<^i *i\*4 Division of Agricultural Sciences fvu ' V V UNIVERSITY OF CALIFORNIA %,\^ ^ GROWTH CHARACTERISTICS OF 30 MAJOR RETAIL FOOD CHAINS «- 1953-1963 K. D« Duff . . . D. B* DeLoach CALIFORNIA AGRICULTURAL Ml II FN N 829 EXPERIMENT STATION BULUIIN 8 a CONTENTS Page THE FINDINGS 3 GROWTH CHARACTERISTICS OF 30 MAJOR RETAIL FOOD CHAINS, 1953-1963 5 Purpose of the study 5 Procedure 5 CHARACTERISTICS OF THE MARKET 6 Demographic changes 6 Disposable income 8 CHARACTERISTICS OF THE RETAIL FOOD INDUSTRY 10 THE RETAIL FOOD CHAINS 15 Growth characteristics of chains 17 Effects of Growth 24 Financial interrelationships 26 INDUSTRY PERFORMANCE AND PUBLIC POLICY 29 Methods used 29 Growth patterns 29 Effect of population changes on number and location of grocery stores 34 Conduct and performance of chains 35 GOVERNMENT REGULATION 40 Legal restraints 40 Public opinion 40 Current investigatory philosophy 41 APPENDIX A 43 APPENDIX B 48 APPENDIX C 54 APPENDIX D 58 Application of Markov technique to structure data 58 Limitation to the Markov analysis 60 APPENDIX E 64 LITERATURE CITED 65 JANUARY, 1967 THE AUTHORS: K. D. Duft is Research Assistant in the Experiment Station, Davis. D. B. DeLoach is Professor of Agricultural Economics and Agricultural Economist in the Experiment Station and on the Giannini Foundation, Davis. THE FINDINGS This bulletin reports the study of the operating performance of 30 major gro- cery chains over the 10-year period 1953— 1963, their pattern of growth, and the factors behind their growth. It provides an economic analysis of the concentration of market power in the retail grocery busi- ness. Today, more than enough retail food stores are available to supply the require- ments of the local population in many parts of the United States. The retailers, competing for sales, cater to customer pref- erences for products and services, as much as this is economically feasible. Population shifts, among age groups as well as geographically, affect managerial planning for food distribution. For in- stance, figure 1 shows a projected increase by 43 per cent in the 20-to-29-year age group by 1972, thus constituting a large in- crease in new family units. Similarly, the westward migration of our population forecasts a higher level of demand for food products, marketing services, and facilities in the West. Because market growth offers opportunities for food retailers, new mer- chandising technologies find fertile ground. The firms that survive must pro- vide modern facilities, services, and prod- ucts at competitive prices. Corporate chains, operating only 10 per cent of the grocery stores in the United States, sold more than 41 per cent of the total grocery store volume in 1964. Affili- ated buying groups handled 50 per cent, and unaffiliated stores 9 per cent. About 68 per cent of the total grocery sales volume moved through supermarkets — stores with an annual sales volume of $500,000 or more. The relative volume of all small stores (up to $150,000 annual sales) continued to decline after 1958. This reflects a further shift of patronage to mass merchandising establishments. But the number of relatively small chain store firms increased, which indicates that es- tablished firms do not erect entry barriers, especially since most of these newer chains are found in large metropolitan areas where they are competing with the larger previously established chains. The new chains, however, may find it difficult to obtain capital for entry and to survive in a competitive market. Both from the standpoint of relative sales volume and actual store numbers, the two largest firms — The Great Atlantic and Pacific Tea Company and Safeway Stores, Incorporated — were relatively less impor- tant in 1964 than at any time during the last 20 years. The relative share of annual sales of the 20 major chains was about the same in 1929 and 1962. The most signifi- cant sales growth took place among the medium-size chains. Several financial items were analyzed. Among the significant results are these: • The portion of total assets considered fixed increased slightly, possibly cir- cumscribing some of the flexibility needed in a progressive industry. • The smaller chains seem to be using more short-term credit than the larger chains, particularly for open account purchases. • The net worth position of the larger concerns improved slightly compared to that of the smaller chains. This seems to reflect their retention of cor- porate earning instead of outside financing. • Before 1958, small chains had the largest net profits per sales dollar. Since 1958, this pattern has been re- versed. • Net sales were more responsive to changes in store numbers in the large- chain group than in the smaller-chain groups. • On an average, the addition of one store tended to increase a chain's total annual net profit by $10,000. The small-chain group showed the lowest level of response in this respect. An effort was made to account for the growing importance of retail chains in food distribution. The most obvious con- clusion from the analysis is that, given the state of our technology, the "push for big- ness" that characterizes most businesses, including agriculture, shows that size, sales, [3] and buying volume are basic to minimiz- ing unit costs and increasing total net profits. Nonetheless, retail chain managers are always seeking the optimum size for their individual stores and the optimum number of stores within their given firm. Through the analysis, the following gen- eral tendencies show up: • Group I (large-size chains) has the highest net return on net dollar sales on the average, while many of the large firms seem to be losing their grip on total industry sales. • Group II (medium-size chains) seems to be most dynamic in increasing the scope of its operations. The data in- dicated great variations among the operating trends of the 10 individual chains in this group. • Group III (small-size chains) seems to be rapidly expanding its relative share of the market, while as a group it received the lowest relative net profit as percentage of net dollar sales. None of the chains seems to be satisfied that its present size, both in terms of sales and stores, provides the solution for op- timizing profit. Therefore, experimenta- tion goes on. The authors' attempts to project the outcome of this experimenta- tion led to the conclusion that a further growth in the average size of chains is likely. This does not mean, however, that the large chains will grow still larger, compared to the others, but that the small- and medium-sized groups will increase in size. Most investigations of the food industry have been made in answer to expressed dissatisfactions about some element of in- dustry performance, usually about alleged inequities over prices paid or received at various levels of food production, manu- facturing, or distribution. Most investiga- tions have assumed that the alleged in- equities are associated wth the market con- duct of large-chain firms that use their economic power to influence, unduly, the terms of trade at all levels of marketing. It is the judgment of the authors that technology has fostered and encouraged bigness in almost all of our basic industries, including agriculture. Large-scale, multi- store retail food firms seem to be the only satisfactory arrangement to provide the investment capital required to support the modern technologies used by such firms. The only alternative source of capital of such magnitude would be a federal agency. This alternative has far-reaching implications. The authors consider the growth of large food chains as a function of the com- petitive struggle for customers. Through- out the evolutionary period of mass mer- chandising, consumers have had alterna- tives available to them. If and when con- sumers find a more satisfactory solution to their food supply problems, retailing methods will be adjusted further. This is a basic attribute of the competitive process. r 4 j GROWTH CHARACTERISTICS OF 30 MAJOR RETAIL FOOD CHAINS— 1953-1963 1 v^ongress authorized the creation of the National Commission on Food Marketing on July 1, 1964, to conduct a broad eco- nomic inquiry of our food processing and distribution system and to report its find- ings by June 30, 1965. Ensuing delays in organizing the Commission and its staff made it necessary to extend the reporting date to June 30, 1966. Many leaders in the food industry were apprehensive about the type of inquiry that would be undertaken and believed that the short time allocated to the Commission for a comprehensive study would make a thorough appraisal of the economic problems confronting the industry impossible (Ginsburg, 1966). Many officials of retail food chains — cen- trally owned and controlled retail grocery firms with two or more store outlets — were uneasy about the direction the in- quiry would take. Their past experience led them to believe that many economists, legislators, and farm leaders had already publicly taken a position on three of the main issues submitted to the Commission for study: • That there is an undue concentration of market power in a few retail food chains. • That the chains use their market power to depress farm prices and keep prices to consumers higher than they would be under more competitive conditions. • That additional government restraints or remedial legislation are needed to correct the imbalance among the vari- ous groups engaged in producing, processing, distributing, and using food (Adamy, 1966). Purpose of the study The authors of this report, along with many other economists, are unwilling to accept the foregoing claims without fur- ther examination (Tongue, 1965). In par- ticular we wanted to probe the implied Submitted for publication August 4, 1966. charge that retail food chains strive to in- crease their size (power) primarily to mani- pulate their buying and selling prices. An alternative explanation for growing power of chains seems to be their advantage in obtaining the investment capital required to establish and operate modern retail outlets, including their ability to compete in pricing and promoting their mechan- dise, and their greater flexibility in adapt- ing to the demographic changes in the United States. To learn more about the reason for the growth in size and market power of retail food chains, an analysis of the growth characteristics of 30 major national and regional retail grocery chains was under- taken. This bulletin reports one phase of the study, supplementing U.S. Federal Trade Commission (1960) and Hiemstra and DeLoach (1962). Procedure The procedure for the analysis was dic- tated partly by the nature of the obtain- able data. Fairly complete records are available for 30 grocery chains for the period 1953-1963, inclusive. In addition, the 1954 and 1959 censuses provide bench- marks against which the trade data can be compared. To analyze the growth characteristics of the 30 chains included in this study, we sketched the environmental conditions which give rise to the demand for the number, kinds, and qualities of retail serv- ices offered consumers. To show the im- portance of recent changes in the structure of the retail grocery business, comparative sales data for both the independent (one- store firms) and chains are introduced. In our analysis of the 30 chains, we classify each chain, according to its annual sales volume, into one of three groups of 10 stores each: Group I contains the large chains with $400 million or more annual sales, Group II includes the medium chains with $100 to $400 million annual sales, and Group III has the small chains with [5] less than $100 million annual sales. The groupings are used to show comparative changes among groups I, II, and III and among the chains within each of the groups. Changes in several important operat- ing policies and practices are brought out by a comparative analysis of the financial records of the 30 chains. The performance results of each chain within and between the size groups are shown through a com- parison of net returns on investment and/ or dollar sales. Finally, an attempt is made to appraise the impact on public policy of the growing economic power of retail food chains and to learn whether that policy has been or is an effective force in modify- ing the growth characteristics of chains. The data and other information pre- sented in this report were obtained from the published material listed in Literature Cited, unpublished material referred to in footnotes, and from interviews with indus- try and government officials. CHARACTERISTICS OF THE MARKET Retail food firms must continually adjust their business policies and practices to the changing environment in which they oper- ate. This environment is affected greatly by changes in population composition and geographic movement, and consumers' dis- posable income which can be allocated to the purchase of various products and serv- ices. Demographic changes The Bureau of the Census estimates that the population of the United States will approximate 226 million by 1975. This is about 17 per cent above 1964 and 33 per cent above 1954. The structure of the pop- ulation also is undergoing considerable change, especially among the age groups AGE Under 10 10-19 +28% ^30° ill! Il/ll I TTTTT 20-29 30-39 40-59 60 and «.E -10% -23°, Total Population 20% TT/l II I! II JJJ. +86% ^3% +26% uMZMZM™ ii/n/nua +76% i ////// m +40% ffl 1940-1962 □ 1962-1972 Fig. 1. Changing U.S. age compositions. Source: This Week Magazine, 1963. [6] cS (figure 1). The projected growth of the 20- to 29-age group is particularly impres- sive because it reflects the probable growth of 43 per cent in the number of family units for the period 1962-1972. The ed- itors of the Progressive Grocer (1965) ex- pressed their attitude toward the popula- tion changes with this comment: Youngsters married for the first time form a market which offers special oppor- tunities to retailers and manufacturers. Their marriages create new households that never bought (food) independently before. They are establishing brand and store preferences and buying habits. Be- ing inexperienced in shopping, they are more attracted to those store operators who understand their needs and cater to them. The prize, obviously, is not only their present business, but their future business as well. In addition to the changes in popula- tion numbers and composition, the food industry is paying close attention to re- gional population shifts that can and do affect their sales and facility requirements in various parts of the country. In recent years there has been a strong movement to the southwestern states and a more moderate growth in most other areas. For example, the combined population of Ari- zona and California rose about 6.7 million between 1954 and 1964, or 48.1 per cent. In the Southeast, Florida's population in- creased by 2.3 million, or 67.5 per cent, during the 10-year period. This meant that the food industry had to provide facilities to distribute food to an additional nine million people in these three states alone (appendix table A-l and figure 2). Before 1950, population tended to move from rural to urban areas. Although this urbanization process is still in progress, it has now been overshadowed by a second- ary movement of our city population into surrounding suburbs. Accompanying this movement has been the construction of a multitude of "shopping centers," almost all of which include a grocery store, vari- ous other merchandising outlets, banks, etc., all oriented toward serving the entire range of consumer wants. To offset the suburbanization trend, widespread city housing redevelopment projects have beeen completed, more are underway, and new ones are planned. This kind of redevelopment of cities has had, and probably will continue to have, an important effect on the types of food distribution facilities required to serve the inhabitants. In fact, some industry leaders believe such redevelopments may be the basis for a revival of neighborhood stores designed especially to supply a large group of consumers who will buy on a "hand-to- mouth" basis. The probable impact of population growth and geographic distribution on the number and distribution of grocery stores will be discussed under "Growth Patterns," on pages 29 to 34. Disposable income The net disposable income of consumers is a meaningful statistic to the retail gro- cer (appendix table A-l and figure 3). This income affects both the volume of sales of the food industry and the kinds of products customers are most likely to buy. Although the food and nonfood purchases in grocery stores are an outgrowth of a number of conditions faced by customers, including better dietary information and alternative products, much of the increase in consumption of meats, dairy products, fruits, and vegetables became possible as income rose sufficiently to enable cus- tomers to shift to these relatively higher priced products (table 1). Also, customers, as their incomes rise, are purchasing more luxury and specialty foods, for example high-markup foreign food specialties. Despite the fact that more money is be- ing spent for food today than ever before, the percentage of consumers' disposable income used to buy food has declined from 25.6 per cent in 1947-1949 to 18.6 per cent in 1965 (U. S. Department of Agriculture 1965c and d). Much of the disposable income used for food, household supplies, drugs, cos- metics, and many other consumer goods is spent in combination- and specialty-type grocery stores. Because our primary pur- pose is to examine the food retail distribu- tion system, the next few sections will be directed toward this goal. [8] a a < a> u c 1 _ ,>*N. □ Table 1 APPROXIMATE CONSUMPTION OF FOOD PER CAPITA BY MAJOR FOOD CATEGORIES 1910 1932 1955 1963* 1910-1963 Food category Pounds Percent- age of total Pounds Percent- age of total Pounds Percent- age of total Pounds Percent- age of total Percent- age of change Dairy products Eggs 337.0 37.1 170.2 41.9 16.3 130.6 123.0 7.6 202.5 188.0 14.5 221.1 295.0 87.9 9.6 .211 .023 .107 .026 .010 .082 .077 .005 .127 .118 .009 .139 .185 .055 .006 360.0 38.0 152.8 46.8 16.4 132.5 115.9 16.6 220.0 198.0 22.0 156.3 223.0 110.3 13.8 .238 .025 .101 .031 .011 .088 .077 .011 .146 .131 .015 .103 .147 .073 .009 398.0 46.9 191.8 49.2 16.3 143.4 94.6 48.8 202.7 154.6 48.1 113.7 152.0 106.3 15.0 .269 .032 .130 .033 .011 .097 .064 .033 .137 .104 .033 .077 .103 .072 .010 368.0 40.0 203.3 49.9 16.7 123.7 73.7 50.0 197.6 145.1 52.5 111.4 144.0 110.0 15.6 1,420.0 .259 .028 .143 .035 .012 .087 .052 .035 .139 .102 .037 .078 .101 .077 .011 + 9.19 + 7.81 + 19.44 -1- 6.62 + 2.45 - 5.28f - 40.08 +557 89 Meats, fish, and poultry Fats, oils, and butter Beans, peas, and Fruits Fresh Vegetables Fresh - 2.41f - 22.81 +262.06 - 49.61 Flour and cereal - 51.18 Sugar and sweeten- + 25.14 Coffee, tea, and + 62.50 Total 1,592.0 100 1,512.0 100 1,480.0 100 100 - 10.80 * Preliminary. t Since this pertains to pounds of fruits and vegetables purchased at the retail store, this datum may be somewhat misleading. The greatly increased consumption of processed fruits and vegetables indicates that less wastage is now purchased with the food product. Therefore, "actual" fruit and vegetable consumption may have increased while pounds of fruits and vegetable products purchased has decreased. Rising incomes during this period provide an additional basis for this explanation. Source: U. S. Department of Agriculture, 1965c, pp. 18-19. CHARACTERISTICS OF THE RETAIL FOOD INDUSTRY According to the 1964 Census, retail food store sales amounted to slightly more than $57 billion in 1963, or about 24 per cent of all retail transactions. The 1963 volume was handled by 319,433 outlets, or about 23 per cent of all retail trade outlets. The food stores employed 1,274,395 paid work- ers, or about 15 per cent of those in all kinds of retail outlets (table 2). If the 1939 and 1963 annual dollar sales are adjusted to a 1947-1949 price index, food store sales more than doubled dur- ing the 24-year period. However, the in- creased volume was handled by 43 per cent fewer stores. Significantly the domes- tically produced food sold through these stores was produced on 6.9 million farms in 1939 and 3.5 million farms in 1963, a decrease of about 50 per cent. Because the retail food industry of 1963 is a product of a changing technology and changing needs of its customers, a few comments of a historical nature seem ap- propriate to an understanding of the process by which our present-day retail food industry became what it is. The present system of food distribution in the United States has been associated with our transition from a predominantly self-sufficient, rural society to a largely in- dustrial-urban society and with advances in food technology that have made it pos- sible to handle food as it is now done. In our present industrial-urban society, [10] us O 00 TJ< os 10 ^ OS 00 1 & os 10 co ■>* us co OS rt OS CO CO US •* CM CM -* 00 t^ US ^_ CO 02 ■*" rt 00 0" CM* •*!* 00 C5 OS O N CD to CO US OS OS CO CD N N * CO CO t~- CO CO CO "*1 02 r-< r-4 rt rH 00 »9 >-. 3 o S co s h co CO t- co i-~ r^ CO CO s CO CO CO N co" «o co" CO Tt< r- O CM US 00 00 10 os t~ 00 cm" O us CO t^ i>r os OO N M *< CO -h us co co CO t^- ^ CO CM OS T3-T3 rH* rH rt' l-H r-" ' 3 S os 00 O * N Ct ^1 I— O OS co •<*< CO °a 10 OS 00 00 os co us cm" 00" r^ t- OO «-" O CO co" 0" OS CO cm us us US i-H co CM £i-n O CO CM CO CO t- co 00 si rtrtrt t^ 3 a CO CO -— 1 CO CO ^h O0 0O Ol Tf CM O us os 00 00 CM •>*> CM ^ co ■* CM OO CM OS 00 •>*< 00 us OS cc cm" -h" OS us •«»* t~-T os 00* rC co" CM r^ CM r- O O r^ ■»»< CO e O CM 00 US O 00 CO t^ CD CD *§ OS US — " cm" O -* T)" CO OO OS 1 13 T»« —. CM — ' 1-1 CO "* co OS 3 s « co h co 01 »" cm" OS CO CM CD US co •«< CO co co CM ~ US CO CM r^- us t^ US CD 00 O CO us CO -H US CM .-1 CO t-^ 1 CO CO -h CO CO CO 00 CO CM os -»ti CO CM CO _, co co CO co -f •«* O CM co us t^ -H OS CO aa OS -o IO S CO N ^H US •«* S3 S CM Tt< CO CM CO .3 CO us US ■*? CO 00 co co 00 co" ~ 00 US -f O0 r-t O os O US CM 00 3 03 CO CO >-i 1-1 US t^ i-T CO CD co s ce co t^ CO OS os § H 1(5 O t!< os us O US O CO CO h N t^ OS us O CO CO Xi us OS * Ol CO Ol ^h" US 8 CD O0 _; OO t-h t^ y-i OS 00 US O CM CO CO -h t^ B ^ 3 co n cr t^ CM CO co r^ CO fi 00 co ie CM CO O co t^ US CO CM O" OO i-H CO OS CO co CO CO O "5 C CM* O us co os tC CD us 0- CO 113 CO <— ■ of bo 4 1 a : b a a a 3 ! 1 I 3 CD -5 § 2 "3, M g» 8 a "m I 3 2 .2 C3 i '3 : o* • a : (-1 . •2 • 73 : C co J a s .2 a CD ccj i t c 3 1 3 | c X ■3 « H «" 73 g : csd g ^ crj 4) C «« T3 c S » i ° 2 b •* .2 a 1 «1 fell 1, 3 * : - u 1 p S a 3 a -a 3 b3 CO I, «| Q " a a C +. X "0 s- a "f :3 e 3 EH f= M O < h < hj c consumers are almost totally dependent on our food distribution system as a means of maintaining their way of life. Con- versely, the present system reflects, in a large measure, the acceptability to con- sumers of the system and their preferences among a large variety of services made available to them by competing retail firms that make up the system. In view of the demographic and technological changes taking place in the United States, it is rea- sonable to expect a further evolution in all sectors of the food industry, particu- larly the retail sector. The speed with which this change occurs will be deter- mined by the technical and economic ca- pacity of the food industry to stimulate and then fulfill their customers' desires for new products and services. The evolutionary processes in food re- tailing are well illustrated by events in the United States during the past 200 years. The colonial villages relied on bakers, butchers, and tea and spice mer- chants for items produced outside the home. Later, the itinerant merchant or peddler became the principal supplier of the farmers in the sparsely settled coun- try. As the rural population increased and became more concentrated, the general store type of operation became a feasible source of supply. A further concentration of population in towns and cities made possible the specialty food store — bakery, dairy, fruit and vegetable, and meat — a type of operation in which the individual was both the owner and operator. Today, one sees all types of retail food stores. Competition for urban customers has encouraged a reversion to the general store concept on a larger and more pre- tentious scale. In an effort to improve shopping arrangements, encourage "one- stop" shopping, and reduce unit selling costs, food retailers began offering a com- plete line of foods, a wide range of higher- margin household products, cosmetics, wearing apparel, drugs, and hardware. The widespread use of private auto- mobiles also has been a very important factor in expanding the area from which large retail stores can draw customers. The practice of selling a wide range of food and nonfood merchandise requires more floor space and equipment than is needed by the more conventional grocery store. These large combination stores, often departmentalized, are referred to as the "supermarkets" which now account for the largest percentage of all retail food sales. Zimmerman (1955) defines the super- market as "a highly departmentalized re- tail establishment, dealing in foods and other merchandise, either wholly owned or concession operated, with self-service sales, adequate parking space, and doing a minimum of $250,000 (since doubled) in business annually." The modern super- market embodies the accumulation of pre- vious technological innovations in food retailing and supplements them with im- proved techniques developed by other seg- ments of the nation's economy (Mueller and Garorian, 1961). Table 3 illustrates the declining im- portance of the more specialized-type out- let and the growing importance of the combination-type store. Total sales more than doubled for the combination stores from 1948 to 1963, while there was little or no sales increase for the specialized units. However, the number of specialty food stores decreased by only 18.2 per cent; the number of combination stores decreased by 34.8 per cent. As the combination-type store grew, both in sales and average store size, more small independent operators went out of business (appendix table A-2). These data also show that because of the large de- crease in the number of combination stores, the specialty-store group was a larger proportion of the total food store group in 1964 than in 1955. Nonetheless, total sales volume is of much greater eco- nomic significance than number, and illus- trates better the declining importance of the specialty food store. Figure 4 shows the relative growth in chain store outlets compared to that of the independents. While the number of chain stores increased by 1,960 stores be- tween 1955 and 1964, the number of all food stores decreased by 123,640. The trend towards larger food stores is shown in appendix table A-3. In 1953 food sales in the small-stores group constituted about 20 per cent of total sales and those in supermarkets were 48.3 per cent. In 1961, the last year for which comparable [12] Table 3 RETAIL ESTABLISHMENTS AND SALES, BY KIND OF STORES, SELECTED YEARS Kind of store Number of establishments Sales 1948 1954 1958 1963 1948 1954 1958 1963 350,754 23,920 4,517 13,482 27,165 19,500 7,917 13.658 279,440 22,896 4,458 13,136 20,507 19,034 8,132 13,777 259,796 23,844 4,339 12,689 17,593 19,235 NA 18,012 244,838 16,457 3,630 8,874 14,979 18,631 NA 12,024 thousand dollars Grocery and combination 24,729,717 1,641,087 132,331 394,602 586,592 722,761 308,336 092,438 34,420,764 1,943,969 184,148 484,503 567,955 862,290 479,787 752,439 43,696,343 2,327,038 193,748 505,355 527,752 904,981 NA 867,116 52,565,955 1,529,814 175,666 Fruit and vegetable 412,292 Candy, nut, and confec- 499,268 1,080,282 NA 815,909 Total 460,913 381,380 355,508 319,433 29,207,864 39,695,855 49,022,333 57,079,186 Source: U. S. Bureau of the Census, 1954, 1958, and 1963. 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 Fig. 4. Composition of total food stores. Source: Appendix table A— 2. data are available, the small-store group accounted for about 8 per cent of total sales and supermarkets for 70 per cent. A subsequent (1962) reclassification of stores into small (up to $150,000 in annual sales), superette ($150,000 to $500,000), and supermarkets (more than $500,000) places many of the former superette stores in the small-store class, thereby affecting total sales of all three groups. Despite this change in classification, the supermarkets held roughly 68 per cent of the total an- Table 4 GROWTH IN NUMBER OF SUPERMARKETS Year Number of supermarkets* Size definition of supermarket (minimum annual gross sales) 1952 1954 16,500 21,440 24,700 27,100 28,800 29,900 32,000 33,300 30, 100 27,125 28,400 30,900 dollars 375,000 375,000 375 000 1955 1956 375 000 1957 375 000 1958 1959 375,000 375,000 375 000 1960 1961 375,000 500,000 500,000 500,000 1962 1963 1964 * Progressive Grocer, 1955-1965, and 1963. [13] Table 5 NONFOOD SALES IN SUPERMARKETS AS A PERCENTAGE OF TOTAL SALES, 1958-1963 Item Health and beauty aids Housewares Magazines and books. . . Soft goods Toys Phonograph records Stationery All nonfoods Percentage of total sales 1958 2.28 .74 .21 .26 .10 .10 3.77 1959 2.38 .81 .22 .30 .11 .11 .09 4.02 1960 1.47 .84 .23 .33 .13 .13 .10 4.23 1961 2.57 .88 .26 .36 .15 .13 .11 4.45 1962 2.75 .93 .27 .40 .15 .13 .12 4.75 1963* 2.29 .83 .43 .56 NA NA t NA * Food chains only. t Included in "Magazines and books." Source: Progressive Grocer, 1952-1964. nual grocery store sales. This increasing importance of the large supermarket is even more obvious in table 4. As mentioned, inherent in the present- day supermarket is a characteristic of the old general store — that of selling various nonfood products, which amounted in 1962 to 4.75 per cent of sales, with a con- tinuing upward trend indicated (table 5). The most common nonfood products sold Table 6 SALES OF CHAIN AND INDEPENDENT STORES, BY SIZE GROUP FOR SELECTED YEARS Size of store Sales per store 1953 1958 1959 1960 1961 1962 dollars Small* 70,000 24,945 25,110 272,674 141,210 157,224 976,699 779,070 886,772 62,500 17,949 18,040 297,297 191,964 198,492 1,143,791 1,054,795 1,096,667 62,500 17,448 17,541 306,452 211,321 216,578 1,141,975 1,031,646 1,087,500 62,500 25,298 25,386 300,000 204,301 208,405 1,150,585 1,018,519 1,086,336 NA NA 26,541 NA NA 217,153 NA NA 1,269,103 82,143 58,371 Total 58,738 Superette* 301,786 274,490 Total Supermarket* 277,289 1,501,701 1,292,683 Total 1,401,292 95,739 169,474 179,412 202,308 218,850 238,237 * 1962: Small, under $150,000 annual gross sales; Superette, $150-$500,000; Supermarket, $500,000 and more. Prior to 1962: Small, under $75,000; Superette, $75-$375,000; Supermarket, $375,000. Source: Appendix table A-3. [14] Table 7 UNITED STATES POPULATION, GROCERY STORES, AND POPULATION PER STORE FOR SELECTED YEARS Year Popula- tion Grocery stores Popula- tion per store Increase from 1940 1940 million 132.1 139.9 151.7 175.4 177.1 180.0 183.0 185.8 thousand 444.9 397.0 400.7 285.0 280.5 260.1 248.8 235.9 297 352 378 615 631 692 736 788 index 100 1945 119 1950 127 1958 207 1959 212 I960 233 1961 248 1962 265 Source: U. S. Bureau of the Census, 1963, and Progressive Grocer, 1952-1964. in today's store are classified as "Health and Beauty Aids," including such high- margin items as cosmetics, toiletries, and drugs (Progressive Grocer, 1959). As early as 1957, more than half of the existing food stores sold drugs and cos- metics, magazines, housewares, stationery, toys, beer, hardware, store-baked goods, dietetic foods, soft goods, greeting cards, books, and garden supplies. To accommo- date this increase in the items offered, the stores have continually grown larger. The average undeflated dollar sales per store almost doubled in each of the three classi- fications during the past 10 years (table 6) 2 , and the average population per store rose from 297 persons in 1940 to 788 per- sons in 1962 (table 7). THE RETAIL FOOD CHAINS Because of the controversy surrounding the rising economic power of the national food chains and their policies and oper- ating practices (Ginsburg, 1966), this analysis emphasizes the position of the chains in the structure of the industry. The term "retail food chain" used in this report refers to a firm operating more than one retail outlet, even though the United States Census classification in- cludes firms with 1 1 or more outlets. The retail firm may or may not be integrated back to its source of supply of merchan- dise. The corporate chain, which is the most common form of ownership, is a centrally owned and controlled, multi- unit firm, performing both the buying and selling functions in retailing. Such a firm may be local, regional, or national, depending on its area of operation. Neither independently owned, single- store firms which associate themselves to- gether for cooperative buying, advertis- ing, and public relations, nor wholesaler- sponsored buying groups are considered as chains in our study These groups are referred to as "affiliated" stores. Single- store firms without such a relationship with wholesalers or other retailers are re- ferred to as "unaffiliated" stores. Table 8 CHAIN EXPANSION PATTERN, BY SIZE GROUPS, 1953-1964 Number of stores per Number of chains in group Change in number Number of stores operated by groups Change in number Average number of stores per chain chain 1953 1964 1953 to 1964 1953 1964 1953 to 1964 1953 1964 2-3 2,013 587 171 85 23 2,509 648 201 83 30 +496 + 61 + 30 - 2 + 7 4,654 3,040 2,620 3,449 14,115 5,650 3,448 2,916 4,178 17,058 + 996 + 408 + 296 + 729 +2,943 2.31 5.18 15.32 40.58 613.70 2.25 4-9 5 32 10-25 14 51 26-99 50.34 100 and more. . . . 568.60 Source: Mooney, 1964, p. 38. The retail food price level rose about 8 per cent during this period. [15] Table 8 indicates that the 2- and 3-store chains increased in number more than other size groups between 1953-1964. The 2- and 3-store firms and the 26- to 99-store firms had the greatest percentage increase in the number of store outlets. However, the chains with more than 100 stores had the greatest increase in actual store out- lets. Data from appendix table A-3 and table 9 show that the chains have been steadily improving their position relative to the independents as far as the number of stores operated by the two groups is concerned. Even though chains operate only a small part of all grocery stores, their 8.7 per cent of the total grocery and combination stores now accounts for 41.2 per cent of total sales for that group. Table 9 RELATIVE IMPORTANCE OF CHAINS VERSUS INDEPENDENT GROCERY AND COMBINATION STORES Percentage of total grocery and combination stores 1953 1958 1959 1960 1961 1962 Chain Independent . . . 5.6 94.4 6.8 93.2 7.0 93.0 7.9 92.1 NA NA 8.7 91.3 Percentage of total grocery and combination store sales 1953 1958 1959 1960 1961 1962 36.0 64.0 38.7 61.3 38.8 61.3 38.9 61.2 NA NA 41.2 Independent . . . 59.8 Source: Appendix table A-3. Table 10 GROCERY SALES BY MAJOR CHAINS AS A PERCENTAGE OF ALL GROCERY STORE SALES FOR SELECTED YEARS Year A&P Safeway Kroger Amer- ican Total of 4 largest Second 4 chains Next 12 chains Total of top 20 chains Other chains All chains ■per cent 1929 14.3 13.7 13.4 14.4 12.1 9.9 10.3 10.0 10.3 11.4 11.5 11.7 12.0 11.2 11.7 11.9 11.8 11.7 11.4 11.2 11.4 9.4 8.9 8.9 8.5 2.9 4.6 4.8 4.9 5.0 4.4 4.1 4.6 4.6 5.0 5.2 4.8 4.6 4.8 5.1 5.2 5.2 5.2 5.1 5.0 5.0 4.4 4.3 4.3 4.1 3.9 3.6 3.1 3.2 3.2 3.2 3.3 3.2 3.1 3.4 3.3 3.3 3.3 3.3 3.3 3.2 3.2 3.3 3.8 3.9 4.0 3.5 3.3 3.1 3.2 2.0 1.8 1.7 1.8 1.7 1.6 1.7 1.6 1.7 1.7 1.7 1.7 1.7 1.7 1.6 1.8 1.8 1.8 2.0 2.0 2.0 1.6 1.7 1.7 1.7 23.1 23.7 23.0 24.3 22.0 19.1 19.4 19.4 19.7 21.5 21.7 21.5 21.7 21.0 21.7 22.0 22.0 22.0 22.3 22.1 22.4 19.0 18.2 18.1 17.4 3.6 3.6 3.2 3.4 3.4 3.0 3.1 3.2 3.5 3.7 3.8 3.9 4.1 4.1 4.2 4.5 4.8 5.3 5.6 5.7 6.0 5.1 5.1 5.2 5.5 3.0 2.8 2.6 2.5 2.7 3.1 3.2 3.3 3.3 3.7 4.1 4.1 4.5 4.8 5.1 5.8 6.6 6.8 7.2 5.8 6.0 6.1 6.3 29.2 30.5 28.0 24.6 25.2 25.7 26.4 28.5 28.8 29.1 29.9 29.2 30.4 31.3 31.9 33.1 34.5 34.6 35.6 29.9 29.3 29.3 29.2 8.2 8.6 9.5 8.2 9.3 7.6 7.7 9.3 8.8 9.7 8.6 6.9 7.2 7.7 8.2 7.9 7.5 7.4 7.4 6.1 8.0 8.2 8.3 1935 1940 37.4 1941 39.1 1942 37.5 1943 32.8 1944 34.5 1945 33.3 1946 34.1 1947 37.8 1948 37.6 1949 38.8 1950 38.5 1951 36.1 1952 37.6 1953 39.0 1954 40.1 1955 41.0 1956 42.0 1957 42.0 1958 43.0 1959 36.0 I960 37.2 1961 37.5 1962 37.5 Source: Mueller and Garorian, 1961, and Progressive Grocer, 1952-1964. [16] IUU 95 - 90 85 Chains 80 75 70 65 - 60 7/ f /////// f /777J777777TTT ^ 55 o /////// Unaffiliated Independents ////// t o !: so - £ 45 I 40 35 / Lr 30 / ^^ 25 Affiliated Independents 20 15 10 5 1 1 1 1956 1958 1963 Fig. 5. Distribution of total U. S. grocery store sales. Source: Progressive Grocer, 1952-1964. Among the nation's largest food chains, A & P has been and still is the largest (table 10). This dominance is no longer so noticeable, its share having decreased from 14.3 per cent in 1929 to 8.5 per cent in 1962. Safeway 's share rose from 2.9 per cent in 1929 to 4.07 per cent in 1962. The share of the total sales of the four largest chains decreased from 23 per cent to 17.4 per cent during that period. The ninth through the twentieth largest chains were the most active with respect to fluctua- tions in their sales levels, having more than doubled their share of total grocery sales from 1941 to 1962. When comparing chain store food sales with those of the affiliated and unaffili- ated independents, one finds that the sales of affiliated independent stores have in- creased proportionately, even more than chain store sales. The share of the un- affiliated independents has continued to decline in relation to the other groups (figure 5). Growth characteristics of chains The 30 major food chains selected for this study accounted for about one-third of all grocery store sales in 1963. Table 11 ranks them according to their annual sales within the chain store group for selected years, beginning in 1940. Table 12 shows the relative size of the 30 major chains and the estimated per- centage of all sales made by all food stores during 1963. Growth Indices. As stated, the 30 chains were divided into three size groups — small, medium, and large — of 10 chains each. Both the net sales and the number of stores for each chain for the 1953-1963 period were converted into indices, with [17] Table 11 DISTRIBUTION OF SALES OF 30 MAJOR FOOD CHAINS AS A PERCENTAGE OF TOTAL GROCERY STORE SALES FOR SELECTED YEARS Firm 1940 1955 1958 1959 1960 1961 1962 1963 per cent The Great Atlantic and Pacific Tea Company, Inc Safeway Stores, Inc The Kroger Company American Variety Stores National Tea Company Food Fair Stores, Inc Winn Dixie Stores, Inc First National Stores The Grand Union Company Colonial Stores, Inc Jewel Tea Company, Inc Wrigley Supermarkets Loblaw, Inc Stop and Shop, Inc Penn Fruit Company, Inc Thirftimart, Inc Red Owl Stores, Inc H. C. Bohack Company, Inc Lucky Stores, Inc J. Weingarten, Inc Mayfair Super Markets, Inc Thorofare Markets, Inc Fisher Foods, Inc Purity Stores, Inc Market Basket Shopping Bag Supermarkets Daitch Shopwell Alpha Beta Stores Food Mart, Inc Marsh Supermarkets, Inc 35.8 12.8 8.3 4.0 2.0 .9 .4 4.6 1.1 1.5 .9 .07 .7 .7 NA .2 .4 .4 NA NA NA NA NA NA NA NA NA NA 29.03 13.55 8.55 4.38 4.04 2.88 1.83 3.30 1.54 2.66 1.93 NA 1.33 NA .76 .37 .79 .95 .31 .55 .47 .57 .60 .65 .45 .34 .32 .12 .13 26.6 11.6 9.3 4.6 4.1 3.7 3.3 2.9 2.6 2.3 2.3 2.0 1.5 1.0 26.16 12.24 9.84 4.49 4.26 3.77 3.42 2.73 2.59 2.31 2.28 1.86 1.46 1.00 .86 1.02 .82 .92 .64 .79 .57 .51 .51 .49 NA .43 NA .30 .33 23.81 11.65 8.82 4.63 4.04 3.64 3.40 2.48 2.95 2.10 2.32 1.64 1.56 1.13 .79 .79 1.07 .81 .90 .65 .96 .58 .50 .48 .49 NA .42 NA .28 .37 23.79 11.51 8.36 4.59 4.03 3.81 3.48 2.43 2.74 2.00 2.31 1.73 1.40 1.34 .81 1.25 .75 .91 .63 1.16 .58 .49 .48 .54 NA .38 NA .28 .36 22.64 10.84 8.41 4.47 4.23 3.99 3.34 3.07 2.77 1.94 2.39 1.68 2.15 1.37 .75 1.21 .76 1.00 .59 1.25 .38 .45 .46 .52 NA .46 NA .29 .35 22.01 10.98 8.71 NA 4.38 4.16 3.45 3.09 2.61 1.90 2.56 .67 1.24 .79 1.07 .59 NA .60 .41 .45 NA NA .52 NA .35 Source: Progressive Grocer, 1952-1964. 1953 as base year. These indices were used to calculate an "expansion ratio" which was obtained by dividing the sales index of a firm for a given year by its index of the number of stores for the same year. The resulting "expansion ratio" becomes a rough measure for the proportional increase in annual sales per store over the observation period. Figure 6 shows a fairly consistent in- crease in the average annual net sales for Group I for the period 1953-1963. After 1956, the increase for Colonial Stores, Inc., was very small; however, the most revealing fact is that an average of the five largest chains (1-5) had sales expan- sion indices lower than the Group I aver- age. This indicates that, with respect to sales, the nation's five largest chains were maintaining a lower sales growth pro- portional to the average of Group I. It must be noted, however, that a base pe- riod analysis gives an advantage to firms with a smaller base and discloses nothing about increases in absolute sales volume. Food Fair Stores, Inc., and Winn Dixie Stores, Inc., experienced the greatest rela- tive growth in the number of food stores in Group I (figure 7). The number of stores for the two largest chains changed very little, while the absolute number de- creased for three of the ten chains. These [18] Table 12 PERCENTAGE SHARE OF ALL 1963 GROCERY STORE SALES FOR EACH OF THE 30 MAJOR CHAINS Chain Percentage of U. S. grocery store sales (1963) Chain Percentage of U. S. grocery- store sales (1963) A&P Safeway Kroger American National Food Food Fair. . . . Winn Dixie. . . First National Grand Union. Jewel Tea Loblaw Colonial Wrigley Stop & Shop.. Red Owl 4.12 3.27 1.68(1962) 1.64 1.56 1.29 1.16 .81 .71 .62 .62 .47 Mayfair Markets Lucky Stores Thriftimart Bohack Penn Fruit Weingarten Thorofare Market Basket. . Daitch Shopwell Purity Stores Fisher Bros Food Mart Marsh, Inc Shopping Bag. .. Alpha Beta .47(1962) .40 .34 .29 .25 .22 .22 .20(1962) .19 .17 .15 .13 .12 NA NA Source: Progressive Grocer, 1952-1964. store expansion data do not distinguish between new stores and acquired or pur- chased stores, nor do they show the closing of old, small, or inefficient stores. As mentioned, those chains in the eleventh through twentieth categories seemed to be the most dynamic (figure 8). Although the Group II average net sales index moved steadily upward after 1953, the sales of individual firms within Group II diverged considerably from the average. Lucky Stores, Inc., and Mayfair 550 500 450 -400 a. £350 ^ 300 z ^250 4> c .2 200 B D ^150 100 50 (1) The Great Atlantic and Pacific Tea Company , Inc. (2) Safeway Stores, Inc. (3) The Kroger Company (4) American Variety Stores (5) National Tea Company (6) Food Fair Stores, Inc. (7) Winn Dixie Stores, Inc. (8) First National Stores ( q ) The Grand Union Company (10) Colonial Stores, Inc. — — average 1953 1954 1955 1956 1957 1958 1959 I960 1961 1962 1963 Fig. 6. Expansion index of net sales, Group I chains. Source: Appendix table A— 4. [19] 500 450 400 G a. 3 J 350 V) I 300 >/> -5 I 250 z o J 200 JZ I 150 c a a. ■" 100 50 - (1)A&P (6) Food Fair (2) Safeway (7) Winn Dixie (3) Kroger (8) First National (4) American (9) Grand Union (5) National Tea (10) Colonial average - j (6) - / ^^ {1) .. 1 (9) --** ' — - _ ■ ^) ^==-=* ^^ ■*=-- (10) ~ ^(3) ( 2 ) 1 1 1 1 1 1 *^ (4) (8) 1 1 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 Fig. 7. (above) Expansion index of number of stores, Group I chains. Source: Appendix table A-4 Fig. 8. (below) Expansion index of net sales, Group II chains. Source: Appendix table A-4. 650 600 550 500 ^450 OL 3 o 400 £ 350 «> z *° 300 250 - 200 150 100 50 - (11) Jewel Tea Company, Inc. (12) Wrigley Supermarkets (13) Loblaw, Inc. (14) H. C. Bohack Company, Inc. (15) Penn Fruit Company, Inc. (16) Red Owl Stores (17) Lucky Stores, Inc. (18) J. Weingarten, Inc. (19) Mayfair Super Markets, Inc. (20) Thorofare Markets, Inc. — — — — • average Am .(19) - (13) - y^'^ / 1^ (20) r*/06) 7^(12) ^-(11) (18) "(15) ^ — - . (14) 1 1 1 1 iiii 1 1 1 1953 1954 1055 1956 1957 1<»58 1959 I960 1961 1962 1963 [20] 1953 1954 1955 1956 Fig. 9. (above) Expansion index of number of stores, Group II chains. Source: Appendix table A-4. Fig. 10. (below) Epansion index of net sales, Group III chains. Source: Appendix table A-4. 700 (21) Fisher Foods, Inc. 650 (22) Purity Stores, Inc. (23) Thriftimort, Inc. (24) Market Basket (29) 600 — (25) Shopping Bag Superma (26) Daitch Shopwell rkets 550 (27) Alpha Beta Stores (28) Food Mart, Inc. (29) Marsh Supermarkets, 1 rtc. / (26) // (28) I 450 (30) A. J. Bay less Markets , Inc. / — average / /^ (23) -5 400 " 350 o i 300 c o 1 250 o a. ^ 200 150 /^^^"^ JLt^^^^ / J27) ^<^Z^^ (30) ,' -~~ (24) . (21) -(77, 100 Z^^^ (25) 50 i i i i , i I l i i 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 [21] 600 (21) Fisher Bros. (26) Daitch Shopwell 550 (22) Purity Stores (27) Alpha Beta (23) Thriftimart (28) Food Mart (24) Market Basket (29) Marsh, Inc. 500 - (25) Shopping Bag (30) Bayless ^450 1 a. 2 400 o average 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 Fig. 11. Expansion index of number of stores, Group III chains. Source: Appendix table A-4. Super Markets, Inc., experienced the greatest proportional growth; H. C. Bo- hack Co., Inc., experienced the least. Chains 13, 17, and 19 seem to have ex- panded sales at an increasing rate, while the remaining firms' expansion rate tapered slightly in more recent years. Except for Bohack, Group II chains in- creased their number of stores during the observation period. Chains 15, 17, and 19 Fig. 12. Expansion index of net sales and number of stores, by group. Source: Appendix table A-4. 400 350 1 300 o c5 -£"250 01 -a 1200 o Ui c 1 150 UJ 1 100 < 50 1953 Net Sales Number of Stores m i 1954 1955 1956 1957 1958 [22] 1959 1960 1961 1962 1963 were near the top of the expansion, while 14 was at the bottom (figure 9). Group III showed a less erratic expan- sion of net sales than Group II. Never- theless, there were substantial differences among individual chains within Group III. As shown in figure 10, the ten chains seem to fall into three general growth patterns. Chains 21 and 22 had very little proportional growth in net sales, chains 24, 27, and 30 experienced nominal growth and the remaining four chains had net sales growth far above the group aver- age. For all three groups, 1955-1959 was the period of greatest sales expansion and 1960-1962 the period of greatest irregu- larity. In figure 11, the expansion in number of stores generally falls in the same three growth patterns mentioned above. How- ever, chains 21 and 22 had very little growth of net sales and actually had a decline in the number of stores. This raises the question of whether the small proportional sales growth was the cause or effect of the decrease in the number of stores. The decline in store numbers seems to have picked up momentum during the latter part of the observation period. In view of this, one would suspect that the relatively small sales gain in the earlier years actually caused these chains to close, merge, or sell some of their old or unpro- ductive operations. Figure 12 shows the average expansion index of net sales and number of stores by group. In almost every year of the ob- servation period, the proportional expan- sion rates for the two largest groups were lower than those of the smaller group. To reiterate: The expansion ratio is the quotient of the net food sales index di- vided by the number of stores index for a given firm. Hence, the quotient will be affected by disproportionate changes in the divisor and dividend. Figure 13 shows the 1963 expansion ratios for all 30 of the nation's larger food chains. The aver- age expansion ratio for all 30 chains is approximately 1.6 which, loosely inter- Fig. 13. Expansion ratio, for all chains, groups, and individual chains, 1963. Source: Appendix table A-5. 12 3 4 5 6 7 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 Number of Chains Largest > Smallest [23 1 preted, indicates that average sales per store in the 30 chains have increased 60 per cent from 1953 to 1963. This figure is consistent with the increases in the aver- age physical size of food stores and average number of items sold per store. For ex- ample, over the same period, the average number of items sold per store increased about 45 per cent and store size increased about 35 per cent. The average expansion ratio of Group I was above that of Group II, but both were below Group III. In general, it seems that the smaller chains made the greatest advances in sales per store, while Group II had the lowest increase in sales per store. This should not seem overly sur- prising in view of the great increases in both sales and number of stores in Group II. There is little or no similarity in the expansion ratios among chains of like size. Group II has the greatest range of variation, as would be expected, in view of the dynamic characteristics of this group. Effects of growth In view of the different rates of growth of the three groups of chains, the re- searchers turned their attention to the relationship between growth and the fol- lowing important financial items: • Profits, • Merchandise inventory, • Fixed assets (buildings and equipment), • Accounts payable, and • Net worth. The graphic presentation in figures 14 through 18 depicts the effects of growth of chains by size groups on the financial factors listed above for the period 1953- 1963. Profits. Profit rates per dollar sales for each of the three size groups rose sharply from 1953 through 1955 and leveled off through 1957 (figure 14). A significant de- cline occurred in Group III chains from 1957-1961. A similar though less drastic decline affected Group II chains begin- ning in 1959. The larger chains in Group I, collectively, showed much less decline than the small and medium chains. The researchers do not have a complete ex- planation for the sharp decline in net profits in two of the groups. In their judgment, however, the pressure of rising labor costs, technological changes requir- ing greater capital investment, and price- quality competition for customers and sales volume, which prevented prices from going much above those of the Group I chains, are the main reasons for the de- cline. Merchandise inventory. For 1953, and from 1957 onward, the Group II chains' merchandise inventory as a percentage of current assets averaged lower than the other groups (figure 15). A factor that could have affected the inventory policy of the Group II chains was their greater reliance on wholesalers to perform the merchandise storage function. On an aver- age, the 30 chains studied maintained a sales inventory ratio of 2 to 1 (see pages 26 to 29). Fixed assets as a percentage of total assets generally rose for all chain groups from 1953 to 1960-1961, and for groups I and III the trend was upward through 1963. The sharp drop of percentage in Group II is explained partly by a rela- tively greater increase in leasing rather than ownership of buildings used in the merchandising operations (figure 16). Accounts payable was the principal item of indebtedness in each of the chains studied. The rising importance of this method of financing expansion of food chains is shown in figure 17. After 1953, groups I and II began to rely more on open accounts as a source of credit financ- ing. Group III also sharply increased its use of this form of credit after 1957. Net worth. This marked increase in creditor financing by accounts payable should be considered in conjunction with the relative decline in net worth or owners' equities in relation to the total assets used by the business (figure 18). For Group II, a sharp downward change in owner financing began in 1955. A less drastic downward change occurred for Group III after 1959. Group I chains financed their expansion by increasing both forms of equities. The rapid growth of groups II and III chains would appear [24] 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 Fig. 14. (above) Net profit as a percentage of total sales. Source: Fairchild Publications, Inc., 1953- 1963. Fig. 15. (below) Merchandise inventory as a percentage of current assets. Source: Fairchild Publica- tion, Inc., 1953-1963. 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 [25] 39 38 - ^1 37 / / \ / 36 1 / \ A / / / * /*■ 35 " / 1 r ^' 34 / \ / // J A !/:j 33 / if / /I / A 32 - / / 31 // /7 30 ^/ o1 1 1 1 i i i i i i i 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 Fig. 16. Fixed assets as a percentage of total assets. Source: Fairchild Publications, Inc., 1953-1963. to be supported by liberal credit from grocery wholesalers and /or brand-product manufacturers who were greatly depend- ent on small- and medium-size chains for access to consumers. Financial interrelationships The financial reports of the 30 chains studied provide one basis for a better understanding of certain management practices and their results. For this phase of the analysis, the three-group classifica- tion of chains (I, II, and III) is again used. While the preceding discussion con- sidered financial trends over time, the fol- lowing appraises certain financial inter- relationships which might exist within a chain or group of chains, irrespective of time. For example, this analysis attempts to determine what relationship one finan- cial factor has with another and how this relationship differs among chains and groups of chains. Regression analyses (appendix C) were used to find the following relationships: • Net dollar sales as a function of the number of stores, • Net dollar profit as a function of the number of stores, • Inventory as a function of net sales, • Current assets as a function of inven- tory, • Fixed assets as a function of total assets, and • Total liabilities as a function of ac- counts payable and current liabili- ties. [26] 74 - ,m 73 / 72 - 71 70 — A ■' 69 - / \ / 68 67 66 / / ■J V I 65 64 / / c 63 u 62 LA 0> tL 61 60 59 58 57 56 55 54 53 A / \ * /* s s s s *-- XT V r V// 52 6 1 1 1 1 i i 1 1 1 1 1 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 Fig. 17. (above) Accounts payable as a percentage of current liabilities. Source: Fairchild Publications, Inc., 1953—1963. Fig. 18 (below) Net worth as a percentage of total assets. Source: Fairchild Publications, Inc., 1953—1963. 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 [27] The 30 chains. A condensed presenta- tion of results from the analysis for the 30 major retail food chains shows the follow- ing: • As expected, a statistically strong, positive relationship was found be- tween net sales and number of stores. Excluding the effects of loca- tion, a chain could expect on an average a total net sales increase of slightly more than $1 million by add- ing one retail outlet. • Also, as expected, a positive rela- tionship existed between net profit and number of stores. By increasing the number of stores by one, a chain could expect an increase in total net profit of approximately $11,000. 3 Paragraphs 1 and 2 tell us that for every new store, one could expect about 1.1 per cent of net sales to be retained as net profit (11,000 -f- 1,- 000,000). • If net sales were to increase $1 mil- lion, it would be expected that the necessary inventory would increase by $500,000. This indicates that a 50 per cent backup in inventory is necessary for each dollar's worth of product sales. • Associated with a $1 million increase in inventory, there would be a $1.5 million expected increase in current assets. This indicates the predomi- nance of inventory in that current asset group. • For a $1 million increase in total assets, one would expect a $390,000 increase in fixed assets. These results are consistent with figure 8 which shows fixed assets approaching 40 per cent of total assets. Unfortunately, the data do not disclose any informa- tion regarding the recent "buy-build- lease-back" trend used by chains to reduce the large amount of invest- ment in fixed assets. • Associated with a $1 million increase in accounts payable, there would be a $1.65 million increase in total lia- bilities; and for a $1 million increase in total current liabilities, total lia- bilities would increase by $1.95 mil- lion. This latter comparison indicates that short-run debt, other than ac- counts payable, fluctuates more ex- tensively than does accounts payable. The three groups. A comparison of the three groups on the six points men- tioned shows the following: • Net sales were more responsive to store numbers in larger chains than in smaller ones, indicating that large chains are catching up with small chains in terms of average store size. • Net profit responded to store num- bers at approximately the $10,000 per store level, except for the smallest size chain group. However, store numbers, accounted for very small amounts of variation in net profits for groups II and III. • The inventory response to net sales changes was relatively similar for all three size groups, coefficients ranging from .05 to .08. 4 This is reasonable because the "inventory margin of safety" should be about the same for all groups, considering the availa- bility of products and the time re- quired for delivery. • The relationship between current assets and inventory change was con- sistent among the three groups, with coefficients of from 1.52 to 1.64. In all three groups, more than 90 per cent of the variation was explained by the equation. This relationship is to be expected since inventory accounts for a major part of the total current assets of retail food firms. • In the fifth equation also, consistent results were obtained from the analy- sis of the three groups. Fixed assets were shown to change at about one- third the magnitude of total assets, and about 90 per cent of the varia- tion was explained by the factor con- 3 Slightly overestimated because of the approximately 10 per cent increase in the retail food price index during the observation period 1953-1963. 4 Coefficient statistics such as these indicate the extent and direction of a response by the dependent factor, given a one-unit increase in the size of the independent factor. For example, given a $1 million increase in net sales, it can be expected that inventory requirements will increase (positive coefficient) by $500,000 to $800,000. [28] sidered. Again, this correlation fails to distinguish differences among firms with respect to the ownership or lease method for obtaining build- ings and equipment. • In the sixth equation, different re- lationships were found for the three groups; however, in most cases the results were statistically significant. This indicates that total liabilities responded quite differently to ac- counts payable and total current lia- bilities in the three size groups. Even though there was considerable varia- tion among groups I, II, and III, the results indicated a heavy reliance on open account purchases of merchan- dise for resale by all the chains. The individual chains. As mentioned, Group II was the most expansion ori- ented. For this reason, each of the 10 chains in this group was analyzed sepa- rately, using the same six equations. A brief glance at the regression coefficients is sufficient to show the large extent of variations among chains' operations. The magnitude, as well as the direction of the relationships, varied considerably among chains. Relationships among equations number 3, 4, and 5 were consistent. Equa- tion 3 shows that each chain's inventory responded similarly to net sales; equation 4 shows an impressive relationship be- tween current assets and inventory; and equation 5 again points out the 30-40 per cent responsiveness of fixed assets to total assets. It is worth noting that in the second relationship, the net profit of some of the Group II chains responded favora- bly to increased store numbers, while others did not. This suggests a great dif- ference in individual new-store profita- bility. INDUSTRY PERFORMANCE AND PUBLIC POLICY Methods used The preceding discussion shows the more favorable earnings record of the chains in Group I. The smallest chains, Group III, had the least favorable earnings rec- ord. On the other hand, the medium-sized chains, Group II, were expanding faster in terms of both sales and outlets. It could be that the management objective of firms in both groups II and III was to compete for sales volume by means of prices and/ or services. Here are three management practices that might keep groups II and III earnings below those of competing chains in Group I: • "Follow-the-leader" pricing, even though the delivered on-the-shelf cost of products might exceed those of chains in Group I. This would result in a lower net, assuming handling costs are comparable. • Lower markup and prices on mer- chandise which would yield a lower net, assuming handling costs are com- parable. • "Follow-the-leader" on prices for products, plus extra customer services which add to the retailers' handling costs and thereby reduce net earnings. Undoubtedly, the various firms use all three of these practices to meet competi- tive conditions relative to specific products or product groups. For example, a firm might have a "follow-the-leader" pricing system for meat and a low markup and pricing policy for fresh fruits and vege- tables. The lower prices would be designed to draw customers and would constitute a form of sales promotion cost. This is consistent with the objective of a firm that wishes to expand or perhaps hold its busi- ness, the ultimate goal being to achieve higher profits by means of the scale econ- omies of buying and selling merchandise. Growth patterns The total sales of the food chains have continued to rise faster than nonchain retail firms. This increase began during the mid-1940's and gained momentum with the removal of price controls and ration- ing in 1946 (figure 19). This trend toward concentrating more sales in fewer retail food firms has brought forth charges of an unwarranted concentration of economic [29] 44 Price controls and rationing removed 1 I l_l L 1929 1935 1940 1945 1950 1955 1960 1962 Fig. 19. Chain store sales as a percentage of all grocery store sales for selected years. (Chain stores are considered firms with 1 1 or more outlets.) Source: Progressive Grocer, 1952—1964. power that has been, and will continue to be, harmful to our nation's economy (U. S. Federal Trade Commission, 1960, p. 1.) Let us examine the major aspects of the al- legation. Although it is true that the corporate chains are rising in the grocery business, additional chains entered the market be- tween 1953 and 1964 and are competing successfully for a greater share of the total grocery store business (table 9). For ex- ample, the 1962 sales of each of the two largest chains, A 8c P and Safeway, propor- tionately were below their 1954 share of the total. Table 13 and figure 20 indicates the rise in the relative importance of the small chains within the group of 30 firms under study. The substantial increase in the number of small chains (table 8) sug- gests the absence of an effective barrier to entry in food retailing (Mooney, 1964). The growth pattern of chain sales as a percentage of all grocery store sales is shown in figure 19. The change in the dis- tribution of sales among the 30 large chains, according to size, is brought out in table 13 and figure 20. According to the data that could be obtained, the Group II and III chains are growing relatively faster than the Group I chains. Should this trend continue, there would be a virtual elimination of the very small chains now represented by Group III. In an effort to get some further statistical verification of the probability that smaller chains will continue to move into a higher sales level and achieve a greater relative share of the total sales of the 30 chains under study, the authors experimented with various tech- niques, one of which was the "Markov Chain" procedure, which is discussed in appendix D. The calculations also pre- sented in appendix D and graphically in figure 21 indicate the following: [30] Table 13 CONCENTRATION OF NATION'S 30 MAJOR RETAIL FOOD CHAINS Chain Average total net sales (1953-1963) $15,389,399 Total net sales (1962) $19,039,659 Total net sales (1954) $11,167,604 Percentage of total Accumulated total Percentage of total Accumulated total Percentage of total Accumulated total per cent 1 30.35 30.35 27.51 27.51 35.72 35.72 2 14.40 47.75 13.17 40.68 16.24 51.96 3 10.63 55.38 10.22 50.90 9.93 61.89 4 5.17 60.55 5.42 56.32 5.41 67.30 5 4.88 65.43 5.13 61.45 4.66 71.96 6 4.10 69.53 4.84 66.29 3.96 75.92 7 3.53 73.06 4.04 70.33 3.12 79.04 8 3.49 76.55 3.72 74.05 3.12 82.16 9 2.76 79.31 3.36 77.41 2.18 84.34 10 2.66 81.97 2.90 80.31 2.04 86.38 11 2.61 84.58 2.61 82.92 1.81 88.19 12 2.18 86.76 2.60 85.52 1.49 89.68 13 1.85 88.61 2.04 87.56 1.19 90.87 14 1.21 89.82 1.50 89.06 1.16 92.03 15 1.09 90.91 1.46 90.52 .96 92.99 16 .94 91.85 1.22 91.74 .84 93.83 17 .93 92.78 1.09 92.83 .79 94.63 18 .93 93.71 .92 93.75 .77 95.39 19 .90 94.61 .90 94.65 .63 96.02 20 .71 95.32 .70 95.35 .56 96.58 21 .65 95.97 .63 95.98 .55 97.13 22 .64 96.61 .56 96.54 .52 97.65 23 .62 97.23 .55 97.09 .49 98.14 24 .57 97.80 .54 97.63 .43 98.57 25 .48 98.28 .46 98.09 .36 98.93 26 .47 98.75 .45 98.54 .34 99.29 27 .32 99.07 .42 98.96 .24 99.51 28 .32 99.39 .36 99.32 .23 99.74 29 .32 99.71 .34 99.66 .14 99.88 30 .29 100.00 .32 100.00 .13 100.00 preceding analysis is that chain firms will tend to become larger. In view of this tendency and the allegations in the U. S. Federal Trade Commission Resolution of October 9, 1958, that the concentration of economic power in the food industry may restrict competition by collusive price ac- tion and unfair competitive methods (U. S. Federal Trade Commission, 1960), the probable growth pattern deserves further attention. Growth in the size of food chains has been accomplished by means of (1) inter- nal growth — a reinvestment of earnings, expansion of borrowings by the firm, and/or sale of stock; (2) purchase or ac- quisition of other firms; and (3) merger. The usual purpose of this drive toward 5 Small-chain entries into the market during this period are possible, but we are unable to analyze such movements in our "closed market" study (see appendix F). • Smaller retail firms will tend to in- crease their relative share of the total sales of the 30 chains. • Larger firms can be expected to lose some of their relative share of the sales of the group of 30 chains. • Both the small and large chains that were operating in 1954 will move to- ward or into size classification E which, relative to total sales, is smaller than the largest size group F. • Given the assumptions on which the projection is based, the 30 chains move toward a structural equilibrium. In other words, smaller firms will be- come larger and larger firms will be- come relatively smaller. 5 The most obvious deduction from the [31] Number of Chains Largest > Smallest Fig. 20. Retail food chain concentration among 30 major chains. The 45° line represents an industry of 30 firms in which every firm is of equal size, i.e., 5 per cent of the firms have 5 per cent of total market sales, 10 per cent of firms have 10 per cent of market sales, etc. Source: Table 13. [32] 2b 24 23 22 21 20 19 18 Group Classification A .1000 - .2714 percent of total chain sales B .2715 - .7366 percent of total chain sales C .7367 - 1.9991 percent of total chain sales D 1.9992- 5.4256 percent of total chain sales E 5.4257 - 14.7251 percent of total chain sales F 14.7252 - 39.9639 percent of total chain sales 17 16 » 15 f 14 o 13 5 12 1 "E 11 z 10 9 8 7 - 6 _ — 1 5 4 - 3 2 _ A B C D E F A n B c D E F A B C D E F A B C D E F A B C D E F 1 — ~ r -i r — \ _c I ) I ) I ) Y V V ^-v— J I < ' \ 19 54 19 62 19 70 1978 1986 Fig. 21. Distribution of 30 major chains by size in 1954 and 1962, and projections by eight-year intervals through 1986. Source: Appendix D. "bigness" has been to improve the total net earnings of the firms through mass buying and selling of merchandise. Because one of the effects of mergers and acquisitions is to reduce the number of firms and ownership entities, such actions could be regarded as a method of reducing competition. Conversely, they could be looked upon as the only means of eco- nomic survival in an industry in which certain operating efficiencies, e.g., purchas- ing and advertising, are a function of sales volume. Even single-store grocery firms that do not want to integrate in an own- ership sense are often achieving the basic economies of mass buying and warehous- ing through retailer-owned cooperative as- sociations or wholesaler-sponsored buying groups. Internal growth is usually characterized by reinvesting earnings or borrowed funds in larger stores and other facility improve- ments. This usually forms a part of a general modernization and replacement program of the firm. Capital for these pur- poses is less frequently obtained from the sale of stock. Food firms have been seeking to in- crease their sales volume and net earnings or profits in a number of ways other than those discussed on page 29. Rapidly grow- ing chains often are attaining their goal of increased sales by expanding the num- ber of store outlets. Others are increasing the number of stores and the size of stores, either by modernizing old or building larger new stores. The well-established large chains appear to be going through a store modernization and consolidation process, the outcome of which seems to be a temporary reduction in the total num- ber of stores operated by a firm. In addi- tion to varying the size and/or number of stores, another approach to increasing sales and profits has been to find the most de- sirable locations for the stores operated by the firm. This has meant adjusting loca- tion plans to population movements. A [33] third approach has been to vary the num- ber and quality of services offered to con- sumers. This has been characterized by the development of small, convenience-type stores (superettes) featuring quick service and long store hours; and minimum serv- ice or "discount" stores." Effect of population changes on number and location of grocery stores The assumption that "trade follows peo- ple" is especially applicable to the busi- ness of supplying food to our population. The almost complete dependence of our people on retail grocery stores is indicated by the following discussion. The past 10 years have been character- ized by a rather steady increase in the United States population and a sharp de- cline in the number of retail grocery stores. Based on the available data, the authors estimate that the number of people per grocery store has increased from an aver- age 465 in 1955 to 828 in 1964. Demog- raphers currently predict an increase of 24,240,000 people by 1975. It also seems highly probable that grocery store numbers will continue to decrease; however, the rate of decline may taper off somewhat. 6 In general, therefore, we should be able to make some statements as to the future movement of the "population per store" statistic. As shown in figure 22, this statis- tic has increased in an almost linear fash- ion over the past 10 years. A linear extra- polation of this trend leads us to expect that in 1975 the average United States grocery store will be serving approximately 1,300 people. One's first impression would tempt him to deflate this estimate in view of this tapering off in the rate of decline of the number of grocery stores. However, this tendency is counterbalanced with the assumption that those new stores, which will be built in the future, are likely to be even larger in size than most current stores. Given the United States population pro- jection and the average size (1,300 persons 6 Past declines in grocery store numbers can be partially attributed to the business failures of the small "corner grocers." Since corner grocers are no longer so numerous, their decline in numbers will be of a smaller magnitude. 7 "Performance or service" is used here to indicate a measure of availability, quantity, and quality of food products currently available to the average customer in the average United States grocery store. [34] per store) of the retail grocery store in 1975, the estimated number of additional retail grocery stores which will have to be built in the United States over the next 10 years in order to maintain the current level of service or market performance equals 18,648. 7 This change is approxi- mately an 8 per cent increase over current store numbers. The typical procedure generally would be to apply the above technique to more specific areas such as a particular state or even a metropolitan area in order to ob- tain more applicable results. Unfortu- nately, this procedure cannot be followed because the data used in this study deal basically with industry, structural changes, and market characteristics on a national basis. Therefore, any attempt to apply the general conclusions and /or projections to a smaller geographic area can only result in reducing the validity of the end prod- uct. To illustrate further this graphical lim- itation, we shall choose a particular state and a metropolitan area, discuss its par- ticular and differentiating market charac- teristics, and then demonstrate the result- ing inconsistencies. Demographers predict that California can expect an increase in population of 4,590,000 people by 1975. By using the na- tional projection of expected store size, one could conclude that California would need approximately 3,500 additional gro- cery stores. However, the California retail food industry has experienced a growth pattern quite different from that of the nation as a whole. Since the 1940's, Cali- fornia has experienced a massive growth in its population. This population has spread itself over areas which were unin- habited desert a short time before. The re- tail food industry has been forced to fol- low this population movement by building a large number of stores. While the major- ity of other states have had a decrease in the number of grocery stores, California has experienced a 50 per cent increase in grocery store numbers over the past 10 1330 • 1250 ,' 1170 * * ,' 1090 * 101C + ' ~ 930 — s jK / • 850 y a 770 — o t 690 / - ^y — Projected 610 — >/ 530 - y 450 7 , , , , , 1 Fig. 22. Population per grocery store in U. S., 1955-1975. Source: Progressive Grocer, 1955-1965; Appendix table D-3; and Bureau of Census, 1965. years (see appendix table B-l). As a result, population per store has remained rela- tively constant at about 870. Therefore, if this present population to number of gro- cery stores ratio continues, California will require approximately 4,300 (as compared with 3,500) additional grocery stores by 1975. A similar inconsistency would occur if national data were applied to an ever smaller area such as the San Francisco- Oakland Metropolitan Area. While the estimated population increase of 1,890,000 people implies a substantial increase in store numbers, authorities have often de- scribed this particular area as "over-stored" and capable of meeting greatly increased consumer demands with the present store facilities. This further illustrates the need to analyze each market area individually, thereby avoiding the possible misfortunes resulting from the application of national projections. Conduct and performance of chains Mass distribution methods for food have been encouraged by technological ad- vances; higher capital requirements; higher salaries, wages, and fringe benefits; rising consumer disposable income; and population concentration. These forces also have been behind the mass production of food and nonfood products. In any case, mass distribution practices and chains seem to be interwoven. This implies a con- centration of economic power in the food field. The real economic and legal issue is how the chains use this power which comes from their dominance of sales. With the few exceptions which are a matter of court record, there is not much to support or refute any charge against the food chains of any overt misuse of their market power, either as buyers of goods to be resold to consumers or as retailers. The available statistical evidence shows that the medium and large retail chains are winning the competitive struggle for cus- tomers and sales. It does not show that competition is being eliminated. In fact, the research department of the largest se- curity investment firm in the United States rates only one firm's common stock — Great Atlantic and Pacific Tea Company — as a recommended high-quality, investment- type risk. All others are average or specu- lative-type investments (Merrill Lynch et al., 1966). The basics reasons for the low investment appraisal are keen competition and overexpansion. In view of the charges [35] that competitors are being eliminated by a misuse of economic power, the advice of the investment firm to its clients is inserted here. In relation to other segments of the econ- omy, the food-chain industry fared poorly in 1965; and indications are that business will continue to be characterized by in- tensely competitive conditions and pres- sure on profit margins during the current year. . . . Despite this relatively poor per- formance, food chains continue to expand outlets, causing overstoring in certain areas of the country and intensified price competition as the chains try to build ade- quate volume in the new units. In some areas, the situation has been further ag- gravated by discontinuing trading stamps and by converting many units to a "dis- count" image. Under existing conditions, many companies are having difficulty pass- ing on increases in wholesale costs of food, particularly meat. On the positive side, a number of com- panies continue to register sastisfactory earnings gains as a result of new-store ex- pansion, above-average population growth in their operating areas, and the ability to increase their share of markets by ef- fective merchandising. Conclusive evidence shows a growing concentration of sales in the retail chains, and also a corresponding concentration of buying power. However, this structural characteristic does not connote an arbi- trary use of market power to frustrate nor- mal marketing and pricing processes. It is possible that the large retail chains would be at a disadvantage during periods of short supply in obtaining the quantities of uniformly graded products to meet their demand. This has happened a number of times since 1941, particularly during the World War II period when price ceilings and product rationing were used. Many of the complaints about "unfair" buying practices and "market manipula- tion" originate with growers and shippers of fresh fruits and vegetables. Because of repeated complaints from lettuce growers and shippers in Salinas, Miklius and De- Loach (1965) chose to examine the basis for the charge of "market manipulation" by chains. The researchers had three ob- jectives in mind: to investigate available evidence regarding the change in the num- ber of buyers and sellers in the market; to estimate the relative volumes bought by the principal chains; and to investigate the purchasing patterns of the three principal buyers in order to learn whether their pat- terns resembled those which could be ex- pected in an oligopsony market. The prin- cipal findings were: • Although the records are incomplete, it appears that the number of lettuce buyers declined during the decade 1953-1963. The change was neither drastic nor abrupt, and it was offset, to some extent, by a decrease in the number of shippers. • The three principal chain buyers ac- counted for a relatively small share of the market. During the 92 days of the 1963 shipping season, they purchased about 10 per cent of the daily supply of Salinas lettuce. • The principal buyers paid either average or above-average market prices. Their purchasing patterns did not show interdependence in the var- iation of quantities purchased. • The study did not disclose any overt attempt by the buyers to follow a strategy of withdrawing from the mar- ket during periods of falling prices, which would have encouraged a fur- ther price decline. The authors recognized the weaknesses of the approach used; nonetheless, they believe that the same research approach would be applicable to other commodities and the results would be very useful to a better understanding of the conduct of the large retail food firms. Viewing the chains as sellers, one must face the problem of a concentration of economic power in terms of market dom- ination. The declining importance of the small, independent retailers is evident from the statistics already presented. It also is equally clear that the small store, regardless of ownership arrangement, is under severe competitive pressure. But this pressure usually arises from the ina- bility of the small operator to fit into a pattern of mass buying and selling. The best illustration of this economic pressure [36] on small retailers is found in the current policy of most retailer-owned cooperative buying groups which either deny member- ship to small retail firms, set up subsidiary organizations to supply them, or charge special service fees for small volume trans- actions. Cooperatives make no effort to en- courage small retailer members to retain their membership. Since the small stores lack sales volume, they also lack purchas- ing volume and bargaining power, even among independents. The claim that chains have resorted to unfair selling practices generally means that the claimant believes that chain prices, particularly on specials, often are lower than wholesale costs to the small operator, or that the quality of chain products and other consumer services are inferior. Many states have sought to end below-cost pricing by means of minimum price laws. Also, state and federal laws forbid unfair selling practices. In view of the fact that relatively few court cases have been recorded as a result of actions to en- force these statutes, one might infer that most claims of unfair practices are ques- tionable. If one chooses to accept the con- sumer as the judge of such questions as prices, services, and product qualities, one will see that they have largely shifted their patronage to large stores, either independ- ent-affiliated stores or chains. Moreover, this shift took place when alternatives were available and it is continuing while alter- natives are available. The Bureau of Labor Statistics Food Price Index reflects the upward movement of retail prices for food (table 14). The United States Department of Agriculture's Market Basket Index reflects the rising spread between farm prices and retail prices of fresh and processed products sold in retail stores. Both indexes are used as a basis for the charge that "distribution costs too much," and that buying and sell- ing practices of large retail firms are the principal reason (Supermarket News, 1966). The U. S. Department of Agriculture has published an informative list of re- ports over the last 15 years dealing with retail margins on food and the compo- nents of those margins (Hamilton, 1959; Table 14 AVERAGE RETAIL FOOD PRICE INDEX (1947-1949 = 100) Year Index Year Index 1964 1963 1962 1961 1960 1959 1958 1957 1956 1955 1954 1953 1952 1951 1950 1949 1948 1947 1946 1945 1944 1943 1942 125.5 124.0 122.0 121.1 119.7 118.2 119.4 115.4 111.7 110.9 112.6 112.8 114.6 112.6 101.2 100.0 104.1 95.9 79.0 68.9 67.4 68.3 61.3 1941 1940 1939 1938 1937 1936 1935 1934 1933 1932 1931 1930 1929 1928 1927 1926 1925 1924 1923 1921 1920 1919 52.2 47.8 47.1 48.4 52.1 50.1 49.7 46.4 41.6 42.8 51.4 62.4 65.6 64.8 65.5 68.0 65.8 60.8 61.4 63.5 83.6 74.2 Source: U. S. Department of Labor, 1951 and 1952; 1961-1963. U. S. Department of Agriculture 1957, 1965a and b). These reports have not con- sidered profits in any great detail. On the other hand, they establish the fact that the rise in retail margins parallels the rise in the costs of the two components, labor and capital. The Department's reports state that the increases arose mainly from rates for labor and capital and a very sub- stantial increase in the customer services included with the product purchased. Considering the alternative channels available in most markets, there is no really sound basis for stating that retail store customers are being forced to pur- chase unwanted services other than those rendered in response to those customers. In a previous study, one of the present authors chose to approach the problem indirectly. He attempted to compare the prices and services of four stores of a large, highly successful consumer coopera- tive with those of four stores of compara- ble size operated by different food chains in the same California communities where labor contracts were operative and rentals [37] were fairly uniform. 8 The net price dif- ferences, including patronage dividends to cooperative members, were negligible. The customer services in the form of pre- cooked food, high-priced specialty prod- ucts, and advisory services were slightly greater for cooperative patrons. Since there are relatively few successful con- sumer cooperatives in urban areas in the United States and the cooperative objec- tive is to serve patrons at cost, the fore- going comparison implies that retail mar- keting margins are fairly well governed by input costs and that prices to customers must be competitive even between coop- eratives and profit-type firms. For the 30 chains for which data are available, the authors were able to estab- lish a positive relationship between the size of the firm and net returns per dollar of sales or investment. Industry-wide data also are sufficient to compare food retail- ing profits with those of a few other indus- tries; however, this type of comparison pre- sents difficulties. Because our economy consists of indus- tries of varying sizes, dollar profits vary considerably. Hence, comparisons are in terms of profit ratios or percentage rates of return, either on sales or net worth. Figures 23 and 24 contain a comparison of profits for selected industries for 1954, 1959, and 1964. The story about profits is rather simple. In condensed form, one may say: • Profits as a percentage of both net worth and dollar sales rose steadily for four of the industry groups in- cluded in the comparison. For food retailing and agriculture, 1964 ab- solute profits were above 1954; how- ever, 1964 profits were below those of 1959 for both retailing and agricul- ture. Fig. 23. Comparison of profits as a percentage of sales, selected industries, 1954, 1959, and 1964. Source: Appendix table E-l. See footnote in that table for method used to arrive at net profits for agriculture. 18 17 16 15 14 13 - 12 11 10 - Agriculture Durable goods industry Other consumer goods industry Food processing industry Retai I food industry A Nation's 500 largest industrial corporations 8 DeLoach, D. B., A comparison of prices among selected retail food chains and a consumer Cooperative, 1964, 25 pp., unpublished manuscript. [38] Retail food industry Durable goods industry! Food processing industry?.' Other consumer goods industry?/ Agr iculture o- Fig. 24. Comparison of profits as a percentage of net worth, selected industries 1954, 1959, and 1964. Source: Appendix table E— 1. See footonote in that table for method used to arrive at net profits for agriculture. • The food retailing industry operated with the lowest percentage return per sales dollar; agriculture had the high- est return per sales dollar. Conversely, agriculture had the lowest percentage return on net worth; the food retailing industry's return on net worth was favorable. This latter comparison shows the effects of sales volume on the net return on capital, reflecting larger total profits for a large-volume, small-profit item than for a small volume, high profit product. In this respect, it is worth men- tioning that the average U. S. farm pro- duces enough food for 28 persons; the average retail grocery store supplies an average of more than 700 persons. The food retailing industry usually shows its net profits as a percentage of dol- lar sales; agricultural groups state theirs in terms of return on net worth. These im- plied preferences are associated with the desires of the groups to convey a notion of "fairness" or "unfairness" of returns to them for the activities in which they are involved. The food retailing industry, in common with other industries, has no criteria for determining what is a "fair" rate of re- turn. Returns have been sufficiently high since 1946 to induce investors to provide adequate funds for expansion, either from new capital from outside sources or from retained industry earnings. In reference to a "fair" return on own- er's investment, various government agen- cies have arrived at about 10.5 per cent as a justifiable net profit rate. 9 By doing this, the government agency became a judge of "fairness." For the three years — 1954, 1959, and 1964 — food retailing was within an 8.8 to 13.1 per cent range. This percentage return is significant in view of the previ- ously mentioned claims that the food chains are using their concentrated market power to hold farm prices down and retail prices up in order to gain excessive profits. Retailers also cite studies of their costs of operations to show increasing prices are a result of higher costs (England, 1959). The retail food industry also claims that it takes only $15 a year per family in the form of net profit for its services. 8 The Interstate Commerce Commission recently used this figure in analyzing the possible existence of excess profit in the nation's public transportation sector. Similarly, the Federal Aviation Authorities used approximately the same return for domestic airlines. [39] GOVERNMENT REGULATION The rapid growth in the economic power of grocery chains has forced some major adjustments in marketing channels and methods. The policy of larger chains to circumvent wholesalers and purchase di- rect from manufacturers and packers or growers, or even to engage in manufactur- ing and packing activities, has descreased the importance of wholesalers. The con- sumers' support of chains, as indicated by their patronage, concentrates selling power in relatively few firms, thereby enhancing the bargaining power of some chains in dealing with manufacturers. Finally, the shift of consumer patronage to chains or to large independent supermarkets has considerably diminished the economic power of small, independent grocers, but it has not lessened their political power. In a competitive struggle for business survival, those who have been affected ad- versely by the growth of chains have sought to buttress their weakening economic posi- tion by appealing for public support or seeking legal constraints based on the pre- cept of "preserving competition." by means of such activities they have had con- siderable influence in the development of whatever public policy we now have in the area of food distribution. It is not the purpose of this paper to analyze state and federal activities which restrain the development and use of eco- nomic power in the market place. How- ever, a few brief comments about this sub- ject should suffice to describe the environ- ment in which the growth of grocery chains is occurring. Most of these government ac- tivities are federal in origin; hence, the comments will relate to these only. Legal restraints The economic growth of the grocery chains has been influenced greatly by the various antitrust laws — Sherman (1890), Clayton (1914), Federal Trade Commission (1914), Packers and Stockyards (1921), and Robin- son-Patman (1936). The essence of the antitrust legislation is to prevent or break down any undue concentration of market power which is, or threatens to become, a monopoly in restraint of trade or to regu- [ late business conduct and trading prac- tices that could be, or are, injurious to competitors. All such legislation is nega- tive in the sense that it follows the prin- ciple of "thou shalt not." Administration of the Sherman Act is a function of the U. S. Department of Jus- tice. This department also shares with the U. S. Federal Trade Commission the re- sponsibilities for enforcing the Clayton Act. The Robinson-Patman and U. S. Fed- eral Trade Commission Acts are assigned to the U. S. Federal Trade Commission; the Packers and Stockyards Act to the Sec- retary of Agriculture. In the past, investigators centered pri- marily on the large meat packers and gen- eral food manufacturers. Investigations have been centered on the food chains in more recent years, largely because they are disturbing to the small retailer and large meat packers and food manufacturers who have developed market interests which are being undermined by mass buying and selling practices. With the exception of antitrust actions against A & P, the chains were relatively free of significant investigations until the early 1930's when Congress undertook a broad inquiry that subsequently resulted in the enactment of the Robinson-Patman Act. Since 1946, a number of Congressional and special Commission inquiries, U. S. Federal Trade Commission and Packers and Stockyard Administration actions, and Department of Justice suits have been in- stituted. Some have been concluded. A few observations regarding the current inquiry of the National Commission on Food Marketing, the activities and impli- cations of the U. S. Federal Trade Com- mission, and court rulings on cases brought by the Department of Justice or other plaintiffs are contained in the fol- lowing section. Public opinion When legal remedies are lacking, those who claim injury from competitive prac- tices often seek relief by appealing to their legislative representatives for an investiga- tion or an enactment of remedial legisla- 40] tion. A second procedure is to request an administrative agency to conduct an in- quiry with a view toward recommending new legislation (U. S. Federal Trade Com- mission, 1960). A third procedure is to ap- point a special committee or "prestige" commission to conduct an inquiry into an allegation of unfair practices and to pre- sent its findings to the Congress. This lat- ter procedure is often used when estab- lished government agencies are unwilling to undertake such an inquiry or legislators or administrators wish to circumvent es- tablished agencies. Frequently, the real motive for the three types of inquiries is to focus public attention on the subject of the inquiry and thereby induce an alleged offender to mend his ways. In a sense, the public hearings conducted by various gov- ernmental bodies become a source of news for the several news media. The hearings can be and have been extremely damaging to the "image" of a particular firm or an economic group such as grocery chains (U. S. Congress, 1959). The current inquiry of the National Commission on Food Marketing is an out- growth of the unrest over rising food prices in retail store and low prices, particularly on beef cattle, at the farm level. Briefly, the Commission was asked to concern it- self with past and future changes in the food industry; what is required to assure an efficient system of production, process- ing, and marketing while maintaining in- dustry competition; methods for achieving a "desired distribution of power" and "de- sired levels of efficiency" within the system; the adequacy of government services and regulatory activities; and the effect of im- ported food on domestic producers, pro- cessors, and consumers (Ginsberg, 1966). The National Commission on Food Mar- keting has been assembling and analyzing evidence bearing on the foregoing sub- jects. It seemingly has encountered all of the statistical difficulties faced by previous investigators who were trying to measure performance among the various sectors of food production and marketing. Among these difficulties are a lack of comparabil- ity of many products and customer serv- ices and rather wide variations in account- ing procedures. This lack of good statistics handicaps any investigational body that is seriously seeking the best answer to the various eco- nomic problems associated with food pro- duction and marketing. Because the whole problem of the validity and adequacy of data has plagued both the industry and various governmental agencies for many years, it would appear that this problem needs to be resolved as a basis for a more complete understanding of the issues in- volved. Current investigatory philosophy Both the U. S. Department of Justice and the U. S. Federal Trade Commission have used their investigational authority to as- semble and analyze statistics relating to the operating policies and practices of all sectors of the food industry. The objective of such inquiries generally has been di- rected toward either obtaining evidence about trade practices of a given firm, or about concentration of economic power. Inasmuch as the antitrust laws adminis- tered by the two agencies were passed to prevent undue concentrations of eco- nomic power and encourage competition, the administering agencies were given the task of defining "undue concentration" and "competition." As is true with any legislation, the ab- sence of precedents means that the admin- istrators of quasi-judicial bodies such as the U. S. Federal Trade Commission build precedents by the manner in which they select firms to be charged with violations, the nature of their evidence, and their subsequent decisions. Similarly the De- partment of Justice has the responsibility for discoving and assessing the evidence of violations of specific statutes, obtaining indictments, and bringing the cases before the courts. The decisions of the courts are then used as precedents. By virtue of the authority of administrative agencies to in- stitute or not institute an antitrust action, one finds a great variation in the applica- tion of the law and the importance the administrative authority attaches to dif- ferent forms of business conduct. The published literature on government regulation of monopolies indicates that there is far less than a concensus within and among our courts and various admin- istrative agencies on what constitutes a [41] monopoly in restraint of trade. Certainly there has been uncertainty regarding what constitutes undesirable competitive prac- tices. However, a measure of relief should emerge from the recent practice of the Antitrust Division of the Department of Justice to scrutinize proposed mergers. In an effort to preserve and protect small businesses, the concept of the "op- timum social competition" has emerged which means a bypassing of many basic economic considerations. It is under this social concept that the U. S. Federal Trade Commission has been instituting actions to prevent mergers on the theory that such mergers might be harmful to competition. For example, Section 7 of the Clayton Act recently has been used by the Commission to prevent structural changes which create or aggravate sub- stantial market power in one firm, includ- ing changes that fall short of monopoly actions, in order to forestall anticompeti- tive behavior. Unlike other sections of the Act, Section 7 does not focus on whether or not an illegal merger or purchase action has occurred. Insted, it calls for an economic prediction of the competitive effects that are likely to fol- low from a proposed structural change that would further concentrate economic power. In other words, the Commission does not have to prove that further concen- tration will result in monopolistic pricing practices. It only needs to show that, on the basis of economic logic and industrial experience, such behavior is "reasonably probable." 10 Recently, the Commission handed down a divestiture order based on the above re- lationship. 11 The case later was brought before the U. S. Supreme Court which re- versed the Commission's decision and stated, "Probability can best be gauged by what the past has taught. We are con- vinced that the Commission has mistakenly rejected what the record demonstrates as to what actually happened in the past in favor of a future possibility based on con- jecture and speculation." Despite this court action, Robert A. Hammond 12 (a proponent of the philoso- phy accepted by the Commission) states, "The costs of such an approach (the insis- tence on the existence of a high degree of certainty prior to any antitrust statute enforcement) . . . might well result in the transformation of the structure of our economy in a way totally inconsistent with our reliance on competition as the pri- mary economic regulator." Retail food chains have shown consider- able initiative and ability in adjusting to the legal constraints under which they must operate. The most outstanding exam- ple has been the need of firms to accom- modate themselves to the Robinson-Pat- man Act. In this case, specific buying prac- tices are forbidden as unfair and destruc- tive to competition. Since it is the func- tion of management to develop and use whatever legal and ethical business strat- egies it can in order to find an optimum solution to its problems, retail food firms turned their attention to finding other management procedures for keeping their procurement costs down. Frequently this meant the acquisition of their suppliers' businesses or the development of their own food processing or manufacturing operations. 10 Brown Shoe Company versus United States, 370, U.S. 294 (1962); and United States versus Continental Can Company, 12L ed. 2d. 953 (1964). 11 United States versus Kennecott Copper Corporation, 231-F, Supplement 95, (S.D.N.Y.) 1964. 12 Hammond, R. A., Antitrust in an Expanding Economy, paper presented at conference of National Industrial Conference Board, March 4, 1965. [42] APPENDIX A Table A-l POPULATION AND DISPOSABLE PERSONAL INCOME, BY STATES, 1954 AND 1964 State Population 1954 Percentage change Per capita disposable income 1954 1964 thousands dollars Connecticut Maine Massachusetts New Hampshire New Jersey New York Pennsylvania Rhode Island Vermont Illinois Indiana Iowa Kansas Michigan Minnesota Missouri Nebraska North Dakota Ohio South Dakota Wisconsin Alabama Arkansas Delaware District of Columbia Florida Georgia Kentucky Louisiana Maryland Mississippi North Carolina Oklahoma South Carolina Tennessee Texas Virginia West Virginia Arizona California Colorado Idaho Montana Nevada New Mexico Oregon Utah Washington Wyoming Total U. S 2,208 917 4,857 554 5,249 15,818 10,798 817 375 9,087 4,266 2,631 2,031 7,040 3,127 4,139 1.329 623 8,962 662 3,599 3,051 1,781 366 828 3,462 3,651 2,912 2,887 2,642 2,079 4,185 2,157 2,234 3,364 8,449 3,483 1,927 932 12,738 1,520 589 613 215 784 1,652 762 2,539 300 161,915 2,816 996 5,505 639 6,688 17,903 11,764 927 405 10,618 5.019 2,808 2,224 8,617 3,673 4,501 1,490 641 10,174 716 4.177 3.409 1,929 511 810 5.798 4,268 3,143 3,516 3,480 2,299 4,920 2,483 2,561 3,809 10,602 4.414 1,768 1.670 18,576 1,990 699 708 401 996 1,980 1,001 3,080 346 193,468 27.5 8.6 13.3 15.3 27.4 13.2 8.9 13.5 8.0 17.7 6.7 9.5 22.4 17.5 8.7 12.1 2.9 13.5 8.2 16.1 21.8 31.7 10.6 17.6 15.1 14.6 13.2 25.5 26.7 - 8.3 79.2 1,984 1,343 1,663 1,378 1,872 1,821 1,614 1,579 1,283 1,854 1,649 1,484 1,482 1,686 1.418 1.476 1,446 1,384 1,752 1,379 1,580 1,010 955 2,102 2,156 1,313 1,126 1,103 1,204 1,491 817 1,035 1,272 1,022 1,118 1.457 1,290 1.185 1,381 1,816 1,499 1,338 1,604 1,984 1,313 1,590 1,379 1,744 1,750 2,791 1,869 2,535 2,137 2,605 2,708 2,216 2,197 1,942 2,610 2.222 2,122 2,041 2,359 2,087 2,266 2,144 1,894 2,235 1,855 2,214 1,538 1,558 2,783 3,003 1,957 1,755 1,687 1.650 2,480 1,388 1,704 1,892 1.487 1,678 1,904 1,931 1,781 1,960 2,656 2,252 1,819 2,004 2,998 1,779 2,230 1,935 2,273 2,214 2,225 Source: Sales Management, 1953-1965. [43 . © © c e © © c c - c © CD © © o o o o o o © o o © © © © © © c © © © © © c 8 OS HO O ._! OO OS i— I t-- «5 C CM CN CM CN CO CM CM r-l rH rt rH O CO * «5 o «■ «5 © IT © C CN CN 1 •» 8888 gg m * o o © o co •rtH lO O - 1 W . CM OS M< os c- © c ■*t< CO* ffl H N lO* C lO .— 1 CN CM CS "* CM rH rH rt r^ O CO CM OJ 1 CN * 88 S © c © c 1 o 1 © © © © © © C © © © © © C CO © © c oo oc oo lO ■"*! © 0 G © © os CO* rH* tti 00 C ~ s s s t~-T oT cm" t^T in r- rH U5 CO rH -rj CM H rH rH rH O lO rH CN CM co ©©©©<= © © <= 1 <=> © © © © © C © © © © © C © lO © O © u" 8 © o CD OS ti lO © oc © © <= © t^ l^ ° "3 OS co" © »o 00 c CM t^ O 1 o (M" 00 rH © -HH* \r CO CO t~ r-l CC oo CM rH rH CM rH 0C CO rH CN 1 CM CO r3 s s s © © © © c © © c © © © © © © c <=> OS © i© i© © c O CD © © CC © © c CO T-H "»» 8 lO >o © © © c 8 OS O* H (D 0O if cm* i>" a lO cm" oo" cm" ©" to" oc CM rH rH IM rH OC co" t^ t^ rH CC 00 t^ rH Cs CM 8888§ 8§i 8 8 888 8li 8 W © »o »o © e lO CO oc oo lO US © © © c oo OS O *" !D 00 C 0O N H OC CN? CO oc oo" cm" od CM* © »o" oc co" r-l r- OS CM rt rH CM rH 0C oo rH CN CM CO © © © © C © O O © C 88 = © © © © © © c © © © © © © © c >o »o © © © c 8 «5 © © © © C © © c © os !>■* O N 00 CJ to in ot ©" CM OO CO © t^ r- rH 00 CC H O rH CN CO CM rH rH CM rH O © © © © © C © © c © © © © © © © C © © © © © © C © © © © © c CO r)< C © © lO lO O "O UJ IT i— 1 CO OC CO © © CO © © t^- C CO OS "O » i« N <» in ma CO lO ©" CO oo" »o" cm" rt © OS rH CN r-l r- CM CM rH CM rH C rH rH C"- CO rf ©©©©<= © o c © ©©©©>©©(= ©©©©©©<= © 8 S 8 S 1 © "o ■h* CD C CO co © oo co os i>- c CD os ■* 0O CO N «■ m co o «3 h CO © CO CM* C o CM CM rH CO rH r- CD rH rH C" CO "* w w rt o Eh co K o H <1 S c T3 Pi 4> r-l o GO *o (3 ca o "3 £ c 1 a a 2 o o Q c CD a CD .5 CI cS .s a c o H Q o c {J s a i o cs g a spendents rocery stores (with ambination stores auntry general stoi c 2 K a | «( c3 e- 1 a I 02 CD l! •11 » .2 < -° fe a c h r_i O H g '3 rC o L ,cs -c ^' c S> 03 1 I c I en CD rQ cd ~t: bo - > rr- T3 C ' a « CO 1? cS 3 ' >i r. r? a < H 0- r3 < <3 Eh H s g 1 ■oooofl rC C O .S A r4 o " S J, cJ o 1 2 '3 § * rf rH" » CO CO OS t— t^ ,_, «j. CM co 00 CO •r o -* MS oo CO ^ CM CO o O co CN MS 0C N CO CC § CM CO O 01 a OS ms •^ o ^ -H O — CM 00 C MS CN CN co co cc CO MS CM ss os o- CM oo "5 •tfl O ■*< r~ o r- CO CO C © MS MS CO CO o~ os o- ms m; — 1-4 CO OO CM 8 o o c CO O cc co co a- CO rtoq« oo cm a r- t~- CM CN CO lu o CD N m a CD - a < a CD - 73 ►- (3 < "S CD - xit- fi < mall* Chain . Indepe TOT uperette Chain. Indepe TOT > F -2 CD C 1 &3 CO CO < CM >o o- MS o •o co re so ^h CM CO OS 00 I— CO ^« CD -h CO CM CM CO CO CD MS CM MS f- 00 CM C O co co os o- CO MS CN OS OS oc MS MS C MS O MS o X NW o <= CD t^ cn ms r- CN ^H CO MS <<% <0 N O OJ N (D CM OO rH rH th OlMtOMOOOOtDUSW (M (N rt rt m NOlNCXlrtWNOllO i-H rH r^ io t^ r^ O CO t}i o CO CO 00 t^ rH 00 O IO © Ol i— I t^ "O CS T(l «l N N N Ol O) M lO O OO W co t^ © co oi © t~ t— o oi h "5 CO rt ■"*l CO IO t-- ffl IN N O Ol «5 * U5 llj h IN U) 00 N OIOOMIOIOOIOJO- I CO i-H rH COt^OOOTtii— i©©coiOthco Ol CO CM i— I CO CM CO i— I J>. CM O lO to 115 pq TH CM Ol IO O N M OO CO 0O f~ t^ IO io r^ -* oi io cm t-h O co O co «5 O Ol W N to N o to to IO Ol CO Tj< r^ IO to to 11} Tti to © IO CM IN 00 o CM 0B IQ co OS CN 3 co — «Q Th co OHU)l(5>JlH01rtN MOOOlONOOt-OOCO »H T-H O t- CO CO th CM CO O O —i cm" r co ffl M « M Ol CM CM CO CO Ol CO tD Ol tO 113 CO CO CM veqwujcoNHrt NOOtOMrtCOflCOlOlO ~~ CO IO CO CO CO CO CM CO CO CO CO CO CO CO CO © O0 © rH Ol rH Ol CM CO O 00 Ol ■*■ CO tO IO N H COMB CO © © 00 Tfl M CO 1l< O) rt —l CO O Ol ^- CM O Ol IO Ol 00 o O Ol CM CO CM rf CO ■— ' t-t t- CO Ttl IO rH Tt I— I I— I t-H as a .r¥ « S 03 r fl S 8 s ? s '^ 8 .2 .S .a -g S3 g j2 o S go 3 1 -I J 3 s 1 O W Ol !D M<5 h S N o t^cO0S-^00»0OO01O CO CO OS OJ «-l m CO O m •>*< —c m ^l W W N IO CO CO m o to eo co IO O OO ■<*< CO >n oo oo o o> - a> co co oo 2 S O lfi> CO ■* n n n rt CO Oi CO CO CO" Co" i-H Co" rH Co" CO* ^HNlflNNOONOlO 00>— l M N N H U) O) OO CO-^OOOCOO— itXCO^h — < OS CO © OO >— IO3t^«*l«0 oo m co co co co © m O m CO CT> CO in CO 1-* CO CO — i «-i 1-1 co © t- r- NNinNMNffflTfin m CO © OS •«* © OS t^ ■** Tf CO CO tO 00 w t>. OO^OOMWHONOO OSCOOCO-*t-— i — "0 COCO'-H'-l'-l'-i^COCOO t^--^(CO»-i^HOOC0^HCOC0i-i r C0O05OC0rt05C0Tt<00 OCOOOJOO-HCOO^OCO •-H00O3C000COOO'— ico t^r^oioiooosoocooo^ in^Noi^ioxtOTfi- co«no3cot^Tfio>050oco CO CO Co" •** Co" Co" CO © •«*" -H i-H OO >-< -^ i-h" CO t--C0Oimi^-^COr- C000OO3C0»OTfC0C005O OOTfNOOinmtO'J'OON COCOCOOO—i-HOOmcOCO© ooooHHNn^iio) oiNioiniowNtDNinco CO Co" CO" Tf" Co" CO CO OO CO i-H Co" ^H rt co" oo -h co — i © os © co ct> -»-i CO CO CO t-c * m oi m O) rH rt N CO ^h ^h 1^ co o> © co m in oo rf n co" oo « t^ cot^i^Tfincocoincocoo cocoooooococococDom OO^NOlliOONVN^ co co © (^ •*o)innoNNNHNH Nom^inin^NONoo N N M rt « rt-^rtCO oiinoraoNNocoH mt^o>ooco©comcoco Tf<"*lCO'<* g g 02 H _ o | | | jj " 2 $ U o a C C oco "Cr W a CD tD +s * CO 3 a o PL, 1 &W- ^^y. :?;£& &:'V V'-^ ■^^" ;MT- S?T* : * u - o a. u> c o a. 8 S ° S » S 1 g s| 5 5 ID o o la o Q. CN o CJ> o a > a o > II c c o 4) o ■£ w _c «/> > o ■^ o o V) o o a> 1/1 ;-. u «•>" ^ .! D APPENDIX C Table C-l REGRESSION ANALYSIS RESULTS— ALL 30 CHAINS* Equation Dependent variable Independent variable Constant "b" Coef- ficient Coefficient of corre- lation £-ratio n I Net sales Number of stores 25.9030 1.072900 .97860 84.04300 313 II Net profit Number of stores 1.1431 .011167 .92408 42.64000 313 III Inventory Net sales 3.9665 .057290 .95143 54.49900 313 IV Current assests Inventory 1.4138 1.587000 .99550 185.49300 313 V Fixed assets Total assets 1.8153 .387000 .96972 70.03500 313 VI Total liabilities Accounts payable Current liabilities 9.3857 1.656890 1 959800 .99136 8.95870 16.32000 313 VII Net worth Total assets Net sales 47.9168 .003349 .19330 3.47480 313 VIII Fixed Assets Total assets Total assets 31.3950 .011456 .13786 2.45479 313 * Observations are based on annual data covering the period pertaining to the 30 major retail food chains, t Degrees of freedom 312 — 1.959. 1954-1962, inclusive. Observations consist of data [54] Table C-2 REGRESSION ANALYSIS RESULTS— BY SIZE GROUP Equation Group Dependent variable Independent variable Constant "b" Coeffi- cient* Coeffi- cient of corre- lation <-ratio n I (1) (2) (3) Net sales Number of stores -32.0583 84.5949 42.0390 1.0970 .8940 .6366 .94610 .27540 .21530 43.5585 6.3467 5.0506 110 108 95 11 (1) (2) (3) Net profit Number of stores 2.4506 1.1119 .7986 .0106 .0119 .0024 .79714 . 12720 .02150 20.6010 3.9302 1.4278 110 108 95 III (1) (2) (3) Inventory Net sales 10.7090 - 4.5370 - 1.0578 .0546 .0969 .0842 .88610 .46730 .80830 28.9923 9.6437 19.8010 110 108 95 IV (1) (2) (3) Current assets Inventory - .5813 4.2617 .0683 1.6001 1.5205 1.6424 .99160 .90110 .97140 113.0970 31.0706 56.1980 110 108 95 V (1) (2) (3) Fixed assets Total assets 6.7927 - .0910 .3670 .3145 .3558 .3174 .90230 .95650 .86660 31.4380 48.4858 24.5815 109 109 95 VI (1) (2) (3) Total liabilities Accounts payable Current liabilities 28.8510 1.9958 1.5949 .98060 7.5473 9.0168 110 5.2052 - .3594 3.2296 .97850 .8628 12.1524 108 2.4363 3.9019 .4672 .91900 8.1795 1.2846 95 VII (1) (2) (3) Net worth Total assets Net sales .4861 50.1149 37.0850 .0031 - .0056 .1328 .07600 .00230 .07690 2.9823 .4950 2.7652 110 108 95 VIII (1) (2) (3) Fixed assets Total assets Total assets 32.1226 32.7098 20.8407 .0076 .0206 .5851 .01550 .01380 .02150 1.3021 1.2196 5.0458 110 108 95 D.F. t 109 1.959 107 1.959 94 1.959 [55] Table C-3 REGRESSION ANALYSIS RESULTS— BY INDIVIDUAL CHAIN, GROUP II Equa- tion Group Dependent variable Independent variable Constant "b" Coef- ficient* Coeffi- cient of corre lation t -ratio n I (ID Net sales Number of stores - 94.09500 2.30100 .915000 9.84600 11 (12) 347.99400 - .37250 .044380 .57010 (13) 417.91600 - .81600 .351200 2.20730 (14) -257.43900 2.17240 .329600 2.10350 (15) 97.09500 .70678 .532800 3.20380 (16) -621.31700 5.16150 .578900 3.51780 (17) - 36.25000 1.85000 .968600 1.66690 (18) 13.38800 2.15700 .952800 13.48290 (19) 17.64400 1.42200 .982400 21.13890 (20) 40.11450 .90320 .130400 1.16160 II (ID Net profit Number of stores - 3.50300 .04807 .923700 10.43800 (12) 4.31880 - .00360 .023800 .41270 (13) 9.92980 - .03320 .617900 3.81500 (14) 10.12070 - .04560 .398200 2.44000 (15) 2.47700 - .01360 .384000 2.36870 (16) - 3.86800 .03590 .306600 1.99500 (17) - .06670 .02210 .928800 10.08340 (18) 1.33700 - .00760 . 109900 1.05450 (19) . 13370 .01250 .969400 15.92900 10 (20) 2.28600 - .01360 . 105700 1.03160 III (11) Inventory- Net sales - 7.12330 .08720 .854900 7.28260 (12) 8.57350 .01902 .058800 .66104 (13) - 49.87200 .30430 .745300 5.13160 (14) - 8.28550 .11800 .530800 3.19060 (15) - 1.00330 .07110 .922100 10.32200 (16) - 1.61210 .07650 .868300 7.70390 (17) - .02720 .06150 .901000 9.05100 (18) - .62150 .07910 .859500 7.42100 (19) .97100 .06010 .985500 23.35700 10 (20) - 1.40300 .07290 .526800 3.16500 IV (ID Current assets Inventory - 1.72050 2.21760 .985800 25.02600 (12) 31.96840 .37910 .076700 .76270 (13) - .20690 1.45600 .993700 37.85500 (14) 1.47800 1.49800 .926700 10.66800 (15) 1.53500 1.48960 .921100 10.24800 (16) .09710 1.54190 .981700 22.02670 (17) 2.50860 1.66520 .926400 10.64480 (18) .71910 1.82080 .864200 7.56950 (19) .74850 1.60700 .963200 14.47830 (20) 1.69350 1.63480 9880b0 27.29500 V (11) Fixed assets Total assets - 2.83950 .34680 .983900 23.52100 (12) - .36430 .33830 .942100 10.67100 (13) - .35620 .35870 .959600 14.62200 (14) - 2.71650 .53370 .877600 8.03530 (15) - .30000 .39820 .968300 16.58950 (16) - .73440 .31730 .905100 9.26800 (17) - .44110 .37810 .966000 16.00660 (18) 1.14450 .32380 .737100 5.02320 (19) - 2.61910 .46160 .989200 27.08300 10 (20) .92730 .25160 .909700 9.52260 11 [56] Table C-3 (Continued) Coeffi- Equa- tion Group Dependent variable Independent variable Constant "b" Coef- ficient* cient of corre- lation 2-ratio n VI (11) Total liabilities Accounts payable Current liabilities 6.96689 -1.94610 4.04490 .985600 .71936 2.91730 11 5.12470 3.19970 (12) 6.47280 - .42980 .909200 .38950 9 1.03810 .48330 (13) - 3.08690 2.41260 .993000 1.78270 11 -1.53230 .72340 (14) 4.81780 5.08150 .904100 3.31780 11 -2.04123 .98122 (15) 2.60520 6.56010 .928900 4.42900 11 -1.17110 .53600 (16) - .36460 4.06800 .979000 3.04980 11 -1.86400 1.16240 (17) 2.16110 3.88500 .966000 3.24430 11 2.04750 2.91840 (18) 6.58840 2.67580 .922400 9.71120 11 5. 15910 1.82450 (19) - 8.54680 - .18410 .972500 .09060 10 -1.59730 .48030 (20) 3.05670 3.71480 .889500 1.88260 11 VII (ID Net Worth Net sales .33020 .03980 . 103700 1.02050 11 (12) Total assets 9.75380 . 10770 .827800 5.80100 9 (13) 41.32760 - .01010 .006800 .24789 11 (14) 27.55200 - .11640 .018800 .41580 11 (15) 80.65900 - .09450 .325900 2.08600 11 (16) 38.50100 .01980 .008900 .28387 11 (17) 50.65200 - .00043 .000045 .02014 11 (18) 73.65500 - .24570 .690900 4.48560 11 (19) 47.55200 .02910 .117070 1.02990 10 (20) 33.98990 .22270 .619340 3.82660 11 VIII (ID Fixed assets Total assets 11.84200 . 17620 .657800 4.15800 11 (12) Total assets 2.22400 .44040 .884200 7.31100 9 (13) 33.40400 .00800 .008100 .27190 11 (14) - .43340 1.29260 .982300 17.76700 11 (15) 37.23400 .04910 .041800 .62690 11 (16) 10.63900 .49810 .594000 3.62830 11 (17) 38.07100 - .05220 .027700 .50600 11 (18) 38.44700 - .05690 .007600 .26220 11 (19) 25.34600 .29690 .818700 6.01090 10 (20) 33.67400 - .15350 .043200 .63760 11 D.F. 10 t 2.228 2.306 APPENDIX D Application of Markov technique to structure data Economists are often interested in the characteristics of institutional changes over time as well as in the paths these char- actistics are likely to follow in the future. If, in a given sequence of events, the out- come of each event depends on some chance occurrence with a known prob- ability, then such a sequence is called a "stochastic process" to which a Markov Chain Analysis may be applicable (Judge and Swanson, 1961). Such an analysis has been attempted in our study. Table D-l shows the changes in the market shares of the 30 largest retail food chains during the 1954-1962 period. Our first step was to develop six classifications according to market share. With these classifications, we could construct a cross- classification table showing how many chains moved into or out of each classifi- cation during this 1954-1962 observation period. Given the information in table D-2, it is then possible to construct a possibility tree and attach branch weights that de- scribe the process as it moves through any finite number of steps. Alternatively, the transition probabilities (Pa) can now be calculated and represented in the form of the following transition matrix "P" (table D-S). Where's P i} = 1 and Pi^O for all i and i =. 1 /. The elements of "P" denote the prob- ability of moving from some class / to class i in the next step. Since the elements of this matrix are nonnegative and the sum of the elements in any column is I, each column of the matrix is called a probabil- ity vector and the matrix "P" becomes a stochastic matrix. This matrix, together with an initial distribution W° pletely defines our Markov Chain Process. Given this information, we can determine the outcome of, say, the nth period in the Table D-l SALES OF INDIVIDUAL CHAINS AS A PERCENTAGE OF TOTAL SALES OF 30 MAJOR CHAINS, 1954 AND 1962 Chain Percentage of total sales of 30 chains 1954 1962 A&P 35.72 16.24 9.93 5.41 4.66 3.12 2.04 3.96 1.81 3.12 2.18 1.12 1.49 1.16 .84 .96 .36 .63 .55 .56 .77 .79 .43 .52 .49 .23 .34 .14 .13 .24 27.52 13.18 10.23 5.44 5.14 Food Fair 4.85 4.06 First National 4.05 3.37 2.36 2.90 2.05 2.62 .93 Penn Fruit .91 Red Owl 1.47 1.22 .71 1.51 .46 .55 .56 Thriftimart 1.10 .64 .32 .56 .18 Food Mart .35 .43 .33 Total 100.00 100.00 future, i.e., in matrix language: rn w°P - w 1 where w 1 - w^P = w 2 where w 2 - w n - x P - w n where w n - infinite future dis- tribution. Alternatively, this process may be written: w n - w opn [58 J Table D-2 CROSS CLASSIFICATION TABLE Size of firm (1962) Firms moving from: Firms moving to: A B C D E F Size of firm (1954) Total .1000- .2714 per cent .2715- .7366 per cent .7367- 1.9991 per cent 1.9992- 5.4256 per cent 5.4257- 14.7251 per cent 14.7252- 39.9639 per cent A .1000- .2714 per cent 1 1 B .2715- .7366 per cent 4 4 2 10 C .7367- 1.9991 percent 3 3 6 D 1.9992- 5.4256 per cent 3 6 9 E 5.4257-14.7251 percent 1 1 1 3 F 14.7252-39.9639 per cent 1 1 Tc >tal 4 8 8 7 1 2 30 If: P is a transition matrix for a "regu- lar (Hart and Prais, 1956) chain, then the powers of P n approach a matrix T, each row T is the same probability vector w, and the com- ponents of w are all positive. If: P is a transition matrix for a regular chain and T and w are as described above, then the unique vector w is the unique probability vector such that wP = w. In other words, this states that if P is a transition matrix for a regular chain, there exists a unique vector w that is both a fixed vector for P and a probability vector whose distribution at time n moves toward this equilibrium vector irrespective of the initial distribution or starting state: e.g., P n — » T as n — > 00 m i and T = e*w where e = equilibrium vector or distribution. and w is the probability knowns from which the unique values of w can be derived. Based on the changes that took place during the 1954-1962 period, we are then able to use the Markov technique to de- termine: 13 1. the probable structural characteristics of the nation's 30 largest chains at any time (eight-year multiples past 1954) in the future; and Table D-3 STOCHASTIC TRANSITION MATRIX "P" Firms mov- Firms moving from: ing to: A B C D E F A 12.50 B 100.00 50.00 25.00 C 37.50 37.50 D 37.50 85.71 E 14.29 100.00 50.00 F 50.00 Using this information, one can com- bine equations and form a system of n linearly independent equations with n un- 13 For more complete descriptions and applications of the Markov Chain Process, see the following: Padberg, 1962; Kemeny, J. G., et ah, 1957; Anderson and Goodman, 1957; and Collins and Preston, 1961. [59] 2. the length of time it would take for the dynamics of structural change to reach an equilibrium. 14 Note, how- ever, this term "equilibrium" does not indicate that the 30 food chains will no longer grow or decrease in size or change classifications. In fact, chains will continue to move from one class to another; however, the probability of any chain moving into a given classification, in equilibrium, is ex- actly offset by the probability of a chain moving out of that same classi- fication (Adelman, 1958). Limitation to the Markov analysis All statistical endeavors such as this have some limitations — some created by data deficiencies and inaccuracies, others in- herent in the statistical process itself. In the more common and rigorous ver- sion of the Markov Chain Analysis, an extra column and row are added to the transition matrix. This allows for the in- clusion of data pertaining to the possibil- ity of chains entering or leaving the exist- ing classifications. Unfortunately, however, information on this type of an occurrence was not available. Without the data neces- sary for this type of process expansion, it was necessary that our analysis deal with a type of "closed industry" where barriers of entry and exit prohibited a more rigor- ous study. This restriction admittedly has reduced the applicability of our projec- tions and this must be taken in account with respect to their reliability, especially with respect to any long-run projections. 15 According to Adelman (1958), two major limitations in the Markov technique itself must be recognized and, if possible, avoided. The first relates to the compara- tive size of the firms falling within the total range of one's classifications. For ex- ample, it is supposedly much easier for a firm with $500,000 annual sales to raise its total sales by $1,000 than it is for a firm with only $100,000 annual sales volume to accomplish a similar increase. In other words, in a given set of circumstances, it is probable that the sales volume of a large firm will fluctuate by greater absolute amounts than that of a smaller firm. To eliminate this distortion created by the use of "absolute" measurements, all data were converted into "relative" measurements. All sales data were converted into per- centages relative or proportional to total chain sales. The second limitation is related to the establishment of the classification bound- aries. It is obvious that by establishing re- latively many classifications with small ranges, one will find the firms rapidly mov- ing from one classification to another. Conversely, few classifications with ex- tremely large ranges will create an environ- ment where only a very few firms will ever change classifications during the observa- tion period. Additionally, the larger firms again supposedly have the potential to expand (or decrease) their relative sales more than their smaller competitors. Therefore, Adelman states, "The class in- tervals were constructed so that their ab- solute width was greater for large than for small enterprises." Even this procedure, however, should not be subject to the per- sonal whims of the statistician or the "niceties" of the data accumulations. It must basically be a mathematical process, free from personal bias and chosen with some degree of scrutiny. The class interval procedure used for this analysis was a mathematical technique called "proportionality from the upper limit." 18 It follows on the opposing page. Next, a limitation on the results them- selves must be noted. One must recall that the percentages used herein were relative measurements. Hence, table 20 show that in the year 2010 each of the 30 chains 14 However, in this particular application, the matrix of transitional probabilities (P) did not meet the requirement of "regularity." In this case, any discussion of an equilibrium be- comes meaningless since all firms move into the absorbing state (or states) as n — > 00. 15 In a similar way, Collins and Preston (1960) applied the Markov process to the food processing industry. Fortunately, their information allowed them to consider the so-called "not-on-list" column and row. 16 Pankey, Victor, "Proportionality from the Upper Limit," mathematical technique, Davis: University of California, Department of Agricultural Economics, October 1965, unpublished manuscript. [60] PROPORTIONALITY FROM THE UPPER LIMIT a b c ! I I I I I I 12 3 4 5 % % % % % 10% 40 = upper limit 40 — e = kiO .: k = e = first division of classes e = 40 — £40 40(1 - k) 20% 40 - e e = upper limit He) where (e) = 40 - &40 d = second division of classes e — He) = d e(l - ft) - d 40(1 - AX1 -k) = d upper limit d-c = k(d) where '<*) = 40O - ft)(l - k) 40(1 - fe)« third division of classes c = d - kd c = (1 - &)rf c = (1 - k)* 40 c = upper limit c — b = k(c) where (c) = 40(1 — &) J b = fourth division of classes b = c — kc b = (1 - k)* 40 upper limit fifth division of classes Hb) where (b) = (1 - &)« 40 (1 - &)» 40 30% a = upper limit . 1 = lower class limit .1 = Wo) where (a) = 40U .1 = a — ka .1 = (1 - &)M0 40% (1 - fc)« 40 1 -k Vlo k - l -\j 40 = .6318 Range 0.1-40.0% Size classes Per cent of total sales a = .2714 A .1000- .2714 6= .7367 B .2715- .7367 c= 1.9992 C .7368- 1.9992 rf = 5.4257 D 1.9993- 5.4257 9 = 14.7252 E 5.4258-14.7252 F 14.7253-40.0000 Table D-4 STEPWISE PRODUCTS OF MATRIX MULTIPLICATION Firms moving from: Firms moving to: 12.50 100.00 50.00 25.00 37.50 37.50 37.50 85.71 14.29 100.00 50.00 50.00 [61] 12.50 6.250 3.125 50.00 46.875 21.875 37.50 32.813 23.438 14.063 46.204 73.453 5.359 26.538 100.00 75.00 25.00 5.859 4.736 2 490 37.891 32.275 16.943 29.883 25.415 13.843 24.358 32.083 47.847 53.967 2.010 5.490 18.876 46.033 100.00 93.750 6.250 2.882 2.439 1.293 19.513 16.518 8.758 15.517 13.136 6.967 41.027 40.983 38.488 29.124 21.061 26.924 44.495 70.876 100.00 99.609 .391 .760 .643 .341 5.144 4.355 2.309 4.091 3.463 1.836 27.100 24.762 18.010 8.482 62.904 66.777 77.504 91.518 100.00 99.998 00 .002 .053 .045 .024 .358 .303 .160 .284 .241 .128 4.515 3.977 2.523 .719 94.790 95.435 97.165 99.281 100.00 99.999 .002 .002 .001 .002 .001 .001 .056 .005 99.940 99.969 99.969 99.995 100.00 99.999 [62] should have sales equaling about 5 per run economic projections (more than 25 cent of the average total annual chain sales years) are quite speculative. In view of for the period 1954-1962. Since total chain the dynamic nature of our food industry, sales in 2010 are likely to be much more such projections are even more precarious, than 1962 sales, these results seem reason- Nevertheless, short-term (5-10 years) ap- able (5 per cent refers to level of total praisal may prove useful in business plan- sales in 1962). ning, since it indicates the probable direc- Finally, it must be recognized that long- tion and magnitude of future changes. [63] APPENDIX E Table E-l COMPARATIVE PROFIT DATA, SELECTED INDUSTRIES, 1964, 1959, 1954 Retail food industry Food processing industry Durable goods industry Other con- sumer goods industry Agriculture Nation's (All food stores) (63 largest firms) (61 largest firms) (39 largest firms) (All farms) 500 largest industrial corporations 1964: dollars 20,583,834 254,048 29,860,031 1,019,900 85,230,345 18,003,273 5,768,288 1,328,232 41,700,000 6,700,000* 266,497,511 17,237,516 Net profit (after taxes) per cent Profit as percentage of sales. Profit as percentage of net 1.230 10.810 3.416 11.450 6.767 13.696 7.377 13.008 16.067 3.641 6.468 12.105 (28 largest firms) (61 largest firms) (85 largest firms) (34 largest firms) (All farms) 1959: dollars Sales 16,880,304 214,785 24,253,174 722,369 66,860,685 4,249,029 11,878,864 823,625 37,479,000 11,279,000* 197,394,885 Net profit (after taxes) 11,986,772 per cent Profit as percentage of sales . Profit as percentage of net 1.272 13.142 2.978 10.786 6.355 12.309 6.933 12.869 10.486 2.246 6.072 11.003 (29 largest firms) (67 largest firms) (84 largest firms) (35 largest firms) (All farms) 1954: dollars Sales 11,039,253 115,139 19,373,995 456,976 42,347,046 2,612,713 7,385,349 508,540 34,355,000 4,951,000* 136,782,913 Net profit (after taxes) 8,266,557 per cent Profit as percentage of sales. Profit as percentage of net 1.050 9.880 2.358 NA 6.169 NA 6.885 NA 14.411 3.482 6.043 NA * Method Used to Calculate Agriculture's Net Income Cash receipts from farm marketings . Government payments to farmers . . Nonmoney income 1963 billion dollars 36.9 1.7 3.1 Realized gross farm income . Farm production expenses. . Farm operators' realized net income. Net change in farm inventories Farm operators' total net income Farm operators' and family labor charge. 41.7 -29.2 12.5 .5 13.0 -6.3 Farm operators' actual total net income (profit) . 6.7 Source: Fortune, 1955, 1960, 1965; Oarlock, et al„ 1965; and U. S. Department of Agriculture, Farm Income Situation, 1964. [64] LITERATURE CITED Adamy, Clarence G. 1966. Executives jolted by draft report. Supermarket News 25(1 2): 25. Adelman, Irma G. 1958. A stochastic analysis of the size distribution of firms. American Statistical Assoc. Jour. 53(284): 893-904. Anderson, T. W., and L. A. Goodman 1957. Statistical inference about Markov chains. Annals of Math. Statistics 28:89- 110. Collins, N. R., and L. E. Preston 1960. Growth and turnover of food processing firms. Proceedings annual meeting of Western Farm Econ. Assoc, Stanford University, pp. 297-314. 1961. The size structure of the largest industrial firms, 1900-1958. American Econ. Rev. 51:986-1011. England, W. B. 1959. Operating results of food chains in 1958. Boston: Harvard Business School Tech. Bui. 156. Fairchild Publications, Inc. 1953-1963. Fairchild's financial manual. New York. Fortune Magazine 1955. 1960, 1965. Directory of the 500 largest U. S. industrial corporations. July issues. Garlock, Fred L., Phillip T. Allen, Lawrence A. Jones, William H. Scofield, and Harvey Shapiro 1965. The balance sheet of agriculture 1964. U. S. Dept. of Agr., Econ. Res. Serv., Agr. Inf. Bui. 290. Ginsburg, David 1965. What can we expect from government? A view to 1970: A five year look ahead. Proceedings of the 1965 mid-year conference, Super Market Inst., Wash- ington, D.C., pp. 5-9. Hamilton, W. E. 1959. A current look at the farmer's percentage of the consumer's food dollar. Nat. Planning Assoc, Special Report 55. Hart, P. E., and S. J. Prais 1956. The analysis of business concentration: A statistical approach. Royal Sta- tistical Soc Jour., Series A, 119:150-75. Hauser, Philip M. 1965. Is the market moving away from you? A view to 1970: A five year look ahead. Proceedings of the 1965 mid-year conference, Super Market Inst., Washing- ton, D.C., pp. 10-13. Hiemstra, S. J., and D. B. DeLoach 1962. Growth patterns in the retail grocery business. University of California Agr. Exp. Sta. Bui. 786:1-74. Judge, G. G., and E. R. Swanson 1961. Markov chains: Basic concepts and suggested uses in agricultural economics. Urbana: University of Illinois, Agr. Res. Report AERR-49. Kemeny, J. G., J. L. Snell, and L. A. Goodman 1957. Statistical inference about Markov chains. Annals of Math. Statistics 28:89- 110. Kemeny, J. G., J. L. Snell, and G. L. Thompson 1957. Introduction to finite mathematics. New York: Prentice-Hall, Inc., pp. 171- 77 and 217-32. Merrill Lynch, Pierce, Fenner and Smith, Inc. 1966. Security and industry survey — A guide for investors. Merrill Lynch, Pierce, Fenner and Smith, Inc., New York, p. 31. [65] Miklius, W., and D. B. DeLoach 1965. Do lettuce buyers exert oligopsony power? Agr. Econ. Research, U. S. Dept. of Agr. 17(4): 101-07. Mooney, Edward (ed.) 1964. Small chain boom. Chain Store Age 40(7). Mueller, Willard F., and Leon Garoian 1961. Changes in market structure of grocery retailing. Madison: University of Wisconsin. Musolf, Lloyd D. 1965. Government and economy. Chicago: Scott, Foresman, and Co., pp. 34-45. Padberg, Daniel I. 1962. The use of Markov processes in measuring changes in market structure. Jour, of Farm Economics 44(1): 189-99. Progressive Grocer 1952-1964. Annual reports, April issues. New York. 1955-1965. Facts in grocery distribution. New York. 1959. Super valu study. P. S-12. New York. 1963. Food distribution in the 1960's — a forecast. New York. 1 965. The newly weds. 44 ( 1 2) : K-60. New York. Sales Management 1953-1965. Survey of buying power. New York, N.Y. Supermarket News 1966. Several popular marketing tests questioned in the NCFM draft report. New York, N.Y., 15(12):25. This Week Magazine 1963. Contribution to profit. Tenth Biennial Grocery Study. New York. Tongue, William W. 1965. Competition in the affluent society. Jour, of Farm Economics, 47(2):433. U. S. Bureau of the Census 1954, 1958, 1963. Retail trade: U. S. summary. Census of Business, Washington, D.C. 1963. Statistical abstract of the United States, p. 5, Washington, D.C. 1965. Estimated population: 1900 to 1965. Statistical Abstract of the United States, p. 5, Washington, D.C. U. S. Congress 1959. Small business problems in food distribution, Part II, vols. 1 and 2. House subcommittee No. 5 of the Select Committee on Small Business, Hearings, 86th Congress, on House Resolution 51, San Francisco. U. S. Government Printing Office, Washington, D.C. U. S. Department of Agriculture 1957. Farm-retail spreads for food products. Agricultural Marketing Service, Market- ing Res. Div., Misc. Publ. 741, Washington, D.C. 1964. Farm income situation. Economic Research Service, FIS-195, Washington, D.C. 1965a. Developments in marketing spreads for agricultural products in 1964. ERS- 14, Washington, D.C. 19656. Farm-retail spreads for food products, 1947-1964. ERS-226, Washington, D.C. 1965c. U. S. food consumption: Sources of data and trends 1909-1963. Economic Re- search Service, Stat. Bui. 364:1-184, Washington, D.C. 1965