1' ”gm: 1:. may...” , g . _. .-‘ f programming for 171? ”mire c 'c «.2;- .. . =7: ~ «aw mo Tim! eraft of tr .e. sag-«Irma. “fuming Inc-HM S amused 385 17!? vet, ring -2’ wife"— ‘SIIW L Earthy of {1c :sed beveragx- }am:9q.pe¢~. fir. I. of the Mtioml {Seer Wholesaler! fisfiaigtt'fia of h) C. sea of the (new States 8mm Wis-u: 1;, ' ”Jaylor of ates M. ' 3h _.r ‘53,. ~.-_-- in" - ‘. vh ~m } - ‘ R" ', r: R -.- ‘ I - .n- , ‘Iz. ACKNOWLEDGMENTS I am deeply indebted to Professor E. Jerome McCarthy for his guidance and support over the past several years. He has had a pro- found influence on my graduate education, and his efforts on my behalf have far exceeded his role as major professor. Professor Frank R. Bacon, Jr., has provided much needed perspective on the direction of this study and, on a broader scale, contributed greatly to my understanding and appreciation of applied research. Professor Leighton A. Price has been particularly helpful in selecting and interpreting the computer programs used in the study. In addition to my guidance committee, special thanks must be given to Professor Donald A. Taylor who provided a generous amount of funding for computer services; to Professor William L. Nilkie of the University of Florida who provided valuable counsel during the pro- posal stage of the study; to Ms. Anne Neiberding who assisted me in using the ISODATA program; to Mr. Richard Hewens who performed all of the computer programming for the study; and to Ms. Nancy Robinson who typed the final draft of this manuscript. The following individuals provided me with very helpful infor- mation: Dr. Steve L. Barsby of Licensed Beverage Industries; Mr. Russell G. Hopkins of the National Beer Nholesalers' Association of America; Mr. Philip C. Katz of the United States Brewers Association; and Mr. Thayer C. Taylor of Sajes Management. iii I would also like to pay special tribute to my friend and coun- selor the late Mr. John L. Roberts of Wayne State University who was a constant source of strength and encouragement during the earlier years of my marketing education. Last of all, and most of all, I want to thank my wife Patricia who has read, typed, and proofread countless rough drafts of this thesis and who has somehow managed to remain the beautiful person she is despite all the many sacrifices she has had to endure during the course of this project. TABLE OF CONTENTS LIST OF TABLES ......................... viii LIST OF FIGURES ........................ ix Chapter I. INTRODUCTION ....................... 1 Overview of the Problem ................ 1 Historical and Economic Perspective ......... 2 Is Race an Effective Basis for Market Segmentation? 5 Arguments in support of a separate black—consumer market ...................... 6 Arguments against a separate black-consumer market . 8 Why the issue remains unresolved .......... 12 Objectives and Scope of the Study ........... 14 Organization of This Report .............. 15 II. LITERATURE REVIEW .................... 17 Observed Differences and Similarities in Black-White Consumption Patterns ................. 18 General Expenditure Patterns ............. 18 Product Usage and Brand-Choice Behavior ....... 25 Automobiles .................... 26 Appliances ..................... 28 Clothing ...................... 33 Packaged goods ................... 35 Alcoholic beverages ................ 40 Store Shopping Behavior ............... 42 Where blacks shop ................. 43 Factors influencing store selection ........ 45 Explaining Observed Variations in Black-White Consumption Patterns ................. 47 Income ........................ 49 Social Class ..................... 51 Motivation ...................... 52 The Black Dilemma .................. 54 Conclusion ...................... 56 Summary ........................ 58 v III. RESEARCH PERSPECTIVE .................. 59 Race as a Segmentation Variable: What the Literature Reveals ....................... 59 Is a New Research Perspective Needed? ......... 61 Behaviorally Oriented Versus Decision- Oriented Research ................. 63 A Priori Assumptions about Race ........... 65 Research Implications ................ 67 In Search of an Alternative to A Priori Racial Segmentation .................... 69 The Concept of Market Segmentation ......... 69 Application of the Concept ............. 71 Different approaches to segmentation ....... 72 Evaluating the different approaches ........ 75 Research Implications ................ 79 Is Benefit Segmentation the Answer? ......... 80 People buy benefits, not products ......... 80 Do blacks and whites seek different benefits? . . . 84 Summary ........................ 85 IV. RESEARCH FRAMEWORK AND METHODOLOGY ........... 86 Statement of the Problem ............... 86 Objectives and Scope of the Study ........... 86 Guiding and General Research Hypotheses ........ 89 Research Design .................... 90 Products to Be Studied ............... 90 Research Hypotheses Restated ............ 92 Statistical Hypotheses ............... 92 Alcoholic beverages ................ 93 Beer and malt liquor ............... 94 Scotch whiskey .................. 95 Sample Selection .................. 96 Data Collection ................... 100 Planning ..................... lOO Implementation .................. 103 Response rates .................. 105 Plan of Analysis .................. 110 Forming segments ................. 110 Testing segments ................. 123 Evaluating bases for segmentation ......... 127 Comparing bases for segmentation ......... 131 V. RESULTS OF THE STUDY .................. 132 Alcoholic Beverages .................. 132 Description of Segments ............... 132 Striver/nonstriver segments ............ 132 Benefit segments ................. 134 Findings ...................... 137 Evaluating bases for segmentation ......... 141 Comparing bases for segmentation ......... 141 vi c’ Beer and Malt Liquor ................. Description of Segments ............... ‘Findings ...................... Evaluating bases for segmentation ......... Comparing bases for segmentation ......... Scotch Whiskey .................... Description of Segments ............... Findings ...................... VI. SUMMARY, LIMITATIONS, CONCLUSIONS, AND IMPLICATIONS FOR FUTURE RESEARCH ................... Summary ........................ Objectives of the Study ............... Methodology ..................... Results ....................... Alcoholic beverages ................ Beer and malt liquor ............... Scotch whiskey .................. Limitations of the Study ............... Conclusions . . . . .................. Implications for Future Research ........... Appendix A. DEMOGRAPHIC PROFILE OF THE U.S. BLACK POPULATION Population ...................... Geographic Location .................. Education ...................... Employment ...................... Income ........................ B. SUMMARY OF STRIVER AND NONSTRIVER RESPONSES TO STRIVING INDEX BY RACE ........ . Calculating Lambda .................. Testing the Significance of Lambda .......... D. MAIL QUESTIONNAIRE AND COVER LETTER ........... E. PERSONAL—INTERVIEW QUESTIONNAIRE AND LETTER OF INTRODUCTION ...................... LIST OF REFERENCES ...................... 143 143 150 ISO 152 152 158 160 160 160 161 163 163 163 163 164 165 167 170 170 171 174 175 180 182 182 186 188 195 208 LIST OF TABLES II—l. Black Vs. White Distribution of Family Expenditures for Current Consumption: 1960-61 .......... 21 II-2. Summary Comparisons of Black-White Expenditure Differences (Per Household) ............. 22 II—3. Percentage Change in Selected Consumption Expenditures in Response to Percentage Change in After-Tax Income for Black and White Families: 1950 and 1960-61 . . . 24 II-4. Percentage of Black and White Households Possessing Selected Appliances: 1970 .............. 30 II-5. Average Black and White Expenditures Per Household on Cars and Selected New Household Durables: 1968-72 . . 31 II—6. Percentage of Blacks and Whites Who Had Recently Purchased or Who Owned Various Household Products. . . 37 IV-l. Summary of Types of Data Collected for Study ...... 101 IV-2. Response Rate for Mail Questionnaires ......... 106 IV—3. Response Rate for Personal Interviews ......... 107 IV—4. Demographic Characteristics of Subsamples by Race (in Percentages) ................... 108 IV-5 Potential Benefits Sought by Product Category ..... 117 V-l. Distribution of Striving Scores ............ 134 V-2 Summary of Statistical Hypothesis Tests for Alcoholic Beverages ...................... 139 V-3. Summary of Research Hypothesis Tests for Alcoholic Beverages ...................... 142 V-4. Summary of Statistical Hypothesis Tests for Beer and Malt Liquor ...................... 149 V-5. Summary of Research Hypothesis Tests for Beer and Malt Liquor ..................... 151 A-l. Percentage Distribution of the Black Population by Region, and Blacks as a Percentage of the Total Population in Each Region: 1950, 1960, and 1970 . . . 172 A-2. Distribution of Families by Income: 1965, 1969, and and 1973 ....................... 176 A-3. Median Family Income in 1973, and Black Median Family Income as a Percentage of White, by Region: 1967, 1969, 1971, and 1973 (in Current Dollars) ...... 178 8-1. Summary of Striver and Nonstriver Responses by Race (in Percentages) ................... 180 C-1. Hypothetical Contingency Table ............. 183 C-2. Alternate Contingency Table .............. 186 viii III- 1. IV- 1. V- < < I I 4:- w —l a LIST OF FIGURES A Continuum of Consumer Characteristics ...... Summary of Plan of Analysis ........... . . Alternative Market Grids for Alcoholic Beverage Market (Number and Percentage of Respondents in Parentheses) .................... Market Grid Based on Race Stratified by Motivation to Strive (Number and Percentage of Respondents in Parentheses, N=176 ................ Benefit-Importance Scales by Cluster: Alcoholic Beverages ..................... Market Grid Based on Benefits Sought for Alcoholic Beverages and Most Important Benefits by Segment (Number and Percentage of Respondents in Paren- theses, N=276) .................. Alternative Market Grids for Beer and Malt Liquor Market (Number and Percentage of Respondents in Parentheses) .................... Benefit-Importance Scales by Cluster: Beer and Malt Liquor ....................... Market Grid Based on Benefits Sought for Beer and Malt Liquor and Most Important Benefits by Segment (Number and Percentage of Respondents in Paren- theses, N=24l) ................... Alternative Market Grids for Scotch Whiskey Market (Number and Percentage of Respondents in Paren- theses) ...................... Benefit-Importance Scales by Cluster: Scotch Whiskey Market Grid Based on Benefits Sought for Scotch Whiskey and Most Important Benefits by Segment (Number and Percentage of Respondents in Paren- theses, N=153) ................... 82 111 133 135 136 138 144 147 148 153 156 157 CHAPTER I INTRODUCTION Overview of the Problem In recent years, many marketing researchers have tended to dis- count the value of demographic variables as bases for market segmenta- tion due to their typically low degree of correlation with actual consumption behavior. One notable exception to this trend has been the increased attention given to ra§g_as a segmentation variable, expecially with regard to America's black population.1 Marketing academicians and practitioners have devoted countless articles and even entire books to the topic of black consumption behavior, while trade journals have literally abounded with references to the "black travel market," the “black leisure market," the “black clothing mar- ket," and so forth. Indeed, the question "Is there really a black— consumer market?" has been debated so often in the literature that marketers might well be accused of having contributed to the identity crisis which allegedly afflicts blacks in America! 1The term "black" is used exclusively in the text and tables of this study, except that the terms "black" and "Negro" are used interchangeably when quoting from other sources. The terms "Negro and other races" and "nonwhites" describe persons of all races other than white and are used whenever data for blacks alone are not available. Such data generally reflect the condition of the black population, since blacks make up about 90 percent of the population of "Negro and other races." -30..- 2 Obviously, nobody is really questioning the existence of large numbers of black consumers who, in the aggregate, purchase huge amounts of goods and services each year. This would be sheer folly in a nation where some 24 million blacks account for at least 11 percent of the total population. The question many marketers arg_asking is whether the needs and buying behavior of black consumers are so different from those of other consumers that black consumers should be viewed as com- prising a separate and distinct market segment for purposes of market- ing strategy planning (i.e., for purposes of selecting target markets and developing appropriate marketing mixes). Some historical and economic perspective may be necessary to help the reader understand why marketers have shown this growing interest in the possible existence of a black-consumer market. After all, the U.S. population is made up of many different racial and ethnic groups. Yet, one seldom encounters any reference to the ~"Italian-American market," the “Polish-American market," and so on and so forth. In particular, one does not hear mention of a "white- consumer market,“ and indeed, no competent market analyst would ever dream of lumping together all white consumers into a single market segment. Historical and Economic Perspective Interest in black consumption behavior is not a very recent phenomenon in the United States. As Bauer and Cunningham (1970b, p. 7) have reported, several black—related marketing topics in vogue today were already being discussed as far back as 1931. These included: . . . The use of Negroes to sell to Negroes, the need to display concern for the welfare of the Negro, the differential product 3 usage of Negroes, choice of media, the interest of Negroes in personal-care products, and the amount of their spending power. Nevertheless, interest in black spending habits did not really become widespread until the past decade or so, when mounting social pressure-- together with a search for new profit opportunities--led marketers to focus greater attention on the needs and wants of black consumers. The mounting social pressure stemmed from the social revolution that rocked the nation during the turbulent sixties. The civil rights movement and the Johnson Administration's "war on poverty," together with provocative books by writers such as Galbraith (1958), Harrington (1963), and Caplovitz (1967), drew considerable attention to the plight of poor and disadvantaged consumers. What really captured the atten- tion of marketers, however, were the destructive urban riots (or "con- sumer revolts," as some have called them) that reached their peak during the long, hot summer of 1967. Suddenly, many marketers began to show concern over the plight of the more visible, and perhaps more volatile, black consumer. This concern was heightened by the growing strength of a ”Black Power" movement which advocated economic self- sufficiency and, accordingly, boycotts of white—owned businesses.2 2In general, however, this concern proved to be somewhat tempor- ary in nature. As the riots subsided, the nation gradually turned its attention away from the ghetto toward other problems which had arisen. Some of these problems, the ecology movement for example, were regarded as irrelevant by many blacks who feared that public concern for these problems would divert resources away from the problems of poverty and racism (Business_ngk, 14 November 1970, p. 49). They may have been right to some extent, for according to Vernon Jordan, Executive Direc- tor of the National Urban League, "the condition of black Americans, once the benchmark of America's commitment to equality and justice, is now the object of malign neglect and hostile disregard“ (Jarrett, 1976; sec. 2, p. 6). In fact, some futurists have warned of the possibility of internal confrontations and further polarization along racial and economic lines if large numbers of blacks continue to be frustrated in their social and economic aspirations (Wheeler, 1974). 4 It would be grossly unfair, however, to Suggest that the recent interest marketers have shown in the needs and wants of black consumers can be attributed solely to social pressure and coercion. Certainly, many marketers reacted with something more than just a feeling of social responsibility when Gibson (1969) pointed them toward a "$30 billion Negro market" which he said represented a comparatively untapped source of sales and profits. Today, alert marketers recognize that although the black population is composed of a disproportionate number of people who are poor, undereducated, and unemployed, its over- all purchasing power has risen Substantially from a scant $1.6 billion in 1931 (Sglgs Management) to an estimated $59 billion in 1975 (AQXEE; tising Age, 14 July 1975, p. 34)3. In fact, according to the U.S. Department of Commerce (1975, p. 26), America's nonwhite consumers now comprise the ninth largest consumer market in the world and are expected to purchase almost $120 billion worth of goods and services in 1980. The National Industrial Conference Board has estimated that while nonwhite consumers comprise about 11.5 percent of the U.S. popu- lation, they account for a disproportionately low 8.5 percent of total consumption expenditures (Linden, 1967). However, nonwhite expendi- tures tend to be disproportionately high_for selected categories of goods and services. For example, Gibson (1969, p. 10) tantalized mar- keters with statistics which indicated that: . Negroes, as 11 percent of the total population in the United States, consume over 50 percent of the Scotch whisky imported into nation, consume more than 70 percent of the entire output of the 3A brief demographic profile of America's black population is contained in Appendix A. . P’T' 5 Maine sardine industry, consume more than 49 percent of all the grape soda produced in America, spend 23 percent more for shoes than does the majority white population, and spend up to 23 per- cent more for food sold in Supermarkets to be consumed at home. While some have questioned the accuracy of Gibson's statistics, he did succeed in drawing attention to a group of customers whose needs and wants previously had not been well understood and thus had gone largely ignored. Is Race an Effective Basis for Market Segmentation? To say that a black-consumer market exists for purposes of marketing strategy planning is to assume, in turn, that race is an effective and perhaps key variable for segmenting markets. Because this study is mainly concerned with the validity of this assumption, it is important at the outset to define the term "market segmentation" as it will be used in the study. Market segmentation is the process of isolating smaller, more homogeneous market segments within a [heterogeneous] market for the purpose of selecting one or more target markets and developing a unique marketing mix to satisfy the needs of each (McCarthy, 1975, p. 63). Market segmentation is an extremely vital process, as witnessed by a recent survey conducted by Booz, Allen, and Hamilton which indi- cated that "where companies go astray most frequently in launching new products is in their failure to define the markets for them (Philadel- phia Inguirer, 6 January 1974; sec. C, p. 9). Unfortunately, market segmentation is more of an art than a science, and markets can be seg- mented in an almost infinite number of ways. Human behavior being com- plex and multidimensional, _gy_basis for segmentation is likely to generate a set of market segments for which one can observe some 6 consumption differences between segments and sgmg consumption similar- ities within segments. However, marketing strategy planners are interested only in segments which exhibit both a high degree of homo- geneity of demand within segments and a high degree of heterogeneity of demand between segments, thereby facilitating the development of effective marketing mixes. Returning to the issue at hand, a marketing strategy planner must determine if black consumers tend to exhibit similar needs and buying behavior for his firm's product-market area and whether their needs and behavior are, in turn, differentiated from those of other, mainly white segments. Then he must decide whether his firm is capable of providing a need-satisfying marketing mix for this potential target market-~at a profit, of course. Also, he must consider what his firm's competitors might be doing. Obviously, the paramount question is whether the strategy plan- ner can assume that race is an effective basis for market segmentation and, therefore, that black consumers constitute a separate and distinct market segment. In reviewing the literature pertaining to this topic, the consensus of opinion appears to be that, yes, a black-consumer market gggs exist. However, this opinion is far from unanimous. Therefore, arguments on both sides of the issue must be given consider- ation. Arguments in sgpport of a separate b ack-consumer market Gibson (1969, p. 9) cites “four reasons why the Negro market exists": I. forced identification of the people comprising this market 7 . definable purchase patterns by this group of consumers . the size of this market . the [concentrated] location of this market within the United States th According to Gibson, every aspect of a black American's life, including his or her buying behavior, is influenced either consciously or uncon- sciously by an omnipresent "racial reaction" which reflects a legacy of slavery and racial discrimination. Thus he advises firms to seek to create a "favorable racial reaction" among black consumers by providing them with recognition, identification, and an invitation to buy (pp. l3-21). Tuesday publisher and editor Leonard Evans also believes that a black-consumer market exists, but not entirely for the same reasons as Gibson. Evans argues that black and white consumers have different needs and motivations which have nothing to do with civil rights but instead can be traced to their divergent backgrounds and "completely opposite economic histories" (Media/Scope, 1967, p. 70). Regardless of whose views are more accurate, there is a pleth- ora of documented evidence to Show that black consumption expenditure patterns frequently differ from those of white consumers. This tends to hold true even when the much lower income distribution of black families is taken into account. Although the actual cause of these differences remains very much open to debate, many marketers have interpreted these differences to mean that a black-consumer market does in fact exist. Thus, in recent years one has witnessed more and more finns trying to attract black customers through the use of advertising gimmicks suchas those which black adman Thomas Burrell has labeled: 1. Ihg_Black Face A roagh; Assumes that the only thing that makes lac s and whites different from each other is color. 8 Therefore, reaching the black consumer effectively is merely a matter of replacing the white faces in an ad with black ones. 2. The "We Got Rhythm" Approach: Assumes that, irrelevant mes- sages and inappropriate strategy notwithstanding, blacks have an innate tendency to respond, like Pavlov's dog, to a funky beat. . . . [Therefore,] a special rhythm-and-blues version of the general market jingle will have the black consumer hustling, bumping, and boogalooing his way straight to the advertiser's product. 3. The Slang—Slinging Approach: Assumes that no matter what you say, no matter what you're talking about, just say it "cool, man" and the boss brothers and sisters will dig your rap, make the store scene, lay down some heavy bread, cop your goods and get down. 4. The Superstar Testimonial Approach: Assumes that blacks are so high on worshipping their athletes and entertainers, they will loyally emulate whatever they do or whatever they say they do (Cappo, 1975, p. 38). While Burrell does not consider these approaches to be inherently wrong, he believes that selling to black consumers requires greater empathy and more comprehensive marketing strategy planning than most firms have displayed thus far. Arguments against a separate black-consumer market Despite empirical evidence of black-white consumption differ- ences, not all marketers are fully convinced that a separate and dis- tinct black-consumer market really exists. There are several reasons for this. First, the most adament apostles of this doctrine of racially segmented markets have been the black media, whose obvious self-interests tend to cast a shadow of suspicion over what they preach. Second, systematic research has been the exception rather than the rule in studies of black buying behavior. Third, while it is true that black-white consumption patterns tend to be different for many 9 products, it is likewise true that black-white consumption patterns tend to be similar for many other products. Fourth, more than a few researchers believe that so—called black-white consumption differences may be due more to factors such as income or social class than to race. Fifth, as alluded to earlier, it is difficult to defend the logic of lumping together some 24 million blacks into a single homogeneous mar— ket segment while at the same time emphasizing the heterogeneity of the white population. Perhaps the main reason some marketers remain skeptical about the existence of a black-consumer market is that they are uncertain and confused about how to effectively cultivate such a market. In part, this is because researchers have been noticeably inconsistent in describing how blacks react to various marketing efforts. For example, black consumers have been described as being high in brand loyalty, and at the same time, as extremely price-oriented. They have been said to prefer both expensive "prestige brands" and inexpensive dealer brands. Blacks have been called "trend-setters," while also being described as slow to adopt new products. They have been said to regard integrated advertising as “tokenism,” but also to react favorably to the use of black models in ads. And although blacks have been described as less mobile than white shoppers, it has also been said that they frequently leave their local area to seek out higher quality goods at competitive prices. Along with conflicting stereotypes of black buying behavior, marketers have been provided with contradictory advice concerning how to treat black consumers. Consider, for example. an article published in Grocery Mfr. (1967) which, on the premise that "for most national 10 companies, the Negro market has been disguised as part of the mass market,“ offered marketers the following advice: Negro purchasing habits are sorely misunderstood. Traditional assumptions and stereotyped approaches must be discarded in favor of a sounder approach based on market research. Few companies have reached their attainable potential in this market. Many opportun- ities have been overlooked (p. 50). In this same article, however, a specialist in "ethnic marketing“ cautioned that: . The most important consideration is to determine the decisive versus the nondecisive way to segment the total market. To slice into Negro and white portions may not be decisive—- regional, income, age or other factors may be better guidelines (p. 52). And still another specialist warned that "many professional marketing men tend to overemphasize [italics mine] the differences between Negro and white markets“ (p. 52). To add to the confusion, while many reputed experts have recom- mended offering blacks a distinctive marketing mix, Bullock (1961a, p. 89) has called for "the creation of an 'integrated' marketing pro- gram that appeals equally to blacks and whites." Moreover, the impor- tance of using the black media to reach potential black customers (Oladipupo, 1970) has been disputed by some admen who feel that the black media are unnecessary and less efficient for this purpose (Adver— . tising Age, 3 April 1972; 10 April 1972). At the same time, however, Wall (1969, p. 25) has argued that while community relations has only a minor influence on innediate sales to most market segments, A company which advertises in Negro media, contributes to the United Negro College Fund, and employs Negroes is perceived as being concerned with the welfare of Negroes, and therefore is entitled to special concern and patronage. A final reason for questioning (or perhaps ignoring) the 11 existence of a separate and distinct black—consumer market relates to the fact that the social and economic status of America's black popu- lation is an extremely sensitive and volatile issue. Apparently, many marketers fear that misguided efforts to woo black consumers could easily backfire and create an adverse reaction toward their firm and its products. In particular, they are wary of treating black consumers as a "segregated market" at a time when black Americans are striving to achieve equality in all aspects of their daily lives. This is despite Gibson's assurances that "customer-oriented programs aimed at Negro customers are not segregation in reverse but simply provide the Negro with what he wants--recognition” (1969, p. 12). The following example concerning United Airlines suggests that some black consumers may indeed be prone to react negatively to differ— entiated marketing efforts. Back in the early 19705, United attempted to zero in on the fast-growing "black travel market" by setting up "Something Special" travel information desks at several major airports. These desks were staffed by black employees and intended to provide special assistance to the relatively inexperienced black air traveler. Thousands of letters were sent out to potential black travelers informing them of this special service and urging them to take advan- tage of it. Apparently, many did just that, as United later reported that the "Something Special" program not only generated additional sales revenue but also enhanced the airline's rapport with blacks (Wall Street Journal, 10 August 1972, p. 1). However a number of civil rights groups, including the Urban League, apparently felt that "special" and "segregated" mean about the same thing when applied to blacks, and thus they vehemently protested that the program represented 12 just another disguised form of racial discrimination (Peters, 1972). This example is no doubt atypical, and some might argue that the more controversial aspects of the "Something Special“ program could have been anticipated and avoided. Nevertheless, United's experience does illustrate the subtle complexity of trying to woo the black con— sumer. Furthermore, as in the case of the so-called "black exploita- tion films” (Jarrett, 1974), it underscores the fact that black con- sumers, like their white counterparts, are not likely to respond in a uniform fashion to any particular marketing effort. Why the issue remains unresolved From the preceding discussion, it should be obvious that the possible existence of a black-consumer market remains very much an unresolved issue. The primary reason for this is that black consump- tion behavior has not been researched in a manner which would allow one to state unequivocally whether race is or is not an effective basis for market segmentation. Many researchers have assumed a priori that race is a relevant variable for segmenting demand among black and white consumers--without subjecting this assumption to a systematic empirical test. Thus, many studies have focused exclusively on black consumers, which not only assumes that blacks are different from whites but also fails to provide any basis for comparison. Other studies have involved aggregate com- parisons between black and white consumer groups, ignoring any possible variations in conSumption patterns within each group. Of course, not all researchers have been oblivious to the pos- sibility that black consumers may be quite heterogeneous in terms of -S—V—v 1"?" 13 their needs and buying behavior. For instance, Barry and Harvey (1974, p. 53) have hypothesized that black consumers can be divided into four basic segments: 1. The Ne re se ment--strives to emulate the white middle class 2. The bIack segment--disregards the values of the white society and 15 in the process of evolving its own set of cultural standards 3. The Afro-American segment-—has not only discarded white middle- class standards but may also be attempting to destroy the established standards of white supremacy 4. The recent black immigrant segment—-identifies closely with the society from which it emigrated However, only a handful of researchers have actually attempted to oper— ationalize this concept of black consumer heterogeneity in their research designs. Those who have tried to do so have indeed found differences in consumption patterns between different groups of black consumers. But here again, the tendency has been to assume without proof that these black market segments are separate and distinct from any existing white market segments. ’ As the situation now stands, when confronted with the informa- tion, say, that blacks consume proportionately more “widgets” than white consumers, the marketing strategy planner must ask himself whether this statistical comparison suggests the most effective way of segmenting the market for widgets. Undoubtedly, ggmg consumption dif- ferences would also be observed if the overall market for widgets were segmented on the basis of some characteristic other than race—~such as sex, age, religion, life style, income, social class, personality, geo- graphite location, brand loyalty, product usage rate, ad infinitum. Maybe one of these alternative bases for segmentation would provide a better explanation of variations in the demand for widgets and, conse- quently, would be more useful for selecting target markets and 14 developing marketing mixes. Perhaps the similarities in the way blacks and whites consume widgets might outweigh the differences. In fact, it is conceivable that there may be as much, or more, heterogeneity of demand among black consumers as there exists between black and white consumers! If 50, race clearly would not be an effective variable for segmenting the widget market. Objectives and Scope of the Study The objective of this study is to explore the effectiveness of race as a basis for market segmentation. However, for reasons dis- cussed in the above paragraph, the study will not focus exclusively on race but will also take into consideration five alternative bases for market segmentation, each of which offers a theoretically plausible alternative explanation for empirical differences and similarities in black-white consumption patterns. The five alternative bases for market segmentation are: Income Social class Race stratified by social class Race stratified by motivation to strive Benefits sought m-wao-I The effectiveness of race as a segmentation variable may depend on the nature of the product-market situation. Therefore, in order to ensure that race-related differences will have a maximum chance to I appear in the research data, this study will focus on alcoholic beverages, a product class for which substantial differences in black- white consumption patterns have long been reported. Of particular importante will be product-choice decisions involving the selection of One type of alcoholic beverage over another (e.g., beer vs. gin), and 3:5. :zI-j.’ . Lflt. s; 15 brand-choice decisions for two types of alcoholic beverages--beer and Scotch whiskey. Data for the study were obtained through a survey of black and white residents of Lansing's Model Cities Neighborhood, as well as black and white members of the Michigan State University facul- ty and staff. For purposes of the study, the two subsamples will be classified as "disadvantaged consumers" and "affluent consumers,” respectively. Qgganization of This Report The introductory chapter has stressed the importance of deter- mining the effectiveness of race as a segmentation variable from the viewpoint of the marketing strategy planner. Also discussed were the objectives and scope of the study. Chapter 11 ("Literature Review“) presents a lengthy literature review of studies pertaining to black consumption behavior. Here, the intent is not so much to summarize everything that is known about black consumers, but rather to examine how researchers have studied black consumption behavior and what their findings imply about the effective- ness of race as a basis for market segmentation. Chapter 111 ("Research Perspective") attempts to synthesize the research findings presented in Chapter II to indicate what they do and do ggt reveal concerning the effectiveness of race as a segmentation variable. The need for a new research perspective is discussed, taking into account the different objectives and evaluative criteria of the behaviorally oriented and decision-oriented schools of segmentation research, as well as the current "state of the art" in market segmen- tation. 16 Chapter IV("Research Framework and Methodology”) discusses the author's research framework and the methodology to be employed in the study. A brief statement of the problem to be researched is presented and the objectives and scope of the study are summarized. The author's guiding and research hypotheses are presented, followed by a discussion of the research design to be implemented in testing these hypotheses. The selection of the products to be studied is explained, and product- specific research and statistical hypotheses are listed. This is fol- lowed by a section on sample selection and data collection. Finally, a plan for analyzing and evaluating the data is outlined, including the use of the ISODATA cluster—analysis algorithm to identify benefit seg- ments and the use of both the chi-square test and Goodman and Kruskal's lambda statistic as hypothesis testing procedures. Chapter V ("Results of the Study") summarizes and discusses the author's research findings concerning the relative effectiveness of race for segmenting the market for (a) alcoholic beverages, (b) beer and malt liquor, and (c) Scotch whiskey. Chapter VI ("Summary, Limitations, Conclusions, and Implica- tions for Future Research") does as its title suggests. The objectives, limitations, and results of the study are summarized. Conclusions are reached as to the effectiveness of race as a segmentation variable, and suggestions for future research are presented. CHAPTER II LITERATURE REVIEW Although widespread interest in black consumption behavior is a fairly recent phenomenon, this interest has already manifested itself in the form of a rather substantial body of literature. This chapter will summarize and review some of the more important research findings in this area.1 The purpose of this literature review is not so much to rehash everything that has been learned about black consumers, but rather to determine how much light these findings have shed on the essential question raised in Chapter I: Is race an effective basis for market segmentation? The reader is cautioned to bear in mind that some of the studies that will be reviewed are quite old, and thus their findings may no longer remain valid given the myriad of social, econom— ic, and political changes blacks have experienced in recent years. The first part of the chapter focuses on observed differences and similarities between black and white consumers in terms of (a) gen- eral expenditure patterns, (b) product usage and brand-choice behavior, and (c) store shopping behavior. The next part examines studies which seek to explain yhy_black-white consumption differences occur and what strategic implications such differences may have for marketers. 1For additional surveys of the literature, see Bauer and Cunningham (1970a), Joyce and Govoni (1971), Sexton (1972), Alexis and Smith (1973), and Sturdivant (1973). 17 ._...—--—~.-. 4 18 Observed Differences and Similarities in Black-White Consumption Patterns General Expenditure Patterns Comparative studies of black-white consumption patterns date back at least to Edwards' The Southern Urban Negro as a Consumer (1932), an empirical study which sought to refute a number of widely held fal- lacies concerning black spending habits. Edwards' pioneering effort was followed by a number of other comparative studies, with those by Sterner (1943) and Friend and Kravis (1957) perhaps being the most noteworthy. While Edwards conducted his own surveys, most other studies were based on data which had been gathered through a handful of national and local surveys of personal consumption expenditures. The nationwide surveys included a 1935—36 National Resources Planning Board study of Family Expenditures in the United States (1941) and a 1950 U.S. Bureau of Labor Statistics (BLS) Study of Consumer Expenditures, Ipcomes, and Savings (1956). Local surveys included a series of BLS budget studies conducted in Washington, D.C. (Humes, 1947); Denver, Detroit, and Houston (BLS, 1948); and Memphis (Ruark and Mulcahy, 1949). The above mentioned studies will not be reviewed here as this has already been done by others. In particular, Alexis (1962) con- ducted a comprehensive literature review to determine whether there was any real basis for the contention that black and white families with comparable incomes tend to allocate their income differently for var— ious household budget items. "When all the data have been digested," he concluded, "the following major findings emergez" 1. Total consumption expenditures of Negroes are less than for comparable whites, or Negroes save more out of a given income .— m-_ 33.5.21. ' ' 19 than do whites with the same income.2 2. Negro consumers spend more for clothing and non-automobile transportation and less for food, housing, medical care and automobile transportation than do comparable whites. 3. There is no consistent racial difference in expenditures for either recreation and leisure or home furnishings and equip- ment at comparable income levels (p. 27). Shortly after Alexis' literature review was published, new data from the 1960—61 BLS Study of Consumer Expenditures (1964) became available and subsequently were analyzed by Bauer and Cunningham (1970a) and by Brimmer (1966). Bauer and Cunningham focused on consumption expenditure patterns in 1960-61, while Brimmer examined changes in expenditure patterns between 1950 and 1960-61. Both studies will be reviewed in detail, as they represent the most recent available data.3 In analyzing the 1960-61 consumer expenditures data, Bauer and Cunningham (1970a) utilized a built-in income control to treat the black and white populations "as though they had identical income dis- tributions with equal numbers of persons in each income category." 2This finding remains highly controversial among economists. Using the 1950 BLS data, Sawyer (1962) calculated marginal propensities to consume for blacks and whites from various sections of the country. He found that "whereas it is obvious that in no case are the marginal propensities to consume for the two races the same, statistical tests reflect that the difference between the two groups is no greater than the difference within each race" (p. 218). Thus he rejected the hypoth- esis that race is a factor in determining consumption patterns. According to economists such as Galenson (1972), blacks do not save more than whites. Rather, whites dissave more than blacks; that is, whites are more able and willing than Blacks to fall back on finan- cial reserves to support increasing consumption levels. Some have also suggested that whites are willing to allocate more of their income or savings for consumption because of factors such as greater job security, income mobility, and credit availability. For more on this topic, see Alexis (1962) and Alexis, Haines, and Simon (1972). 3Data collected in the 1972-73 BLS Consumer Expenditures Survey 'were scheduled to be made available late in 1976. 20 This was done by averaging the percentage distribution of expenditures across eight income categories for each race. The resulting profile of black—white consumption patterns, as shown in Table II-1, was con— sistent with Alexis' findings--with two exceptions: "Negroes in 1960- 61 spent as much as whites on housing ('shelter'), and they clearly spent more on household furnishings and equipment" (p. 20). The major differences between black and white consumption pat- terns are summarized in Table II—2, which shows the absolute percentage of "overspending" and "underspending" by blacks for each expenditure category, as well as the relative proportion of black/white spending.4 As the reader can see, black and white families allocated about 6.6 percent of their income differently. Blacks spent more than whites on clothing, personal care, household furnishings, alcoholic beverages, and tobacco. Whites spent more than blacks for medical care, food, transportation, education, fuel and light, and "other." The two races spent about the same amount for heusing and recreation. Bauer and Cun- ningham suggest that this expenditure pattern indicates that blacks tend tend to allocate a larger share of their inc0me for "maintaining appear- ances" and for "immediate gratification," possibly due to their his- torical lack of equal opportunity in areas such as housing and educa- tion (p. 22). Brimmer (1966) compared the 1960-61 expenditure data with expenditure data from 1950 to determine whether any major shifts in consumption patterns had occurred among black and white families. As a 4Bauer and Cunningham used the terms "overspend" and "under- spend" in the comparative sense; no evaluation of black spending habits was intended. ‘—_ A—u- ‘- ,‘.,._ .. —-._._, '—'< .,:mL~;“ 21 TABLE II-1. BLACK VS. WHITE DISTRIBUTION OF FAMILY EXPENDITURES FOR CURRENT CONSUMPTION: 1960-61 (Controlled for income: $1,000-$14,999 income inclusive)a Percentage of Total Expenditure Category Consumption Expenditures White Black U' Total food expenditures ........... 25. Food prepared at home .......... 20. Food away from home ........... Tobacco ................... Alcoholic beverages ............. Shelter ................... Rented dwelling ............. Owned dwelling ............. Other shelter .............. Fuel, light, refrigeration, and water . . . . Household operations ............ House furnishings and equipment ....... Clothing, material, and services ...... Personal care ................ Medical care ................ Recreation ................. Reading ................... Education .................. Transportation ............... Automobiles ............... Other travel and transportation ..... Other expenditure ............... DE 24.4% 20.0 y—I s—Ie—t HOTNth p—a y—I HNSDHOOW'AWNU'TO‘AOD v—u—I NHHwOHwVNmbwamee—av—Im ................. Nm-bHQOmHmKOOSmmUTHmHVCDI—INV . . . . . . . . . . . hbmsomtowmoomwwml—awwl—Awoa I—l O #0001 O HU'IU'lr—I H R H O (HMO-b O R Expenditure for current consumption . . . Value items received without expenditure Food .................. CNN-5N (A) SOURCE: Raymond A. Bauer and Scott M. Cunningham, Studies in the Ne ro Markgt_(Cambridge, Mass.: Marketing Science Institute, 1970), p. 20. aIncome control was obtained by "averaging averages"--that is, the percentage for each income group was weighted by 1, summed, and divided by 8, the number of income categories. Income categories under $1,000 and over $15,000 were excluded from the analysis. Total sample size was 8,000 families. bFor whites (controlled for income), 25.7 percent of the total expenditures for current consumption was spent on food. 22 TABLE II-2. SUMMARY COMPARISONS 0F BLACK—WHITE EXPENDITURE DIFFERENCES (Per Household) . Proportion Expenditure Category ngggztagfi Eifizezgnce Black/White u Expenditure Clothing ............ 3.6% 1.4 Personal care ......... 1.0 1.4 Household furnishings ..... 0.7 1.2 Household operations ...... 0.5 1.1 Alcoholic beverages ...... 0.5 1.4 Tobacco ............ 0.2 1.1 Recreation ........... 0.1 1.0 Total overspending +6.6% Shelter ............ 0.0% 1.0 Fuel and light ......... -0.2 0.9 Education ........... -O.4 0.6 Other ............. -0.8 0.6 Transportation ......... —1.2 0.9 Food .............. -1.3 0.9 Medical care .......... -2.6 0.6 Total underspending -6.6% SOURCE: Raymond A. Bauer and Scott M. Cunningham, Studies in the Ne ro Market (Cambridge, Mass.: Marketing Science Institute, 1970), p. 21. 1' 23 ."rough measure of the income elasticity of consumption” Brimmer cal— culated the percentage increase betwen 1950 and 1960-61 in consumer expenditures for selective categories of goods and services in response to each 1.00 percent increase in family income after taxes (pp. 281-82). As shown in Table II-3, his analysis revealed that during this period blacks tended to allocate their increased income rather selectively to those expenditure categories for which they historically spent propor- tionately 1g§§_than whites. For instance, they placed more emphasis on education, household operation, housing, automobiles, and medical care; while they placed less emphasis on tobacco, fuel, clothing, and alco- holic beverages. In Brimmer's view, "this shifting pattern of outlays by Negro families represents a gradual conversion toward the consuming behavior of white families" (p. 281). Thus he has predicted that: . . In the future, Negro families will register the strongest demand in those areas associated with overall upgrading in their standard of living. There should be a strong market for housing and household operation, automobiles and medical and personal care. Those areas which have traditionally received a good share of the Negro's patronage—-tobacco, clothing, alcoholic beverages and food--will probably be characterized by relatively slow growth (p. 282). Apparently, Brimmer feels that black-white consumption differences will gradually fade away (1) as the income of black families continues to rise both in absolute terms and relative to white income, (2) as racial discrimination declines, and (3) as marketers pursue black consumers with greater vigor. One naturally wonders to what extent Brimmer's prediction has come true during the past 15 years as blacks have experienced perhaps their greatest social, political, and economic progress. Unfortunately, -.—--- -- --v-v.—.-.—..~ Wm 24 TABLE II-3. PERCENTAGE CHANGE IN SELECTED CONSUMPTION EXPENDITURES IN RESPONSE TO PERCENTAGE CHANGE IN AFTER-TAX INCOME FOR BLACK AND WHITE FAMILIES: 1950 AND 1960-61 - Black White Expe'm' tum ”em Families Families 1. Education ............... 4.22 3.06 2. Household operation .......... 2.39 1.50 3. Housing: owned dwelling ....... 2.25 2.11 4. Automobile .............. 1.93 1.12 5. Medical care ............. 1.80 1.56 6. Housing: rented dwelling ....... 1.74 1.72 7. Personal care ............. 1.66 1.58 8. Food away from home .......... 1.02 0.51 9. Reading ............... 1.00 0.85 10. Tobacco ............... 0.91 0.78 11. Recreation .............. 0.78 0.56 12. Fuel, light, etc ............ 0.73 1.09 13. Clothing ............... 0.64 0.54 14. Alcoholic beverages .......... 0.49 0.75 15. Transportation (exc. auto) ....... 0.43 0.77 16. House furnishings .......... 0.17 0.13 17. Food (at home) ............ O 12 0.27 Total consumption ......... 0.80 0.81 SOURCE: Andrew F. Brimmer, "The Negro in the National Economy,” The American Nggro Reference Book, ed. by John P. Davis (Englewood CI1ffs, N.J.: Prentice—Hall, 1966), p. 281. 25 this question must go unanswered until results from the 1972-73 BLS Consumer Expenditures Survey become available.5 Meanwhile, it appears that--in the aggregate--black and white consumers do in fact allocate their income differently for selected expenditure categories. And the fact remains that income parity between the two races continues to be more of a dream than a reality. As Alexis (1962, p. 27) has noted, "it is impossible to predict with any accuracy when Negroes shall be accorded all the rights and privileges which whites take for granted.“ Product Usage and Brand-Choice Behavior The data discussed in the previous section may be too general to be of much use to businessmen who are conditioned to think and plan not in terms of aggregate expenditure patterns and broad product cate- gories, but rather in terms of market segments and product—market shares. Fortunately, more specific cata are available (if not always reliable). This section focuses on comparative product usage and brand- choice behavior for automobiles, appliances, clothing, packaged goods, and alcoholic beverages. 5The 1960-61 BLS consumer expenditures data were updated by the National Industrial Conference Board in 1966 by applying current demo- graphic data to the 1960—61 figures and then presenting the 1966 esti- mates in the form of "share-of—market" percentage distributions. "This approach assumes that any change which may have occurred in total con- sumption expenditures for a particular product or service between the original survey and 1966 was experienced in ab0ut the same magnitude by all social and economic segments of the nation's household population" (Linden, 1967, p. 7). The Conference Board's 1966 market share esti— mates were in turn used by the Commerce Department to calculate esti- mated nonwhite consumption expenditures for 1973--on the aSSumption that "market shares change to slightly within a decade" (1975, p. vii). This assumption seems quite unrealistic and naive in light of all the changes blacks have experienced since 1960-61, and one wonders why the Commerce Department went to so much bother when more recent data from the 1972- 73 expenditures survey had already been collected and presumably were being analyzed. The author questions the accuracy of the 1973 estimates and does not feel compelled to present them in this study. 26 Automobiles One of the most pervasive and enduring stereotypes of the black consumer is his fondness for ”big, flashy, expensive cars"--Cadillacs, in particular. An editorial published in Epppy_magazine some twenty- five years ago suggested that: . Investment in a Cadillac has its points too as a weapon in the war for racial equality. The fact is that basically a Cadillac is an instrument of aggression, a solid and substantial symbol for many a Negro that he is as good as any white man. To be able to buy the most expensive car made in American is as graphic a demon- stration of that equality as can be found (1949, p. 34). Of course, not every black American owns a Cadillac. In fact, the reader nay recall from the previous section that blacks spend propor- tionately lg§§_than whites for automobiles. The Census Bureau estimates that about 85 percent of the nation's white households, but only 57 percent of the black households, had at least one car ”available" in 1970; some 37 percent of the white households had two or more cars available, compared to only 16 percent of the black households (1972b, p. 97). Further, from 1968 to 1972, expenditures for automobiles aver- aged $3,153 for white households and $1,646 for black households (Cen- sus Bureau, 1974, p. 110). These differences may be partially due to the fact that the black population is heavily concentrated in central cities where mass transit is usually readily available and where a num- ber of hardship factors may act to reduce automobile ownership. How- ever, Bennett found that nonwhite families in metropolitan New York and Chicago spend much less on automobiles than white families with similar incomes, locations, and family composition (1967, p. 849). How about those blacks who gp_own cars——do they, as the ~r—-———————————a I. . - .géxo‘f‘. , 27 stereotype suggests, drive “big, fancy, expensive cars“? Alexis (1959) conducted a detailed multivariate study of car ownership among blacks and whites based on government surveys of consumer expenditures. Con— trolling for income differences, he found no evidence that blacks own automobiles of higher average value, own more automobiles in the medium- and high-price classes, or are more likely to own automobiles that were purchased used. More recently, Larson and Akers both conducted studies of automobile ownership among black and white households, and although the two studies were both conducted in the same city at about the sane time, they produced contradictory results. Larson (1968) surveyed automobile ownership patterns in Chicago and found that “the stereotyped idea that more Negroes own and drive Cadillacs and other luxury automobiles than their white counterparts was not borne out" (p. 210). This conclusion is debatable, however, because Larson's data clearly indicated that 13 percent of the black households, but only 5 percent of the white households, owned either a Cadillac, Chrysler, or Lincoln. Furthermore, although Larson concluded that "the Negro preference for automobile ownership is a function of income rather than race" (p. 210), he made no attempt to control for a $3,950 difference in median family income between blacks and whites sampled in his study. Therefore, his findings shed little light on the question of whether blacks tend to buy higher-priced cars than whites yigh comparable incomes. This question was investigated by Akers (1968) who, using an income-stratified sample of Chicago car—owners, found that: . Negroes in this study tended to own higher price class auto- mobiles, higher priced models regardless of make, and automobiles with more cylinders than comparable income white families. Also, 28 based on multivariate-regression analysis, the race variable is more closely related to each of these three automobile character- istics than income, education, sex, age, family size, or miles driven per week. No differences between the two races were found in age of car owned, length of time owned, or new or used purchase (p. 288). Akers also found that blacks tend to own more Cadillacs than whites with comparable income. Finally, Bauer and Cunningham analyzed automobile ownership data from several sources and reached the following conclusions: 1. Fewer Negroes own cars than whites, even when income and living area controls are used. 2. Those Negroes who do own cars purchase proportionately more medium-priced and prestige cars than do whites, and propor- tionately fewer low-priced models. 3. There is mixed data on whether Negroes in central city areas are less likely to buy their cars new, but it does appear likely that they tend to purchase prestige automobiles used rather than new. 4. Negroes appear to own their automobiles for a longer period of time (1970a. pp. 156—57). In summary, while the evidence is neither consistent nor con— clusive, there is some indication that blacks do tend to own more expen- sive cars than whites with comparable incomes. However, perhaps more than anything else, these studies illustrate the difficulty of trying to generalize about black buying behavior. Obviously, car ownership is dependent on a multitude of factors other than race. Not every black family drives a luxury automobile, just as not every white family drives an economical compact car. VAppliances As one might suspect given their historically lower income distribution and more restricted housing opportunities, black families I 29 tend to own and use fewer major appliances than white families. Table II-4 indicates that with the exception of television sets (mainly black-and-white sets), black households possessed proportionately fewer appliances than white households in 1970. This pattern tends to hold true at all income levels, according to a 1968 study conducted in the Houston area by Stafford, Cox, and Higginbotham (see Table II—6). How- ever, there is evidence that blacks, particularly middle-class blacks, are placing greater emphasis on major appliances as their discretionary income continues to rise while racial discrimination in housing declines. As Table II—S shows, when income differences are taken into account, black h0useholds tended to outspend white households for many big-ticket items during the 1968-72 period. A more revealing comparison of black-white applicance usage was made by Bauer and Cunningham (1970a) who analyzed data from a 1961 Starch survey of appliance ownership and proceeded to categories appli- ances and other products as either necessitities, discretionary items, or luxuries. Necessities were defined as appliances "without which 'no American home is complete,'" and included items such as refriger- ators, stoves, radios, irons, and television sets. FOr all such items, it was found that both black and white ownership was close to satura- tion with very little income elasticity (p. 71). Discretionary items were defined as products whose "existence and benefits are quite universally known, but whether or not one acquires them is considered a matter of personal preference, resources, and of judgment about the alternative demands on one's money." 0f the products judged to fall within this category, white families at all income levels owned more or spent more on automobiles, washing machines, 30 TABLE II—4. PERCENTAGE OF BLACK AND WHITE HOUSEHOLDS POSSESSING SELECTED APPLIANCES: 1970 Percentage Appliance Black White Clothes dryer .................. 12% 74% Dishwasher ................... 3 49 Home food freezer ................ 21 29 Clothes washing machine: Available ............... . . . . 51 74 Automatic or semiautomatic ......... 34 63 Wringer or spinner ............. 16 11 Television sets: Available ................... 92 96 One set ................... 69 67 Two or more sets .............. 23 29 Air conditioning: Available ................... 18 39 Room units ................. 15 27 Central system ............... 3 12 SOURCE: The Social and Economic Status of the Black Population in the United States, 1971. Current Population Reports, Series P-23, No. 42 (Washington, D.C.: U.S. Government Printing Office, 1972), Table 77, p. 97. 31 TABLE II-5. AVERAGE BLACK AND WHITE EXPENDITURES PER HOUSEHOLD ON CAR AND SELECTED NEW HOUSEHOLD DURABLES: 1968-72 S Expenditures . Black Expenditure Category Bia§:;?21te Buying Black White Index All items ........ $2,654 $4,495 0.59 98 Cars, total (net)b ....... 1,646 3,153 0.52 86 New (net) .......... 949 2,062 0.46 77 Used (net) ......... 697 1,091 0.64 107 Selected appliances, total . . . 271 421 0.64 107 Washing machines ...... 52 69 0.75 125 Clothes dryers ....... 14 42 0.33 55 Kitchen ranges ....... 50 52 0.96 160 Refrigerators and freezers . 88 110 0.80 133 Dishwashers ......... 3 27 0.11 18 Room air conditioners . . . . 25 42 0.60 100 Otherc ............ 39 79 0.49 82 Home entertainment items, total . 465 592 0.79 132 Black and white TV sets . . . 60 41 1.46 243 Color TV sets ........ 112 180 0.62 103 Radio, phonographs, and hi-fi equipment ...... 100 108 0.93 155 Home furnishings, total ..... 465 592 0.79 132 Furniture .......... 383 415 0.92 153 Carpets, rugs, and other floor coverings ...... 82 177 0.46 77 SOURCE: The Social and Economic Status of the Black Population in the United States, 1973. Current Population Reports, Series P-23, No.48 (Washington, D. C.: U.S. Government Printing Office, 1974), table 78, p. 110. aReprinted by permission from Thayer C. Taylor, "Black Middle Class: Earn, Baby, EarnJ'Sales Management (8 July 1974), p. A-11. The index adjusts comparative black and white spending to allow for income differences. During the 1968-72 period, black family income averaged 60 percent of white famliy income. Thus on a dollar-for- dollar basis, the black-to—white spending ratio should be 0.60. Where the ratio is higher than 0.60, the index will be above 100, indicating that the typical black family is actually outspending the white family when its lower income is taken into account. bNet expenditures are equal to gross expenditures less trade-in allowances. cVacuum cleaners, blenders, mixers, and sewing machines accounted for approximately one—third of all "other" durables reported pur— chased. 32 telephones, electric clocks, still cameras, vacuum cleaners,_ electric mixers, electric coffee makers, electric toasters, tools worth $50 or more, electric fry pans, and wine. Meanwhile, black families owned or spent as much or more on clothes, furniture, painting or decorating, record players, beer, and liquor (p. 74-74). Luxuries were defined as "items of a broad range of price and newness that have not established themselves as widely 'worth the price.'" Included in this category were clothes dryers, electric blankets, photographic equipment, floor polishers, movie projectors, food freezers, air conditioners, slide projectors, garbage disposals, dishwashers, outboard motors, electric can openers, boats, and color TV sets. (Obviously, many of the products seen as luxuries in 1961 might now be viewed as discretionary items or even necessities). Even at the highest income levels, less than half of the white households owned_any of these items. Nevertheless, in every income category, whites were more than twice as likely as blacks to own any of the luxury goods (p. 76). Bauer and Cunningham's analysis indicates that "as one moves from necessities, to luxuries, the relative spread between Negro and white ownership increases" (p. 79). Moreover, "Negroes with the same household income (and roughly the same per capita income) are consid- erably slower in acquiring such products, particularly as they are seen as luxuries" (pp. 79-80). This latter finding is partially sup- ported by the work of Dalrymple, Robertson, and Yoshino (1971), who found that blacks were slower than whites or Japanese-Americans to adapt new food or small-appliance products but faster to adopt cloth- ing innovations. As they pertain to appliances, these findings can 33 only partially be explained in terms of lower incidences of home ownership among blacks and the availability of adequate electrical and plumbing connections. For example, while black households tended to be saturated with refrigerators and owned proportionately mpyg_record players than white households, whites in a given income catetory were twice as likely to own a toaster, vacuum cleaner, or electric coffee maker. Thus, to Bauer and Cunningham it seemed that ”Negroes did not choose to spend as much of their money as whites on household gadgets of the type that are commonly assumed to increase the convenience and quality of housekeeping and cooking" (p. 75). Obviously, however, we are again dealing with complex phenomena which defy simple explanations. 9.1mm Edwards observed in 1932 that "the Negro buys gaudy, loud mer- chandise, which is durable and wears well" (p. 47). Furthermore, he wears better clothing in proportion to his income than the white man" (p. 48). More recently, Kindel (1970) found that black college students were more brand conscious about shoes than white college students, although the latter were more brand conscious about suits and dress shirts. In general, Kindel found that, in comparison to their white counterparts, black students were more conscious of newer clothing styles, more likely to try out new styles, and more likely to be influ- enced by styles in selecting a retail store. Dalrymple, Robertson, and Yoshino (1971) have also reported that blacks are faster than either whites or Japanese-Americans to adopt clothing innovations. While such stereotypes may not apply to all blacks, government expenditures surveys have consistently revealed that--at all income 34 levels--blacks tend to allocate a larger percentage of their income to clothing expenditures than do whites. The Bauer and Cunningham data reported earlier (Table II-l on page 21)indicate that, after control- ling for income differences, black families allocate about 12.5 percent of their total consumption expenditures to clothing, while white fami- lies allocate only 8.9 percent. There is Some evidence that blacks not only buy proportionately more units of clothing (perhaps due to factors such as family size, age, and occupation), but also that they often pay more for clothing--either voluntarily or due to neighborhood shopping constraints (Bragulia and Rosencranz, 1968). Exactly why blacks allocate more of their income for clothing is a complex question which defies any simple answer.6 One possible explanation, as suggested by Andreasen and Hodges (1976, p. 16) is that "clothing has a much stronger psychological meaning in the black community [and] this is most frequently manifested in higher interest in fashion." Ngwsweek (10 September 1973, p. 54) essentially echoed this line of reasoning in attempting to explain why the black male is a "clotheshorse nonpariel": Through their clothes, black men are celebrating their freedom, displaying their success, exercising their creativity and affirming their racial pride. And among designers and manufacturers, the fashion-conscious black man is considered a trend-setter in men's styles for all races. Black women are also interested in fashion according to Portis (1966). However, although his research revealed that blacks at lower- and middle-income levels were somewhat more interested in fashion than their white counterparts, the overall frequency of fashion-conscious ---. .a-n— 7For an in-depth discussion of this topic, see Kindel (1979) and Andreasen and Hodges (1976). 35 shoppers among black and white women proved to be quite similar. More— over, Portis found little indication that black women follow clothing fashions in ways different from white women. 0n the other hand, he observed large disparities in fashion interest among black women which in turn led to differences among them in terms of how they followed fashion and shopped for clothing. Thus Portis concluded that: . . Negroes, though an important market for fashionable clothing, may not constitute a s ecial market . . . (p. 314). . . . the marketer should recognize that he may be trying to reach some fashion-conscious women who happen to be Negro rather than Negro women who happen to be fashion-conscious. . . . Fashion consciousness is an attitude or need which women possess as individuals and can only be understood on an individual basis. Race is a label applied to groups, and although it provides some indication as to a particular group's social situation, it bears no necessary relation to the shopping habits or needs of indi- viduals (p. 323). No doubt some would disagree with Portis' conclusion. However, one well-known black fashion designer has stated that: "I resent the idea of 'black fashion.‘ Let's just talk about fashion" (Newsweek, 10 September 1973, p. 54). And according to the owner of the nation's largest black-owned department store, "Four years ago there was a definite black fashion look, but that's on the way out. Blacks are, more interested in quality" (Business Week, 13 April 1974, p. 102). All of this can only serve to further confuse marketers as they wonder if and how to cater to the "black clothing market." Packagegugppd§_ Numerous studies have been conducted to compare black-white consumption behavior for packaged goods. While only a representative handful of studies will be reviewed here. this mini-review should be sufficient to once again demonstrate the difficulty of generalizing 36 about black buying behavior. Stafford, Cox, and Higginbotham (1968) sampled 211 black and 1,335 white households in the Houston area to determine if, and to what extent, black-white consumption patterns varied for five product cate- gories--food, soft drinks, liquor, personal hygiene products, and major home appliances. Their results are summarized in Table II-6. ”A major finding of this study was that, for many household products, consumption-pattern differences were small in number and magnitude" (p. 628). Further, although extensive differences were found for a number of products, the authors felt that "a substantial portion of these differences . . . were explainable more in terms of income or sociodemographic variations than by purely 'racial' influences" (p. 629). However, they could find no "economically 'rational' explan- ation" for certain consumption differences-~such as the tendency for blacks at all income levels to consume more butter and Scotch whiskey than whites--and therefore speculated that such differences may be due to compensatory or conspicuous consumption. Taking all these factors into consideration, Stafford et al. concluded that: . From a businessman's point of view . . . A Negro market does exist, not so much identifiable by color as by patterns of consumption. Marketers who assume that product buying in Negro households is roughly a match for that in white familiar of similar economic circumstances are far from correct. A combination of societal constraints; cultural traditions; and differences in values, preferences, and psychological needs have led Negroes . . . to vary their expenditures across different products and, probably, brands compared with whites (p. 630). However, the authors also noted that: . The indications in this study are that the Negro market is not completely homogeneous . . . there has been increasing economic and cultural stratification within the Negro community which, among other things, has led to internal consumption- pattern differences (p. 630, footnote 16). 37 uwc?neou mew: mwmmmpo mEoucw oz» ammo; lead» .mucmucoamme _muou ocean mmmmcucsa coagaom new :uuoum mo mmmucmuemam .mucmucoamme mo census meEm mo mmzmumn .wEo; mg» :_ =mcw>oc= wmmucmuemam .mxmu cm umma cwcuwz vmmmcu .mcpcos NH amen :_;u_3 ammo lem>ma uw~ozoupm meow mcwmmzueza mucmucoamm; _muou ou mompcmugoav .mmoumEOu use .mcoma :mmgm .momg .cgou uwccmu mmuzFucHu .mm—amummw> cm~ocw mo mquu Fpm mmoapucH n .mxmu cw>mm umma mnu cwcumz ummwzucaam .Nmuomo Hammad Lmasmomov me xggmpcmao mucmwum meuom =.mmoemmz new mmu_;3 cone: :mmzumm mmucm -eaeewo eemuoma eo.uae=meou meow: .EQgponewmm_x .m magma was .xou .x ;S_a¥ .ucocemom .m magma ”mumaom m.~n . m.Hm o.m~ m.o~ H.5m ¢.o¢ m.mm m.mo . . . . . . . . mean :30 a_:mgm:zo mac: --- --- cm.m cm.¢m m.H n.m 0.0 m.m . . . . cowmw>mqu Lo_ou --- --- cm.~m ;~.mm “.mm m.mw m.Hm e.~m . . . . . cowm_>m_mu 3am --- w.mm m.m ~.¢H -u- m.m --- o.~ . . . . emzmmzzm_u .ou:< m.~m m.em «.mH N.¢m w.“ m.oH w.m m.NH . . gmxcu mmzpopu .ou:< m.m~ m.mm 0.0m o.mn $.mm m.~m w.mH ¢.~w . mcwcuos newsmmz .ou:< ammucmwpaam wee: c.0m. m.mm m.Ho H.o~ m.om c.om m.mo m.~m . . . . . mucmpummcwmwo v.0w m.mo m.wm m.mm m.mn m.mm o.~m “.mv . . . . . . . :mmzzuaoz «.mm H.mm m.ww m.mm o.om o.m~ m.o~ m.w¢ . . . . . . . mummazuoo» m.Hm n.0m m.~m m.v~ N.m~ m.om H.mm “.mm . . . . . . . “cmcovomo o.om o.~n «.mm m.en o.mm m.mm m.He e.me . . . . . . . . ooqsmgm emuuauogm wcmwm»: Pacomgma m.oe m.o¢ ~.~ N.mN «.mm m.o~ H.mH m.n . . . . . . . mconcsom m.u~ 5.0H o.Vm N.“ H.NN ~.¢ m.m m.m . . . . . . . . acuuoum m.¢m m.om «.me m.mm o.mm “.mm “.mm «.mH . . . . omucmvcoammg —_< Lo:c_4 ¢.m¢ H.Nm H.mm «.mc ¢.HN m.mm m.oc m.o~ . mxcweu “mom xumumvucoz m.mH m.mm H.m~ m.o~ «.mm m.HH v.nH m.m . . mxcweu “mom xgmumwo ~.m¢ m.o¢ v.oe m.~m m.¢¢ m.mm o.mm m.om . . . ommFDoummm> umccmu o.¢m H.me o.¢m o.mm m.om o.m~ ¢.Hm m.om . . . cmwpnoummm> cm~ogm w.~m m.mm n.5m m.mm ~.NN m.mo m.ao m.mm . . . . . . wcwcmmcm: a<.m¢ NH.QH xm.m~ &~.N gm.~m xo.m Nm.m~ ao.o . . . . . . . . . Lmuuzm mmuuzuoga coon mxumrm mmpwcz mxumbm. mmuwcz mxumFm mmumcz mxuahm mouwmz. mcos Lo ooo.mm mam.hlooo.mm omm.m-000.mm ooommm cusp mmm; muuauoem wEooc~ xprsmd —m:cc< mhuaooma o40zmm=ox monm<> omzzo 0:3 mo omm4kzwuum o