87-1693 TDOC April 1991 z TA245.7 F |.EANNESS AND CONVENIENCE DIMENSIONS Or BEEF PRODUCTS = “ an Exploratory Analysis Using Scanner Data The Tnxas Agricult ""~' Fvoeriment Station/ Charles J. Arntzen, Director/The Texas A&M University System/ College Station, Texas [Blank Page in Original Bulletin] Leanness and Convenience Dimensions of Beef Products: an Exploratory Analysis Using Scanner Data Oral Capps, Jr. & and Rodotfo M. Nayga, Jr.‘ " Respectively, professor and research assistant, Department of Agricultural Economics, Texas A8tM University, College Station. H KEYWORDS: scanner data, demand analysis, beef products. Contents Executive Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . lmroduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Literature Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .5 Nature of Scanner Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .7 General Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .7 Problems and Pitfalls . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .8 Present and Potential Uses in Economic Research . . . . . . . . . . . . . . . . . . . . . . . . . . . . .8 Data Source . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .9 Conceptual Framework tor the Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .10 Data Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .11 lndividualUPCs . . . . . . . ... . . . . . . . . . . . . . . . . ..'= . . . . . . . ..’ . . . . . . ..11 CustomerCounts.......;....;..,.... . . . . . . . . . . . . . . . . . . . . . . ..17 Advertisement Space . . . . . . . . . . . . . . . . . . “ . . . . . . . . . . . . . . . . . . . . . . . . . .17 Statistical Procedures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .20 Deletion of Particular UPCs and Data Anomalies . . . . . . . . . . . . . . . . . . . . . . . . . . . . .20 Descriptive Statistics (Individual UPCs) . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . .21 Descriptive Statistics (Commodity Groups) . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . .25 Econometric Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .30 "Empirical Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .31 Fresh Beet and Convenience Beef Products (Individual UPCs) . . . . . . . . . . . . . . . . . . . . . .32 Goodness-of-Fit and SerialCorrelation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .32 ‘yd Own-Price Elasticities . . . . . . . . . . . .’ . . . . . . . . . . . . . . . . . . . . . . . . . . . .33 Cross-Price Elasticities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .34 Own-Advertisement Elasticities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .35 Cross-AdvertisementElasticities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .35 Holidays and Seasonality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .36 Fresh Beet and Convenience Beef Products (Aggregate Groups) . . . . . . . . . . . . . . . . . . . . . 36 Goodness-ot-Fit and Serial Correlation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .37 Own-Price Elasticities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .37 Cross-Price Elasticities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .38 Own-AdvertisementElasticities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .38 Cross-Advertisement Elasticities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .38 Holidays and Seasonality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .39 Conclusions and Implications tor Further Research . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .41 Literature Cited . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .42 Appendix A " List of Individual UPC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .44 Appendix B ~ DescriptiveStatistics of Prices and Purchases per1,000 Customers for the 147 Beet Products . . . . .48 Appendix C EstimationofEoonometricModels . . . . .4 . . . . .3 . . . . . . . . . . . . . . . . . . . . . . .58 Y’ ‘= K's i. f“ l. .v .<}x\/. .»”\ f. l i l ' r’ . -. ~ 1,, ~ ,__ iwK \ \\ r,‘ Y Q1’. <15}. ‘ ‘ J i t: I ‘=- — Executive Summary Although scanner data have been available for several years to marketers, such data represent a new form of information for the meat industry. Marketers and researchers are just beginning to learn how to utilize this information source to make decisions about the meat case. lssues of convenience as well as of diet and health currently are major concerns for consumers. lndustr/y studies show that most consumers now choose foods that are quick to prepare (Morris, 1985).! Moreover, today's consumers are more conscious about diet, health, and nutrition than were yesterday's (Yankelovich et al., 1983, 1985; Burke Marketing Research, 1987). To meet consumer demands, arising in part from health concerns and salient lifestyle changes, the red meat industry is taking steps to foster the development of products that are not only lean but also quick, easy, and convenient to prepare. For instance, the industry is acknowledging these changes in consumer preferences by making available lean beef products, precooked meats, boneless cuts, and microwaveable entrees. Retailers, likewise, moved to reduce fat trim from one-half inch to one-quarter inch (Branson et al., 1986). ln this light, this study constitutes a pilot test of the use of scanner data to investigate the demand for lean, nonlean, and convenience beef products for a local market in Houston, Texas. Although the beef industry recognizes the new realities of the marketplace, little information exists on the factors affecting the demand for lean products and convenience products. More specific knowledge of consumer preferences is essential so that suitable production and marketing adiustments can be made. The determination of key demand variables will allow producers, processors, and distributors to analyze trends in retail markets, improve planning, and provide better service to consumers. In this report, the investigation of lean as well as convenience beef products rests on the use of scanner data from a retail firm in Houston. The time frame is the period January 1986 to November 1988. This particular pilot study reveals much about the potential utility of scanner data in market research on beef products. ln particular, this research demonstrates the feasibility of scanner data in developing econometric models to analyze sales of beef products at the retail level. Traditional analyses of retail demand have generally depended upon aggregate annual, quarterly, or monthly time-series data of purchases and prices. These data often do not represent current market conditions and are too general for product-specific decision making. Consumer panels and consumer surveys provide more detailed data for specific products as well as provide socio-dernographic information but are expensive methods of data collection. Scanner data, however, constitute a readily available, current, and timely source of product- specific information. Scanner data are not without limitations, which are (1) the sheer volume of information, (2) the lack of demographic and income information, and (3) the provision of information only for food eaten at home. Because of problems of data integrity and of too much detail, creating "data overload," empirical practitioners have been less than enthusiastic about the value of scanner data in economic research. Further, despite providing voluminous information, scanner data files must be augmented to monitor advertising or promotional activities as well as to monitor customer counts. importantly, for beet items, food stores supplying the data must have the equipment to generate labels enabling the products to be electronically scanned. ln regard to data integrity, it is unrealistic to expect the scanner to capture 100 percent of all sales. lnforrnation on most sales, however, with proper scmtiny from the retail food industry can be captured, and, consequently, these data may be used for market analysis. Work with scanner data is not a trivial task. Much careful and organized computation is necessary to conduct any analysis successfully using scanner data. Data anomalies are most certainly the rule rather than the exception, particularly for fresh products. This study rests on weekly point-of-sale purchases of 147 individual beef products: 30 lean fresh products, 70 nonlean fresh products, and 47 convenience products (prepared entrees). Additionally, this study considers aggregate commodities, namely, brisket, chuck, ground, loin, rib, round, and all other beef as well as convenience (prepared entrees) steak products, beef entrees, ground beef, beef ribs, and roasts. The weekly observations (150 in all) began on Wednesday and ended on Tuesday to conform to store sales and advertising patterns. The number of supermarkets in operation by this firm over the time interval of this study was 43. importantly, the retail food firm in this study caters to relatively high-income customers. Customer counts per week at this firm (43 supermarkets) ranged from 505,164 to 861,844 over the study period. The average customer count at this firm per week is on the order of 680,000. The advertisement information gathered over the period relates only to fresh beef products, not to convenience beef products. Consequently, in the analysis of convenience beef products (prepared entrees), no assessment of the impact of advertising on convenience beef item movement can be made. Advertisement space (in terms of square centimeters) torthe respective beef products varied considerably from week to week. Ground beef was the most frequently advertised product, whereas beef rib was the least frequently advertised product. On the basis of print space, ground beef received the most attention (on average 62 square centimeters), and rib received the least attention (on average 11 square centimeters). The advertisement frequency of nonlean beef products is three times that of lean beef products. As well, the print space for nonlean items is, on average, slightly more than 10 times that for lean items. Advertisement space for the aggregate of fish, pork, poultry, lamb, and veal items averaged almost 830 square centimeters weekly, roughly 2.5 times that for fresh beef products. As a general rule, lean beef products are more expensive than nonlean products. ln this study, both lean and nonlean products correspond to Choice grades. The lean brand for the firm studied is a Choice grade beef from which fat is trimmed. Lean line brands for other retail firms are generally no-roll, Good (Select) equivalent grades (e.g., "Giant Lean"). Good (Select) grades of meat products are typically priced below equivalent cuts. The average price of lean beef items in the aggregate is $3.47 per pound; in comparison, the average price of nonlean beef is $2.42 per pound. Put another way, the price premium for lean beef is on the order of 40 percent in this retail firm. Except for loin, the price of lean products exceeds the price of nonlean products. The price for lean brisket is about 1.4 times that of nonlean brisket; for rib, the price premium is 80 percent; for round, 30 percent; for ground, 50 percent; and for chuck, 20 percent. The top five lean line products in terms of average purchases per 1,000 customers are (1) gourmet ground round, (2) tailless T-bone steaks, (3) eye round roast, (4) sirloin tip fillets, and (5) beef cube steaks. The top five nonlean products are (1) ground beef chuck #079, (2) ground beef #078, (3) ground beef #080, (4) Choice boneless brisket #062, and (5) chuck boneless pot roast. Similarly, the top five convenience products are, respectively, (1) Armour Chicken Fry Beef Patties, (2) Armour Salisbury Steak, (3) Budget Sirloin Beef, (4) Budget Gourmet Oriental Beef, and (5) Budget Gourmet Pepper Steak with Rice. In the aggregate for this retail firm, the average purchase per 1,000 customers for lean products is almost 14 pounds per week. In comparison, the average purchase per 1,000 customers for nonlean products is about 336 pounds per week. The principal fresh beef product in terms of purchases per 1,000 customers is ground beef (nearly 170 pounds per week), whereas the least important product is rib (almost 20 pounds per week). The average purchase of convenience products (prepared entrees) per 1,000 customers is roughly 23 units. For convenience products, the key items in terms of movement are steak, ground beef, and entrees. The least important convenience items in terms of product movement are roasts and ribs. Budget shares represent the proportion of beef sales attributable to individual products. Within the class of convenience beef products, roughly 58 percent of dollar sales is attributable to steak items; 19 percent to entree items; 16 percent to ground beef items; 6 percent to roast beef items; and less than 1 percent to beef rib items. Collectively, 10 items account for slightly more than 48 percent of the sales of convenience beef products: (1) Armour Chick Fry Beef Patties, (2) Le Menu Sirloin Tips, (3) Le Menu Yankee Potroast, (4) Le Menu Chop Sirloin, (5) Stouffer Oriental Beef Lean Cuisine, (6) Budget Sirloin Beef, (7) Le Menu Pepper Steak, (8) Budget Gourmet Oriental Beef, (9) Classic Lite Steak Diane Mignonette, and (10) Le Menu Beef Stroganoff. Within the class of fresh beef products, by carcass section, ground beef constitutes roughly 37 percent of dollar sales; loin, 19 percent; round, 12 percent; rib, 10 percent; chuck, 6 percent; and brisket, 4 percent. importantly, roughly 6 percent of fresh dollar sales is attributable to lean beef items, whereas 94 percent is attributable to nonlean beef items. Collectively, 10 products account for approximately 65 percent of the sales of fresh nonlean beef products: (1) lean ground beef chuck #079, (2) fresh ground beef #078, (3) extra lean ground beef #080, (4) beef rib eye steak #037, (5) top sirloin steak boneless #032, (6) beef loin T-bone steak #029, (7) boneless strip steak #028, (8) beef chuck boneless pot roast #054, (9) beef round steak boneless #007, and (10) ground beef gourmet #081. Collectively, 10 products account for almost 77 percent of the sales of fresh lean beef products: (1) Lean Line Gourmet Ground, (2) Lean Line Extra Lean Boneless Stew Meat, (3) Lean Line Eye Round Roast, (4) Lean Line Sirloin 11p Fillets, (5) Lean Line Flank Steaks, (6) Lean Line Beef Cube Steaks, (7) Lean Line Sandwich Steaks, (8) Lean Line Shish Kabob, (9) Lean Line Ranch Broils, and (10) Lean Line Eye Round Steaks. Convenience beef products (prepared entrees) generate nearly $36,000 in sales per week. Fresh beef products, however, yield almost $600,000 in sales on a weekly basis. Lean beef products constitute $34,000 per week in sales, whereas nonlean beef products constitute $564,000 per week in sales. With few exceptions, purchases of beef products vary tremendously on a weekly basis. The purpose of econometric analysis in this study is to develop models to explain such variation in product movement. The dependent variable in the respective retail demand relationships is units of movement per 1,000 customers. The respective exogenous (independent) variables are ( 1) own-price, (2) prices of competing products, (3) advertisement variables, (4) seasonality, and (5) holidays. The purpose of the econometric analysis is to identify and assess factors affecting purchases per 1,000 customers. Emphasis is on price and advertisement elasticities. Price elasticities refer to percentage changes in purchases caused by unit percentage changes in prices; similarly, advertising elasticities refer to percentage changes in purchases caused by unit percentage changes in advertising. Observations of elasticities reveals the sensitivity of purchases to price changes and/or to promotion efforts. Remarkably, the models capture significant amounts of variation in purchases per 1,000 customers. Given the relatively large amount of variation to be explained as well as the absence of serial correlation, the econometric models are indeed satisfactory. ln this study, own-price elasticities for lean, nonlean, and convenience beef products are negative to correspond to the inverse relationship between purchases (movement) and price. Further, most of the elasticities are significantly differentfrom zero and have magnitudes greater than 1 in absolute value. Consequently, considerable sample evidence exists to indicate that own-price exerts a notable influence on purchases if everything else is held constant. Techni- cally speaking, the response to price changes is elastic. In fact, the magnitude of the price elasticities is much higher for convenience beef products than for fresh beet products, as expected. By carcass section, lean all other beef and lean round meat are particularly sensitive to changes in own-price. Nonlean brisket, nonlean chuck, nonlean rib, and nonlean round are also sensitive to changes in own-price as well. Finally, convenience rib, roast, ground beef, entree, and steak products are highly sensitive to changes in own-price. In regard to competing prices, purchases of lean beef products are generally not responsive to changes in the price of nonlean beef products. On the other hand, except for brisket, ground, loin, and rib cuts, purchases of nonlean beef are generally sensitive to changes in the price of lean beef. The price of nonlean beef is thus not a key determinant of purchases of lean beef, but the price of lean beef is a prime determinant of chuck, round, and all other beef. As well, the price of convenience products generally bears no relationship to purchases of lean and nonlean beef products. Likewise, except for convenience roast products, the prices of lean and nonlean beef do not significantly influence purchases of convenience beef products. The prices of nonbeef products (pork, poultry, and fish) affect only particular cuts of fresh beef (notably lean brisket, lean loin, and nonlean brisket). The price of poultry negatively influences purchases of convenience ground beef, and the price of fish positively influences purchases of entrees. For fresh beef products as well as convenience beef products, cross-cut prices thus have a relatively minor effect on purchase patterns. Own-advertisement elasticities are positive and in most cases statistically significant. Positive own-advertisement elasticities correspond to the direct relationship between pur- chases and advertising. Own-advertisement elasticities have more influence on purchases of nonlean beef products than on purchases of lean beef products. The magnitude of the own-advertisement elasticities is much smaller than the magnitude of price elasticities. The effect of cross-advertising is marginal. Advertising for fish, pork, poultry, lamb, and veal on purchases of fresh beef products is, in fact, not statistically significant. Similar to those of fresh lean beef, purchases of convenience beef products during holidays are smaller than purchases during nonholidays. However, purchases of nonlean beef during holidays are not significantly different from purchases during nonholidays. Finally, holding everything else constant, seasonal purchase patterns are evident for convenience beef groups (except entrees) and lean beef groups (except chuck and loin). However, only nonlean chuck and nonlean all other beef are subject to seasonality in purchases among the nonlean beef groups. Overall, this research encourages prospects of using scanner data in market research. Despite the apparent success of using scanner data to analyze retail demand relationships, concern lies with generalizing the results to regional or national levels. Scanner data from supermarkets in a particular location represent a "controlled" experimental situation. The community-specific results may not contribute to defensible, broad regional or nationwide inferences. Because of this potential limitation, the results of this analysis should be used not on a stand-alone basis but as supporting evidence in conjunction with a research approach designed to conduct analyses with scanner data on a regional or national basis. Though much recent empirical and theoretical work exists on demand and market analyses, reliable estimates of demand parameters for individual beef commodities are few. With the use of scanner data, retail demand relationships for beef products can be effectively analyzed. Use of scanner data can expand demand analyses. The realization of benefits from the use of scanner data is in the embryonic stage of development, however. In the next decade, analysts will concentrate on scanner data assembly, management, and analysis. Scanner data hold great promise for developing insights in market research. Conceivably, with proper manage- ment, scanner data may well be the ultimate data source of demand and market analyses at the retail level. This particular pilot study highlights the potential utility of scanner data in market research on beef products. »Q\ ;.\ ISSUGS of convenience as well as diet and health (especially those related to fat content) warrant atten- tion in the investigation of the appeal of beef to con- sumers in the United States. Today, consumers want the food they buy to be easy and quick to prepare, a dramatic change from previous times. New technology in food preparation, especially microwave ovens, and concomitant innovations in food processing continue to decrease the time needed for at-home meal prepara- tion. Industry studies show that most consumers now choose foods that can be prepared in less than 20 minutes (Morris, 1985). Consumer national attitudinal research, sponsored by the National Live Stock and Meat Board, indicates that today's consumers are more conscious about diet, health, and nutrition (Yankelovich Y et al., 1983, 1985; Burke Marketing Research, 1987). To meet consumer demands, caused in part by health concerns and lifestyle changes, the red meat industry is taking steps to foster the development of products that are not only lean but also quick, easy, and convenient to prepare. The industry acknowledges fchanges in consumer preferences by making available j lean beef products (closely trimmed choice beef or beef neless cuts, and microwaveable entrees. Motivated ‘tit from carcasses having less fat), precooked meats, “ y the 198s National Consumer Retail Beef (NCRB) study (Branson et al., 1986), retailers reduced fat trim from 1/2 inch to 1/4 inch. Packers followed suit by reduc- ing the standard 1 inch of outside fat to 1/2 inch. Increas- ingly more meat departments are offering consumers a lean "house brand" in addition to Choice grades (Decisions Center, lnc., 1987). Montford of Colorado has been developing "high-quality, convenient" products for the past several years (Wall Street Journal, This section documents the sparse number of studies dealing with the demand for lean and/or convenience beef products. Several studies have been conducted recently to examine consumer attitudes and preferen- ces toward beef. The NCRB (Branson et al., 1986)'study concentrated on the effects of different degrees of beef leanness on consumer demand. Skaggs et al. (1987) and Menkhaus et al. (1988) analyzed the potential of marketing branded, low-fat, fresh beef. The results of these studies indicated that (1) consumer health con- cerns pertaining to the ingestion of animal fats were‘ evident, (2) for a product that was perceived to be more . ‘healthy, consumers were willing to compromise on Introduction Literature Review 1985). The Beef industry Council currently lists the development of value-added beef products resulting from innovations in preparation or packaging as a key research area (personal communication). Increases in real income, declines in household size, and increases in the proportion of women in the work force have contributed to the outward shift in demand for added convenience (products that transfer the time and activities of preparation from the consumer to the processor) in foods purchased for home use (Stafford and Wills, 1979; Capps et al., 1985). Convenience attributes of poultry and seafood products are highly evident in the marketplace. The poultry industry in par- ticular, which increasingly sells processed forms that are easy to prepare, has been in the forefront of this development (The Food Institute, 1986). \ ' fiRecent trends in food consumption indicate an in-s i, creased awareness about nutrition and an increased interest in convenience foods. Not surprisingly then‘, consumer segments exist that prefer lean, low-fat products (Menkhaus et al., 1988; Skaggs et al., 1987) and/or convenience products (Capps and Pearson,‘ 1986; Capps, 1989). Although the beef industry recog- nizes the new realities of the marketplace, little informa- tion exists on the factors affecting the demand for lean beef products and convenience beef products. This research reported herein attempts to fill this void. More specific knowledge of consumer preferences is essen- tial so that suitable production and marketing adjust- ments can be made. The determination of demand variables will allow producers, processors, and dis- tributors to anticipate trends in retail markets, to improve planning, and to provide better consumer service. / taste, and (8) health-related factors influenced the y, decision to purchase leaner meats. A study prepared by Decisions Center, Inc. (1987) for the American Meat Institute focused on the awareness and usage of the lean brand of beef (Giant Lean) offered by Giant Foods, lnc., a chain of stores in the Baltimore, Maryland, and Washington, D.C., area. The particular brand under study was popular with customers who were women, employed, under 40 years old, who had children, and who were concerned about health and nutrition. This study, conducted in November 1986, was based on 300 telephone interviews of customers of the firm. Capps et al. (1985) identified several demographic and psychographic characteristics of consumers who buy lean meat products from a particular retail food chain in Houston, Texas. The source ol data was survey information, gathered by telephone interviews, from 200 shoppers. The analysis was performed using a Probit model. The survey indicated that consumers more than 30 years of age were more likely to buy lean meat products than were consumers 20-29 years of age. Residents of Texas tor more than 10 years were more likely to buy lean meat products than were residents of Texas for less that 10 years. Consumers who attended college were more likely to buy lean meat products than were consumers who had not attended college. Household size and the probability of buying lean meat products were positively associated. Fat-conscious consumers were more likely to buy lean meat products than were nonfat-conscious consumers. There was, however, no statistically significant link between income class of consumers and the likelihood of buying lean meat products. Furthermore, no statistically significant relationship was evident between price consciousness and the likelihood of buying lean meat products. The National Academy of Sciences (Lemieux and Wohlgenant, 1988) suggests that the real solution to human consumption of excessive dietary fat, saturated fatty acids, and cholesterol lies in the production of leaner animals. Using market survey data from a nation- al telephone survey of 200 consumers, Lemieux and Wohlgenant (1988) estimated that the premium con- sumers would be willing to pay lor 10 percent leaner pork was, on the average, 16.6 cents (with a standard deviation of 4.3 cents). Using the 1977-78 Nationwide Food Consumption Survey (NFCS) as the data source, Capps (1989a) addressed the issue of added convenience on the at- home demand for beef, steaks, roasts, and ground beef. The average weekly money value per household for convenience products was roughly $0.37 for beef, $0.11 tor steaks, $0.06 for roasts, and $0.09 for ground beet. For nonoo nvenience beet products, the average weekly money value per household was $5.82 for beef, $2.27 for steaks, $1.45 for roasts, and $1.73 for ground beef. For convenience beef products, more than 90 percent of the sample households reported zero expenditure levels. This descriptive evidence confirms that con- venience beet products for at-home consumption were scarce even in the late 1970s. Income was statistically important in affecting house- hold expenditures on convenience and nonconvenience beef products. Except lor roasts, income elasticities were greater in magnitude in the nonconvenience class than in the convenience class. In the convenience class, the income elasticity for beef was 0.0939; for steaks, roasts, and ground beef, the income elasticities were, respectively, 0.1270, 0.1418, and -0.2261. In the non- convenience class, the income elasticities for beef v' 0.2404; for steaks, roasts, and ground beef, the inccKJV elasticities were, respectively, 0.1644, 0.0997, and 0.0446. Household expenditures on beef products were, however, more sensitive to changes in household size then to changes in income. Household size elas- ticities for beef products in the convenience class were as follows: 0.3737 for beef, 0.5866 for steaks, 0.2737 for roasts, and 0.5980 for gound beet. ln the noncon- venience class, the household size elasticities were 0.7542 for beef, 0.4621 for steaks, 0.3310 for roasts, and 0.6179 for ground beef. ln general, various demographic variates greatly in- fluenced the demand for convenience and noncon- venience beef products. College-educated household managers, unemployed household managers, female household managers, and household managers less than 35 years of age spent significantly less on beef products than their counterparts. Regional and seasonal purchase patterns were evident. Finally, the purchase patterns of nonwhite households were noticeably different from the purchase patterns of white households. Capps (1989a) also forecasted percentage changes of nominal expenditures for convenience and noncor venience beef products over the period from 1980 2000. Growth in convenience beef expenditures was projected to be almost 40 percent, 2.5 times the growth in nonconvenience beef expenditures. Convenience steak and roast expenditures were expected to grow by roughly 35 percent, about 1.5 times the increase in nonconvenience steak and roast expenditures. Finally, growth in expenditures on both nonconvenience and convenience ground beef was projected to be on the order of 40 percent. During the 1980s, a myriad of convenience foods have been introduced into the marketplace. Further research in the area of convenience dimensions in food products is certainly desirable, especially given Capps’ (1989a) projections for beef products. In this report, the investigation of lean as well as convenience beef products rests on the use of scanner data from a retail food firm in Houston. The time frame in question is the period from January 1986 to Novem- ber 1988. Although the application of scanner data for demand analyses is in the embryonic stage of develop- ment, scanner data have been used in market research to investigate brand differentiation (Blattberg and Wis- niewski, 1986; Shugan, 1987; Guadagni and Little, 1983) and to investigate promotional effects on sales of performance (Wittink et al., 1988; Moriarty, 1985). Of particular interest to the beef industry are several cent). First, retail demand relationships for‘ steak, rior applications of scanner data (although very ' ‘round beef, roast beef, chicken, pork chops, ham, and pork lion were examined by Capps (1989b) using scan- ner data. This research demonstrated the feasibility of scanner data in developing short-run predictive models to anticipate sales of meat products. As well, the Center for Agricultural and Rural Development (CARD) at lowa State University, under contract with the Beef industry Council of the ‘National Live Stock and Meat Board (NLSMB), conducted analyses of behavior scan data (Schroeter, 1988). This scanner information was com- piled for the NLSMB by the Chicago-based marketing reseach firm Information Resources, Inc. The motiva- tion of the use of such data was to measure beef General Description Demand analyses require the existence of high- quality data bases. Fundamental elements affecting quality include adequate measures of response vari- bles (sales or consumption levels as well as budget -, ares), adequate measures of exogenous variables, sufficient number of observations, and appropriate time interval. The introduction of scanning check-out systems into U.S. supermarkets in the mid-1970s opened tremen- dous possibilities for generating new data and for using such data in economic research and managerial decision making. According to the Food Marketing ln- stitute (FMI), slightly more than 50 percent of the super- markets in the United States currently employ scanner check-out systems (Progressive Grocer, 1989). Impor- tantly, use of scanner data as a basis for demand analysis has been very limited. Only since 1979 have scanner data, through refinements by manufacturers of electronic scanning check-out systems by retail users, been generated with enough reliability and consistency for application in economic research (Jourdan, 1981). Scanner information constitutes a nontraditional data source for economic applications. The richness of scan- ner data lies in the daily available information on quan- tity, price, and hence expenditure fora multitude of products. The 35,000 to 40,000 items currently avail- able in retail food stores testify to the vastness of scanner data. Scanner data, however, are not within the realm of he public sector. Scanner data series useful for emand analyses are developed and maintained by private sources and are available from several firms Nature of Scanner Data consumption responses to television promotion and advertising. Using scanner data, fresh beef purchases of approximately 1,800 households were monitored in Grand Junction, Colorado, overthe period from October 1985 to July 1987. Combined with the detailed demographic information available for each of the households, scanner data provide a unique capability to assess the impact of the experimental television adver- tising. The beef products under investigation were (1) steaks for braising (chuck steak, round steak), (2) steaks for broiling (most loin steaks, rib steaks), (3) roasts for braising (chuck roast, round roast), (4) roasts for roasting (tenderloin roasts, rib roasts), (5) ground beef (all ground beef including ground round and chuck), and (6) beef for stewing/simmering (stew meat, brisket). who provide primarily information services (e.g., Infor- mation Resources, lnc; The Text Marketing Group; Burgoyne, lnc; A.C. Nielsen; The NPD Group). Scanner data are also available from retail food firms (e.g., Kroger and Safeway). Traditional analysis of consumer demand has generally depended upon aggregate annual, quarterly, or monthly time-series data of consumer purchases and prices. These data often do not represent current market conditions and typically are too general for product-specific decision making. Time-series data, in short, lack disaggregate product and price detail. Con- sumer panels and consumer surveys provide more detailed data for specific products as well as provide socio-dernographic information but are expensive methods of data collection. A key limitation of consumer panels or surveys is their lack of price information. Prices must be imputed from reported quantity and expenditure figures. Analysts question the use of such imputations, particularly estimation of cross-sectional demand functions (Cox and Wohlgenant, 1986). Another key limitation of the use of consumer surveys (not necessarily panels) is the lack of time continuity. To illustrate, the U.S. Department of Agriculture sponsors the National Food Consumption Survey (NFCS). Since its inception in 1936, this survey takes place only once approximately every 10 years (e.g., 1965-66, 1977-78, 1987-88). The U.S. Bureau of Labor Statistics (BLS) sponsors continuing consumer expenditure surveys (making available household panel data since 1980) on a quarterly basis. This source of data from the public sector, a landmark for consumer demand analysis, cir- cumvents the time continuity problem, but nonetheless, data sets from BLS lack price information and product- specific quantity information. Scanner data, on the other hand, constitute a readily available current and timely source of product-specific information. To quote Tomek (1985), “existing second- ary data seem especially inadequate for studying product demand in retail markets, and fundamental work needs to be done to obtain relevant data" (pp. 913-914). "The data associated with computerized checkout systems in grocery stores could become an important source of information for studying retail demand" (p. 913). Scanner data are not without limita- tions, however. The limitations of scanner data are threefold: (1) the sheer volume of information, (2) the lack of demographic and income information, and (3) the provision of information only forfood eaten at home. Problems and Pitfalls Because of problems of data integrity and of too much detail, creating "data overload," empirical prac- titioners have been less than enthusiastic about the value of scanner data in economic research. Each week as few as 10 to 20 supermarkets will generate the equivalent amount of data as would a panel of 10,000 households. Consequently, considerable resources are necessary to reduce the mass of data to useful summary figures for demand analyses. Additionally, data from public agencies are readily available to researchers; data from private firms are not, or if available, only at considerable cost. Despite the volume of price, quantity, and expendi- ture information, scanner data, at least from retail food firms, lack the dimension of consumer sociodemo- graphic data. This sociodemographic information is essential to the derivation of income elasticities. For demand analyses based on scanner data from food stores, the experimental unit is the individual food store (aggregation over consumers), not the individual con- sumer. This aggregation problem may not necessarily be negligible. lf the food store corresponds to a more or less homogeneous group of consumers, however, this aggregation problem is virtually of no consequence. Further, despite their sheer volume of information, scanner data files need to be augmented to monitor advertising or promotional activities. Competitors‘ ac- tions are also important but are extremely difficult to anticipate, measure, and evaluate. Additionally, difficul- ties exist in the representation of nonprice effects (mer- chandising schemes, coupons, services, cleanliness, product selection, and reputation for fresh meat or produce). Consequently, the ceteris paribus (all-other- things-held-constant) assumption (popular with econo- mists) is in jeopardy with the use of scanner data. importantly, for meat, poultry, and fish items as well as for produce, food stores supplying the data must have the equipment to generate labels enabling t‘ products to be electronically scanned. This equipm » is expensive, sensitive, and may not always produce scannable labels. Thus, because of the inability of par- ticular food stores to scan fresh meat or produce, scan- ner data for meat or produce may not be available or if available, not reliable. Fresh meat and produce, how- ever, constitute a sizable chunk of the food dollar per consumer. In regard to data integrity, food industry observer Richard E. Shulman makes this point: "...caveat about scanning data: lt‘s not accurate. lt is representative. Don't expect the scanner to capture 100 percent of all sales. There are dozens of reasons that sales are "lost": bad symbols, poorly trained checkers, etc. The impor- tant thing to understand is that most sales will be cap- tured and the resulting data can be acted upon" (National Grocers Association Technology Newsletter, 1985) Lesser and Smith point out (1986) that scanner data misrepresent item movement (quantity purchased) if the scanning file is not rigorously maintained or if the items cannot be or are not scanned and the Universal Product Codes (UPC) are not entered manually. Furthermore, scanner data may not provide accurate information stock shrink accounts for a substantial portion of th f movement of a product. Because stock shrink generally contributes approximately 1 to 2 percent of supermarket sales, this factor should not be a major issue forthe vast range of products. Consequently, the integrity of the data is a function of the level of discipline of the retail firm in capturing accurate information. Along this line, Lesser and Smith (1986) conducted a study to evaluate the accuracy of scanner data. Their results suggested that "substantial error is possible when examining individual items on a weekly basis. This factor should be considered when using scanner data" (p. 71). Present and Potential Uses in Economic Research Tremendous possibilities exist forthe generation and use of scanner data for applications to economic re- search. Examples of such applications include evalua- tion of shelf space allocation, evaluation of advertising and promotion schemes, evaluation of new items, and estimation of price and total expenditure elasticities. ln fact, as Lesser and Smith (1986) point out, with scanner data, "it is possible to do retail-level analysis routinely which previously required special tabulations" (p. 69). Examples of retail-level analyses requiring special tabulations include in-store pricing experiments (Doyle and Gidengil, 1977), the effects of promotional > _ n, 1974),the measurement of price elasticities(Funk Qbgrams on individual items (Hoofnagle, 1965; Cur- ~ ct al., 1977; Marion and Walker, 1978), the results of space allocation and display (Cox, 1964; Curhan, 1973; Chevalier, 1985), and the effects of interactions among short-run strategy variables such as advertising, space allocation, and pricing (Curhan, 1974; Wilkerson et al., 1982) Except for the work by Jourdan (1981 ) as well as the work by McLaughlin and Lesser (1986), few analyses of consumer demand have been conducted using scanner data. Jourdan (1981) estimated own-price and cross- price elasticities of demand for specific retail cuts of beef (roasts, steaks, ground beef, and nonground beef) by using bi-weekly data over a 25-week period from four retail food stores in Houston. McLaughlin and Lesser (1986) reported on the ex- periment of systematically varying prices and tracking, through the use of scanner data, subsequent movement of potatoes. With this approach, the researchers were able to calculate appropriate store-specific demand elasticities. For potatoes, data over a 42-week period from eight retail food stores in upstate New York indi- cated that consumer response to price changes was cities to assess impacts of promotional activity, to determine optimal space allocation, and to develop sales management models. McLaughlin and Lesser’s (1986) results also suggest that "pricing according to individual stores, rather than according to historical ‘relatively elastic. Retailers could use store-specific elas- Scanner data are primary data that have properties similar to cross-sectional and time-series data. The observations exist over time, usually days, as well as across various cross-sectional units, typically food stores. The source of data for the analyses in this study, similar to the Jourdan (1981) study, is a retail food firm in Houston. The time frame is from January 1986 to November 1988. Weekly observations began on Wed- nesday and ended on Tuesday to conform to store sales and advertising patterns. The number of supermarkets in operation by this firm over this time interval was 43. importantly, the retail food firm in this study caters to relatively high-income customers. Assessment and evaluation of the use of scanner data applied to demand analyses involve several steps. Nearly 40,000 items are currently available in this retail ‘food firm. To ensure computational feasibility, the data source used in this study involves only beef items. Data Source price zones, may be an appropriate profit-maximizing strategy" (p. 9). The common thread in the two-con- sumer demand applications is the interaction with a single firm (although multiple stores) in a local area. Scanner data from the supermarkets in a particular location (for this analysis Houston) presumably repre- sent a "controlled" experimental situation. importantly, however, the community-specific results may not allow defensible, broad regional or nationwide inferences. Because of this potential limitation, the results of local analyses should be used not on a stand-alone basis but as supporting evidence in conjunction with a research approach designed to conduct demand analyses with scanner data on a regional or national basis. Nevertheless, demand analyses can be expanded through the use of scanner data. Though much empiri- cal and theoretical work exists with respect to demand analyses in recent years, reliable estimates of demand parameters for disaggregate food commodities are few. Scanner data may result in the most detailed and defini- tive source of retail food industry statistics available to researchers. However, the realization of benefits from the use of scanner data is in the embryonic stage of development. To paraphrase Branson et al. (1986), the mid-1980s to the mid-1990s will be the learning years for scanner data assembly, management, and analyses. Scanner data hold great promise for develop- ing insights into both applied and theoretical research. Conceivably, with proper management, scanner data may well be the ultimate data source for demand analysis at the retail level. Nonetheless, this data sburce constitutes information for roughly 300 Universal Product Codes (UPCs). Im- portantly, beef products not only are key contributors of sales volume and profit to the firm but also are key elements of the consumer market basket of goods. Scanner data are also available on a daily basis. Aggregation of daily information into weekly information is essential to make computations more manageable. This weekly information also allows for better repre- sentation of store operations. To illustrate, price chan- ges are usually initiated once per week, and store merchandising activities such as newspaper advertise- ments and displays are also usually done weekly (Car- men and Figueroa, 1986). Aggregating observations into longer time intervals also tends to smooth out variability. This study is based on point-of-sale purchases. At- tention is centered on disaggregate beef products, par- ticularly lean and convenience (prepared entrees) items. For documentation of individual UPCs for the respective beef products, see the section titled "Data Description." Pounds sold of the UPC as well as price of the UPC are reported by week for the period in question. For commodity aggregates, the quantities of the various items correspond to the sum of the respec- tive quantities of the relevant UPCs. The implicit prices of the commodity aggregates are weighted averages of all individual UPC prices. The weighting mechanism is the ratio of the sum of all sales over the UPCs to the sum of all quantities. Quality effects may result from such commodity ag- gregation (Houthakker, 1952; Cox and Wohlgenant, 1986). When distinct items are aggregated into com- modity groups, variations occur in the implicit prices. Furthermore, the weighted average prices change with the quantities of the component goods consumed. Al- though the use of implicit prices potentially limits the analysis, given that the beef products in question are relatively homogeneous, quality effects attributable to commodity aggregation are assumed to be negligible. Holdren (1960, pp. 117-123) provides the conceptual framework forthis analysis. Attention is on multiproduct retail demand functions. According to Holdren (1960, p. 123) "the multiple product retail demand function can be characterized by qi = fi(p1, p2, ..., p“, a1, a2, ..., am), (1) where q represents quantity variables expressed in appropriate units, p represents price variables, and a represents attributes of the retailer’s nonprice offer variation. Advertising, sales promotion activities, hours open, and customer services are concrete examples of nonprice offer variation. Additionally, equation 1 may be augmented by considering in-store and competitors’ prices as well as in-store and competitors’ advertising. Changing effective demand related to nearness to payday is a well-known phenomenon in food retailing (Marion and Walker, 1978; Carmen and Figueroa, 1986). Marion and Walker (1978), for example, found that weekly retail meat sales tended to decrease as time since the last payday increased. Seasonal factors also may affect the quantity variables, all other things held constant (Marion and Walker, 1978; Funk et al., 1977; Carmen and Figueroa, 1986). Finally, because they are proxies for tastes and preferences of the collection of consumers who frequent retail stores, the socio- demographic influences in retail demand functions are worthy of consideration. Conceptual Framework for the Analysis 10 Emphasis in our study is on demand relationships at the firm level in lieu of the store level. The prices for each UPC are the same across stores, and sales of me items at the stores are reasonably similar. Hence, d ‘ from all stores in the firm are aggregated to form 150 weekly time-series observations. Funk et al. (1977) examined factors affecting weekly sales of carcass beef and individual beef cuts at two retail food chains in Toronto, Canada. Their analysis used data taken on shipments of beef carcasses, quarters, and primals during a 72-week period. Marion and Walker (1978) used data based on point-of-sale purchases to examine the sales of five meat products (beef round, beef chuck, beef loin, pork loin, and fryers) of two Ohio supermarkets during a 52-week period. The Funk et al. (1977) and Marion and Walker (1978) studies, however, were not dependent upon the use of scanner data. Our study, therefore, deviates from traditional analyses because it examines the potential utility of scanner data in market research on beef products. In light of the previous discussion, the generiu specification of the respective demand models in this study is as follows: Q1, = KP“, P1,, PFISI-It, PPORKt, PPOULTt, SEASON, ADV“, ADVl-t, ADVAOMt). (2) where Qit is purchases per 1 ,000 customers (in pounds) of beef item i in week t; t = 1, . . ., 150; Pit is price of beef product i in week t ($/pound); Pj: corresponds to prices of competing beef products (j refers to the set of com- peting products) in week t ($/pound); PFlSHi, PORKt, and PPOULTi correspond to weighted average prices of fish, pork, and poultry products, respectively, in the retail firm in week t. Wohlgenant (1985) argues for the inclusion of these price variables in demand relation- ships for beef. H refers to a binary variable for holidays (H = 1, if holiday; 0 otherwise); SEASON corresponds to a set of monthly binary variables to measure seasonality; ADV“ corresponds to the amount of print space given for beef product i in the weekly advertise- ment flier (square centimeters); ADV); corresponds to the amount of print space given forthe set of competing beef products in the weekly advertisement flier (square centimeters); and ADVAOMt corresponds to the amount of print space given to fish, lamb, pork, poultry, and veal products (competing meat products) in the weekly ad- vertisement flier (square centimeters). Q Data are converted to a per customer basis. Conse- 00 customers. Because of unavailability of informa- wently, the dependent variables reflect purchases per " "Ton, the model specification excludes competitors’ prices and advertising as well as socio-demographic variables. The variables Pit and P)"; capture own-price and cross-price effects. Own-price effects are hypothesized to be negative. Cross-price effects may be negative or positive to reflect substitutable or complementary relationships among the commodities in question. For disaggregate analyses, the identification of appropriate substitutes or complements a priori is a difficult task. ln this study, cross-price effects are of two types: (1) cross-cut prices and (2) cross-product prices. The former refer to competing beef products, and the latter refer to competing meat products. Because data are from only a single firm, some may argue from the following rationale that price elasticities are not estimable: (1) consumers can respond to price changes by shopping at different stores within a market area, and (2) no information in this study is available on purchases at other stores or on prices charged at other stores. According to the Food Marketing Institute, how- ever, only 27 percent of shoppers compare prices from Marion and Walker (1978) study, our study does not delete observations because of holidays. Monthly dummy variables capture the effects of seasonality. The coefficients associated with these variables may be either positive or negative. As in the Funk et al. (1977) study as well as in the Marion and Walker (1978) study, local newspaper ad- vertising is the only advertising mode considered in our study. Although television, radio, and in-store displays are used by the food store chain, these forms are primarily oriented toward creating a favorable corporate image. Newspaper advertising, on the other hand, is geared primarily to promoting specific products. The basic format and design of the newspaper advertise- ments used by the chain were the same throughout the period. Therefore no measure of "creative aspects" of advertising is necessary. In the Funk et al. (1977) study as well as the Marion and Walker (1978) study, adver- tising data corresponded to the number of advertised items. In our study, advertising data refer to the amount of print space devoted to each item. This study allows the examination of own- and cross- advertisement effects. All other things held constant, own-advertisement effects are hypothesized to be posi- tive, whereas cross-advertisement effects are hypothe- sized to be negative. The respective set of advertise- _ t al. (1977) reported that (p. 534) "multicollinearity ment variables used in the retail demand relationships ‘ etween competitors’ prices and in-sto re prices was too Qtore to store (Cox and Foster, 1985). Additionally, Funk correspond to the set of price variables previously dis- strong to allow for measurement of the separate effects of the variablesflTherefore, in this study, the omission of competitors‘ prices may not be a limiting factor in estimating in-store price elasticities. A dummy variable is used to capture the effects of holidays on per customer beef purchases. Unlike the This section of the report deals with three com- ponents: (1) data for individual UPCs, (2) documenta- tion of customer counts by week, and (3) documentation of advertisement space for beef products. Pulling together price/quantity information on individual UPCs, customer counts, and advertisement space was an exacting task. Individual UPCs Examples of data for individual UPCs are provided in Table 1 (for Lean Line Sirloin Strips) and Table 2 (for Le Menu Beef Stroganoff). The format for all UPCs is similar. importantly, price and quantity information are 0t necessarily available for all UPCs for all weeks. Some products (especially microwaveable entrees) Data Description 11 cussed. Competitors’ advertising is excluded because of resource constraints. Furthermore, because Funk et al. (1977) reported that the impacts of competitors’ advertisement were not statistically significant, this set of variables may be marginal. were not available until well after January 1986, the initial month of the period in question. Other products were available at week 1 of the analysis but were discontinued because of lack of demand. ' A great number of UPCs correspond to beef products. For a description of the various UPCs, see Appendix A. For a schematic diagram of the UPCs, see Figure 1. According to this diagram, the number of fresh beef products is 100, and the number of convenience beef products is 47. Out of the 100 fresh products, 30 are lean products, whereas the remaining 70 are non- lean products. The numbers in parentheses below the beef types correspond to the number of UPCs in the category. Table 1. Data tor individual UPCs (example: Lean Line Sirloin Strips). UPC Units Price Cost Date Week Description 20102000000 45 629 28605 12186 3 Lean Line Sirloin Strips Q/ 20102000000 890 769 684410 12886 4 Lean Line Sirloin Strips 20102000000 523 769 402187 20486 5 Lean Line Sirloin Strips 20102000000 278 769 213782 21186 6 Lean Line Sirloin Strips 20102000000 423 769 325287 21886 7 Lean Line Sirloin Strips 20102000000 503 769 386807 22586 8 Lean Line Sirloin Strips 20102000000 366 769 281454 30486 9 Lean Line Sirloin Strips 20102000000 252 769 193788 31186 10 Lean Line Sirloin Strips 20102000000 i 248 769 190712 31886 11 Lean Line Sirloin Strips 20102000000 ' 143 769 109967 32586 12 Lean Line Sirloin Strips 20102000000 162 769 124578 40186 13 Lean Line Sirloin Strips 20102000000 218 769 167642 40886 14 Lean Line Sirloin Strips 20102000000 147 769 113043 41586 15 Lean Line Sirloin Strips 20102000000 85 769 65365 42286 16 Lean Line Sirloin Strips 20102000000 221 769 169949 42986 17 Lean Line Sirloin Strips 20102000000 164 769 126116 50686 18 Lean Line Sirloin Strips 20102000000 154 769 118426 51386 19 Lean Line Sirloin Strips 20102000000 174 769 133806 52086 20 Lean Line Sirloin Strips 20102000000 239 769 183791 52786 21 Lean Line Sirloin Strips 20102000000 173 769 133037 60386 22 Lean Line Sirloin Strips 20102000000 214 769 164566 61086 23 Lean Line Sirloin Strips 20102000000 187 769 143803 61786 24 Lean Line Sirloin Strips 20102000000 146 769 112274 62486" 25 Lean Line Sirloin Strips 20102000000 222 699 155178 70186 26 Lean Line Sirloin Strips 20102000000 216 699 150984 70886 27 Lean Line Sirloin Strips 20102000000 314 709 222626 71586 28 Lean Line Sirloin Strips 20102000000 164 709 116276 72286 29 Lean Line Sirloin Strips 20102000000 180 699 125820 72986 30 Lean Line Sirloin Strips \4/ 20102000000 212 699 148188 80586 31 Lean Line Sirloin Strips 20102000000 180 699 125820 81286 32 Lean Line Sirloin Strips 20102000000 ' 214 699 149586 81986 33 Lean Line Sirloin Strips 20102000000 151 699 105549 82686 34 Lean Line Sirloin Strips 20102000000 280 699 195720 90286 35 Lean Line Sirloin Strips 20102000000 227 699 158673 90986 36 Lean Line Sirloin Strips 20102000000 199 699 139101 91686 37 Lean Line Sirloin Strips 20102000000 32 699 22368 92386 38 Lean Line Sirloin Strips 20102000000 31 679 21049 93086 39 Lean Line Sirloin Strips 20102000000 34 639 21726 100786 40 Lean Line Sirloin Strips 20102000000 41 639 26199 101486 41 Lean Line Sirloin Strips 20102000000 30 639 19170 102186 42 Lean Line Sirloin Strips 20102000000 30 639 19170 102886 43 Lean Line Sirloin Strips 20102000000 36 639 23004 110486 44 Lean Line Sirloin Strips 20102000000 28 639 17892 111186 45 Lean Line Sirloin Strips 20102000000 34 639 21726 111886 46 Lean Line Sirloin Strips 20102000000 41 639 26199 ' 113086 47 Lean Line Sirloin Strips 20102000000 9 639 5751 120286 48 Lean Line Sirloin Strips 20102000000 39 639 24921 " 120986 49 Lean Line Sirloin Strips 20102000000 18 639 11502 121686 50 Lean Line Sirloin Strips 20102000000 21 639 13419 122386 51 Lean Line Sirloin Strips 20102000000 _ 21 639 13419 123086 52 Lean Line Sirloin Strips 20102000000 31 639 19809 10687 53 Lean Line Sirloin Strips 20102000000 31 639 19809 11387 54 Lean Line Sirloin Strips 20102000000 38 639 24282 12087 55 Lean Line Sirloin Strips 20102000000 37 639 23643 12787 56 a Lean Line Sirloin Strips (continued) 1 » 12 Table 1 (continued). , UPC Units Price Cost Date Week Description “20102000000 18 ' 639 11502 20387 57 Lean Line Sirloin Strips 20102000000 21 639 13419 21087 58 Lean Line Sirloin Strips 20102000000 31 639 i 19809 21787 59 Lean Line Sirloin Strips 20102000000 43 639 27477 22487 60 Lean Line Sirloin Strips 20102000000 22 699 14058 30387 61 Lean Line Sirloin Strips 20102000000 23 699 16077 31087 62 Lean Line Sirloin Strips 20102000000 22 699 15378 31787 63 Lean Line Sirloin Strips 20102000000 17 699 11883 32487 64 Lean Line Sirloin Strips 20102000000 11 699 7689 33187 65 Lean Line Sirloin Strips 20102000000 16 699 11184 40787 66 Lean Line Sirloin Strips 20102000000 24 699 16776 41487 67 Lean Line Sirloin Strips 20102000000 35 689 24115 42187 68 Lean Line Sirloin Strips 20102000000 33 689 22737 42887 69 Lean Line Sirloin Strips 20102000000 31 689 21359 50587 70 Lean Line Sirloin Strips 20102000000 43 689 29627 51287 71 Lean Line Sirloin Strips 20102000000 29 689 19981 51987 72 Lean Line Sirloin Strips 20102000000 27 729 19683 52687 73 Lean Line Sirloin Strips 20102000000 17 749 12733 60287 74 Lean Line Sirloin Strips 20102000000 18 769 13842 60987 75 Lean Line Sirloin Strips 20102000000 14 799 11186 61687 76 Lean Line Sirloin Strips 20102000000 16 799 12784 62387 77 Lean Line Sirloin Strips 20102000000 16 709 11344 63087 78 Lean Line Sirloin Strips 20102000000 17 709 12053 70787 79 Lean Line Sirloin Strips 20102000000 27 709 19143 71487 80 Lean Line Sirloin Strips 20102000000 28 709 19852 72887 82 Lean Line Sirloin Strips 20102000000 24 689 16536 80487 83 Lean Line Sirloin Strips p 20102000000 17 689 11713 81187 84 Lean Line Sirloin Strips * ~>20102000000 16 789 12624 81887 85 Lean Line Sirloin Strips 20102000000 20 799 15980 82587 86 Lean Line Sirloin Strips 20102000000 12 799 9588 90187 87 Lean Line Sirloin Strips 20102000000 18 - 799 14382 90887 88 Lean Line Sirloin Strips 20102000000 24 799 19176 91587 89 Lean Line Sirloin Strips 20102000000 13 799 10387 92287 90 Lean Line Sirloin Strips 20102000000 21 759 15939‘ 92987 91 Lean Line Sirloin Strips 20102000000 10 759 7590 100687 92 Lean Line Sirloin Strips 20102000000 23 759 17457 101387 93 Lean Line Sirloin Strips 20102000000 20 759 15180 102087 94 Lean Line Sirloin Strips 20102000000 11 759 8349 102787 95 Lean Line Sirloin Strips 20102000000 17 759 12903 110387 96 Lean Line Sirloin Strips 20102000000 17 759 12903 111087 97 Lean Line Sirloin Strips 20102000000 13 729 9477 111787 98 Lean Line Sirloin Strips 20102000000 6 739 4434 112487 99 Lean Line Sirloin Strips 20102000000 9 739 6651 120187 100 Lean Line Sirloin Strips 20102000000 15 739 11085 A 120887 101 Lean Line Sirloin Strips 20102000000 19 739 14041 121587 102 Lean Line Sirloin Strips 20102000000 37 739 27343 122287 103 Lean Line Sirloin Strips 20102000000 - 6 739 4434 122987 104 Lean Line Sirloin Strips 20102000000 35 739 25865 10588 105 Lean Line Sirloin Strips 20102000000 25 739 18475 11288 106 Lean Line Sirloin Strips 20102000000 12 709 8508 11988 107 Lean Line Sirloin Strips 20102000000 14 709 9926 12688 108 Lean Line Sirloin Strips 20102000000 12 709 8508 20288 109 Lean Line Sirloin Strips 20102000000 19 709 13471 20988 110 Lean Line Sirloin Strips 20102000000 19 709 13471 21688 111 Lean Line Sirloin Strips 20102000000 24 709 17014 22388 112 Lean Line Sirloin Strips 20102000000 39 709 27651 30188 113 Lean Line Sirloin Strips v (continued) 13 Table 1 (continued). UPC Units Price Cost Date Week Description 20102000000 48 709 34032 30888 114 Lean Line Sirloin Strips 20102000000 41 709 29069 31488 115 Lean Line Sirloin Strips 20102000000 39 709 27651 32288 116 Lean Line Sirloin Strips 20102000000 18 709 13302 32988 117 Lean Line Sirloin Strips 20102000000 35 709 25865 40588 118 Lean Line Sirloin Strips 20102000000 41 759 31119 41288 119 Lean Line Sirloin Strips 20102000000 40 759 30360 41988 120 Lean Line Sirloin Strips 20102000000 18 759 13662 42688 121 Lean Line Sirloin Strips 20102000000 22 759 16698 50388 122 Lean Line Sirloin Strips 20102000000 26 759 19734 51088 123 Lean Line Sirloin Strips 20102000000 19 809 15371 51788 124 Lean Line Sirloin Strips 20102000000 12 809 9708 52488 125 Lean Line Sirloin Strips 20102000000 22 829 18238 53188 126 Lean Line Sirloin Strips 20102000000 15 829 12435 60788 127 Lean Line Sirloin Strips 20102000000 24 829 19896 61488 128 Lean Line Sirloin Strips 20102000000 27 829 22383 62188 129 Lean Line Sirloin Strips 20102000000 13 869 11297 83088 139 Lean Line Sirloin Strips 20102000000 11 829 9119 62888 130 Lean Line Sirloin Strips 20102000000 23 829 19067 70588 131 Lean Line Sirloin Strips 20102000000 18 829 14922 71288 132 Lean Line Sirloin Strips 20102000000 16 829 13264 71988 133 Lean Line Sirloin Strips 20102000000 10 829 8290 72688 134 Lean Line Sirloin Strips 20102000000 21 » 829 17409 80288 135 Lean Line Sirloin Strips 20102000000 21 829 17409 80988 136 Lean Line Sirloin Strips 20102000000 30 829 24870 81688 137 Lean Line Sirloin Strips 20102000000 7 829 5803 82388 138 Lean Line Sirloin Strips 20102000000 8 869 6952 90688 140 Lean Line Sirloin Strips 20102000000 10 869 8690 91388 141 Lean Line Sirloin Strips 20102000000 11 869 9559 92088 142 Lean Line Sirloin Strips 20102000000 18 869 15642 92788 143 Lean Line Sirloin Strips 20102000000 10 809 8090 100488 144 Lean Line Sirloin Strips 20102000000 16 809 12944 101188 145 Lean Line Sirloin Strips 20102000000 12 809 9708 101888 146 Lean Line Sirloin Strips 20102000000 21 809 16989 102588 147 Lean Line Sirloin Strips 20102000000 10 809 8090 110188 148 Lean Line Sirloin Strips 20102000000 25 809 20225 110888 149 Lean Line Sirloin Strips 20102000000 2 809 1618 111588 150 Lean Line Sirloin Strips 14 Table 2. Data for individual UPCs (example: Le Menu Beef Stroganoff). 15 )® UPC Units Price Cost Date Week Description 5100006313 486 357 173502 10786 1 Le Menu Beef Stroganoff 5100006313 482 357 172074 10786 2 Le Menu Beef Stroganoff 5100006313 515 357 a 183885 11486 3 Le Menu Beef Stroganoff 5100006313 443 357 158151 12186 4 Le Menu Beef Stroganoff 5100006313 356 357 127092 12886 5 Le Menu Beef Stroganoff 5100006313 418 357 149226 20486 6 Le Menu Beef Stroganoff 5100006313 437 357 156009 21186 7 Le Menu Beef Siroganoff 5100006313 348 357 124236 21886 8 Le Menu Beef Stroganoff 5100006313 514 357 183498 22586 9 Le Menu Beef Stroganoff 5100006313 351 357 125307 30486 10 Le Menu Beef Stroganoff 5100006313 395 357 141015 31186 11 Le Menu Beef Stroganoff 5100006313 310 357 110670 31886 12 Le Menu Beef Stroganoff 5100006313 316 357 112812 32586 13 Le Menu Beef Stroganoff 5100006313 379 357 135303 40186 14 Le Menu Beef Stroganoff 5100006313 339 357 121023 40886 15 Le Menu Beef Stroganoff 5100006313 357 357 127449 41586 16 Le Menu Beef Stroganoff 5100006313 379 357 135303 42286 17 Le Menu Beef Stroganoff 5100006313 346 357 123522 42986 18 Le Menu Beef Stroganoff 5100006313 371 357 132447 50686 19 Le Menu Beef Siroganoff 5100006313 344 357 122808 51386 20 Le Menu Beef Stroganoff 5100006313 298 357 106386 52086 21 Le Menu Beef Stroganoff 5100006313 357 357 127449 52786 22 Le Menu Beef Stroganoff 5100006313 396 357 141372 60386 23 Le Menu Beef Strcganoff 5100006313 266 379 100814 61086 24 Le Menu Beef Stroganoff 5100006313 306 379 115974 61786 25 Le Menu Beef Stroganoff 5100006313 274 379 103846 62486 26 Le Menu Beef Stroganoff Q 5100006219 261 s79 99919 7099s 27 Le Menu Beef Stroganoff ~. L 5100006313 311 379 117869 71586 28 Le Menu Beef Stroganoff 5100006313 294 379 111426 72286 29 Le Menu Beef Stroganoff 5100006313 273 379 103467 72986 30 Le Menu Beef Strogancff 5100006313 271 379 102709 80586 31 Le Menu Beef Stroganoff 5100006313 314 379 119006 81286 32 Le Menu Beef Stroganoff 5100006313 299 379 113321 81986 33 Le Menu Beef Stroganofl 5100006313 230 379 87170 82686 34 Le Menu Beef Stroganoff 5100006313 238 379 90202 90286 35 Le Menu Beef Stroganoff 5100006313 242 379 91718 90986 36 Le Menu Beef Stroganoff 5100006313 267 379 101193 81686 37 Le Menu Beef Stroganoh‘ 5100006313 303 379 114837 92386 38 Le Menu Beef Stroganofl 5100006313 298 379 112942 83086 39 Le Menu Beef Stroganoff 5100006313 351 379 133029 100786 40 Le Menu Beef Stroganoff 5100006313 286 379 108394 101486 41 Le Menu Beef Stroganoff 5100006313 281 379 106499 102186 42 Le Menu Beef Stroganoff 5100006313 361 352 127072 102886 43 Le Menu Beef Stroganoff 5100006313 383 352 134816 110486 44 Le Menu Beef Stroganoff 5100006313 308 352 108416 ' 111186 45 Le Menu Beef Strogancff 5100006313 332 379 125828 111886 46 Le Menu Beef Stroganoff 5100006313 247 379 93613 113086 47 Le Menu Beef Stroganoff 5100006313 183 379 69357 120286 48 Le Menu Beef Stroganoff 5100006313 265 379 100435 120986 49 Le Menu Beei Stroganoff 5100006313 258 379 97782 121686 50 Le Menu Beef Stroganoff 5100006313 196 379 74284 122386 51 Le Menu Beef Stroganoff 5100006313 154 379 58366 123086 52 Le Menu Beef Stroganoff 5100006313 196 379 74284 10687 53 Le Menu Beef Stroganoff 5100006313 278 379 105362 11387 54 Le Menu Beef Stroganoff 5100006313 290 379 109910 12087 55 Le Menu Beef Stroganoff 5100006313 259 379 98161 12787 56 Le Menu Beef Stroganoff (continued) Table 2 (continued). UPC Units Price Cost Date Week Description 5100006313 296 379 112184 20387 57 Le Menu Beef Stroganoff 5100006313 235 379 89065 21087 58 Le Menu Beef Stroganoff 5100006313 243 379 92097 27187 59 Le Menu Beef Stroganoff 5100006313 256 379 97024 22487 60 Le Menu Beef Stroganqff 5100006313 224 379 84896 30387 61 Le Menu Beef Stroganoff 5100006313 229 379 86791 31087 62 Le Menu Beef Stroganoff 5100006313 213 379 80727 31787 63 Le Menu Beef Stroganoff 5100006313 216 379 81864 32487 64 Le Menu Beef Stroganoff 5100006313 188 379 71252 33187 65 Le Menu Beef Stroganoff 5100006313 190 379 72010 40787 66 Le Menu Beef Stroganoff 5100006313 182 379 68978 41487 67 Le Menu Beef Stroganoff 5100006313 160 379 60640 42187 68 Le Menu Beef Stroganofl 5100006313 221 379 83759 42887 69 Le Menu Beef Stroganoff 5100006313 237 379 89823 50587 70 Le Menu Beef Stroganoff 5100006313 195 379 73905 51287 71 Le Menu Beef Stroganoff 5100006313 188 379 71252 51987 72 Le Menu Beef Stroganoff 5100006313 176 379 66704 52687 73 Le Menu Beef Stroganofl 5100006313 197 379 74663 60287 74 Le Menu Beef Stroganoff 5100006313 185 379 70115 60987 75 Le Menu Beef Stroganoff 5100006313 223 379 84517 61687 76 Le Menu Beef Stroganoff 5100006313 174 379 65946 62387 77 Le Menu Beef Stroganoff 5100006313 194 379 73526 63087 78 Le Menu Beef Stroganoff 5100006313 185 379 70115 70787 79 Le Menu Beef Stroganoff 5100006313 208 379 78832 71487_ 80 Le Menu Beef Stroganoff 5100006313 209 379 79211 72187 81 Le Menu Beef Stroganoff 5100006313 164 379 62156 72887 82 Le Menu Beef Stroganoff 5100006313 211 379 79969 80487 83 Le Menu Beef Stroganoff 5100006313 194 379 73526 81187 84 Le Menu Beef Stroganoff 5100006313 200 379 75800 81887 85 Le Menu Beef Stroganofi \¥ 5100006313 190 379 72010 82587 86 Le Menu Beef Stroganoff 5100006313 210 379 79590 90187 87 Le Menu Beef Stroganofl 5100006313 179 379 67841 90887 88 Le Menu Beef Stroganoff 5100006313 181 379 68599 91587 89 Le Menu Beef Stroganoff 5100006313 ' 186 379 70494 92287 90 Le Menu Beef Stroganoff 5100006313 173 379 65567 92987 91 Le Menu Beef Stroganoff 5100006313 166 379 62914 100687 92 Le Menu Beef Stroganoff 5100006313 214 361 77254 101387 93 Le Menu Beef Stroganoff 5100006313 209 361 75449 102087 94 Le Menu Beef Stroganoff 5100006313 599 299 179101 102787 95 Le Menu Beef Stroganoff 5100006313 280 379 106120 110387 96 Le Menu Beef Stroganoff 5100006313 245 379 92855 111087 97 Le Menu Beef Stroganoff 5100006313 290 379 109910 111787 98 Le Menu Beef Stroganoff 5100006313 289 379 109531 112487 99 Le Menu Beef Stroganofl 5100006313 176 379 66704 120187 100 Le Menu Beef Stroganofl 5100006313 294 379 111426 120887 101 Le Menu Beef Stroganoff 5100006313 238 379 90202 121587 102 Le Menu Beef Stroganoff 5100006313 269 379 101951 122287 103 Le Menu Beef Stroganofl 5100006313 103 379 39037 122987 104 Le Menu Beef Stroganoff 5100006313 245 379 92855 10588 105 Le Menu Beef Stroganoff 5100006313 237 379 89823 11288 106 Le Menu Beef Stroganoff 5100006313 262 379 99298 11988 107 Le Menu Beef Stroganoff 5100006313 236 379 89444 12688 108 Le Menu Beef Stroganoff 5100006313 244 379 92476 20288 109 Le Menu Beef Stroganoff 5100006313 227 379 86033 20988 110 Le Menu Beef Stroganoff 5100006313 174 379 65946 21688 1 1 1 Le Menu Beef Stroganoff 5100006313 198 379 75042 22388 112 1 Le Menu Beef Stroganofi 5100006313 186 379 70494 30188 113 Le Menu Beef Stroganoff (continued) 16 >Q\ Table 2 (continued). UPC Units Price Cost Date Week Description ‘\ 5100006313 212 379 80348 30888 114 Le Menu Beef Stroganoff 5100006313 172 379 65188 31588 115 Le Menu Beef Stroganoff 5100006313 193 379 - 73147 32288 116 Le Menu Beef Stroganoff 5100006313 195 379 73905 32988 117 Le Menu Beef Stroganoff 5100006313 131 379 49649 40588 118 Le Menu Beef Stroganoif 5100006313 158 379 59882 41288 119 Le Menu Beef Stroganoff 5100006313 153 379 57987 41988 120 Le Menu Beef Stroganoff 5100006313 138 379 52302 42688 121 Le Menu Beef Stroganoff 5100006313 162 379 61398 60388 122 Le Menu Beef Stroganoff 5100006313 137 379 51923 51088 123 Le Menu Beef Stroganoff 5100006313 155 379 58745 51788 124 Le Menu Beef Stroganoff 5100006313 130 379 49270 52488 125 Le Menu Beef Stroganoff 5100006313 151 379 57229 53188 126 Le Menu Beef Stroganotf 5100006313 153 379 57987 60788 127 Le Menu Beef Stroganoff 5100006313 176 379 66704 61488 128 Le Menu Beef Stroganoff 5100006313 181 379 68599 62188 129 Le Menu Beef Stroganoff 5100006313 193 357 68901 83088 139 Le Menu Beef Stroganoff 5100006313 162 379 61019 62888 130 Le Menu Beef Stroganofl 5100006313 159 379 60261 70588 131 Le Menu Beef Stroganofi 5100006313 174 379 65946 71288 132 Le Menu Beef Stroganoff 5100006313 161 379 61019 71988 133 Le Menu Beef Stroganoft 5100006313 203 379 76937 72688 134 Le Menu Beef Stroganoff 5100006313 192 379 72768 80288 135 Le Menu Beet Stroganoff 5100006313 172 379 65188 80988 136 Le Menu Beef Stroganoff 5100006313 168 379 63672 81688 137 Le Menu Beef Stroganoff 5100006313 244 357 87108 82388 138 Le Menu Beef Stroganoff 5100006313 176 357 62832 90688 140 Le Menu Beef Stroganofl 5100006313 186 379 70494 91388 141 Le Menu Beef Stroganoff 5100006313 117 379 44343 92088 142 Le Menu Beet Stroganoff 5100006313 153 379 57987 92788 143 Le Menu Beef Stroganoff 5100006313 182 379 68978 100488 144 Le Menu Beef Stroganoff 5100006313 164 379 62156 101188 145 Le Menu Beef Stroganoif 5100006313 154 379 58366 101888 146 Le Menu Beef Stroganofi 5100006313 158 379 59882 102588 147 Le Menu Beef Stroganoff 5100006313 174 379 65946 110188 148 Le Menu Beef Stroganoff 5100006313 176 379 66704 110888 149 Le Menu Beef Stroganofi 5100006313 168 379 63672 111588 150 Le Menu Beef Stroganoff Custqmer Cqu n13- ment space and customer counts are not automatically Figure 2 plots customer counts per week, which for the retail firm studied ranged from 505,164 to 861,844 over the time frame. The average customer count for" this firm per week was on the order oi 680,000. Advertisement Space The advertisement information gathered over the period relates only to fresh beef products, not con- venience beef products. Consequently, in the analysis of convenience beef products, no assessment of the impact of advertising on item movement per 1,000 customers can be made. importantly, information on customer counts and advertisement space must be augmented to the price and quantity information of the individual UPCs. That is, data pertaining to advertise- 17 part of the scanner data pertaining to the individual UPCs collected at the point of sale. Advertisement space (in terms of square cen- timeters) forthe respective beef products varied consid- erably from week to week (Figures 3-11). Descriptive statistics of the advertisement variables are exhibited in Table 3. O1 all the carcass sections (brisket, chuck, ground, loin, rib, and round), ground beef is the most frequently advertised product (46 out of 113 weeks), whereas beef rib is the least frequently advertised product (18 out of 113 weeks). On the basis of print space, ground beef receives the most attention (on average 62 square centimeters), whereas rib receives the least attention (on average 11 square centimeters). The advertisement frequency for nonlean beet products is three times that for lean beef products. As well, the FRESH (100) / \\ LEAN (30) NONLEAN (70) \ \ Brisket Chuck Ground Loin Rib Round AOB (2) (7) (7) (14) (10) (s) (21) Brisket Chuck Ground 1.6m Rib Round AOB (1) (2) (2) (9) (1) (s) (10) CONVENIENCE (47) Steak Entree Ground Rib Roast (28) (12) Beef (1) Beef (3) (3) Figure 1. Schematic diagram of the UPCs. Table 3. Advertisement space’ for beef products by carcass section. Variable Mean Std. Dev. Median Minimum Maximum Frequency N ADLEAN 26.6776 53.0243 0 0 252 31 1 13 ADNOLEAN 290.927 283.731 221.13 0 1343.72 96 1 13 ADBRISK 40.9891 98.7422 0 0 555.65 32 1 13 ADcHucK 54.1262 133.099 0 0 557 21 1 13 ADGBEEF 62.539 139.916 0 0 825.6 46 113 ADLOIN 49.2113 108.059 0 0 598 29 113 ADRIB 11.2954 34.831 0 0 256, 18 113 ADHOUND 52.1808 138.014 0 0 695.2 22 113 ADAOB 47.2527 67.4795 2.75 0 277.2 57 1 13 ADVAOM“ 829.34 387.306 769.93 197.2 2108.82 1 13 1 13 “in square centimeters. bFish, pork, poultry, lamb, and veal. 18 Thousands 900‘ 80% 700 ' i 500 * 1i: 1i“ 1 111111'1111'11111111111111111111111111111111|1I111111111111|11111|11:111111111111111111111111111"1111111111111|11111111111111111111111111111111111 r 1 1 1 1 I Week Figure 2. Customer counts. cm’ 300 250 200 1. 100 5O 111111111 Or 1 1 I I I lllll IIIII MAI ll I I I IAIIIII ll I I III I III/II IIIIIIIIIIIIII l IIIIIIIII I I I I I I I T I I Figure 3. Advertisement space for lean beet products. cm? i400 i200 " iOOO * 800 " 600 " IIII ll IIllIllI Illll Ill llIIIllllIll llll llIil III I lll lll l IIIII ll III I l lIIl I 1 I i I I I I I ii I I I WGGR Figure 4. Advertisement space for nonlean beef products. cm’ 600 500 " 4OO ' 300 " 200 ' iOO" Week 1I/1\11mm)11111111111111111A1111 111/1\41111M111 1A11111A1 1 111 1111111111111 1 1 1 1 1 1 1 1 1 1 1 1 Figure 5. Advertisement space for brisket. cm? 600 500 * 400 - 300 - 200 " iOO" IIIIII I III IIIIIIII O 1 WGGR I I I III Il IIIIII I I IIIII ll II I l Il I IIIIIIIIIIIIIIIIIIIIIIII 1 1 1 1 1 1 1 1 1 Figure 6. Advertisement space for chuck. 1000 800" 600*" 200 O IIIAII I'll" I I I WGGK Figure 7. Advertisement space for ground beef. cm? 700 GOO '- 500 ' 40C ' 200 “ Figure 8. Advertisement space for loin. iOO cm? 200 " 150 " 100 ' 5O l | l I I I ll ll I l l | l | I l I I l I | I I l l I I l I I I l I | || l l | l I I | | I I | “ ll I l I | I l I I l I | I | I l I l l I l l | l | | lll O | | I I I I I r I W68k Figure 9. Advertisement space for rib. print space for nonlean items is, on average, slightly more than 10 times that for lean items. Advertisement space for fish, pork, poultry, lamb, and veal items Deletion of Particular UPCs and Data Anomalies The number of "usable" UPCs for this analysis is 147. Initially, 298 UPCs were available. Because of insuffi- cient observations and/or questionable data entries, 151 UPCs were eliminated from consideration. In the vast majority of the UPCs that were eliminated from consideration, the number of observations available for analysis was less than 30. Because of lack of observa- 300 250 " 200 " 150 " IOO’ Statistical Procedures 2O 800 7OO r 600 ' 500 r 400 " 300 " 200 " 100% ll lill l lllllllllllll llll ll l llllll lll llll lllllllllllll I I llllll I I I I I I I W66k O lLll I llllllllll:l Figure 10. Advertisement space for round. cm? 5O O lll llllill llllll lllAlllgllll l Figure 1 1. Advertisement space for all other beef. r l lll ll l lll lllll l I I l l I lll I i I I I I I W66k average almost 830 square centimeters weekly, roughly 2.5 times that for fresh beef products. tions and consequently lack of degrees-of-freedom in the econometric analysis, these UPCs were not used in this study. Data for analyses of fresh beef products correspond only to weeks 38 to 150. Weeks 1 to 37 were eliminated because of questionable data entries. Specifically, the entries for weeks 1 to 37 were several times larger than those for weeks 38 to 150, and it was not possible to account for this anomaly. Consequently, the advertisement information, described in the pre- vious section, although available overthe entire period, deals only with weeks 38 to 150. This truncation of the vertisement information was necessary for com- 3 [tibility with the price and quantity data. Data for ‘nalyses of convenience beef products, however, cor- respond to weeks 1 to 150. No dataanomalies were observed for convenience items. Descriptive Statistics (Individual UPCs) Because of the confidentiality of the data, it is not possible to report observations for all beef products over the time frame in question. Descriptive statistics and graphical analysis are, however, used primarily to chart customer purchases of the various beef items overtime. Detailed descriptive statistics of prices and pur- chases per 1,000 customers for the 147 individual beet products are exhibited in Appendix B. Descriptive statis- tics correspond to the mean, median, standard devia- tion, minimum, and maximum. The mean and median relate to measures o1 central tendency, the standard deviation corresponds to a measure of dispersion, and the minimum and maximum define the range of the data. To illustrate, consider the UPC 2024500000 (Choice boneless brisket #062). The average price is $1.27 per pound (or 127 cents per pound), and the average pur- chase per 1,000 customers is roughly 17.6 pounds. As general rule, lean beef products are more expensive l i han nonlean beef products. In this study, both lean and nonlean products correspond to Choice grades. The lean line brand for this firm is a Choice grade beef from which fat has been trimmed. Lean line brands for other retail firms are generally no-roll, Good (Select) equivalent grades (e.g. "Giant Lean"). Good (Select) grades of meat products are typically priced below equivalent cuts. The top five lean line UPCs in terms of average purchases per 1,000 customers are (1) 201047 (gour- met ground round, 4.8 pounds), (2) 201023 (tailless T-bone steaks, 1.3 pounds), (3) 201029 (eye round roast, 0.9 pounds), (4) 201031 (sirloin tip fillets, 0.7 pounds), and (5) 201063 (beef cube steaks, 0.7 pounds). The top five nonlean line UPCs in terms of average purchases per 1 ,000 customers are (1) 202601 * (ground beet chuck #079, 68.6 pounds), (2) 202600 (ground beef #078, 50.2 pounds), (3) 202602 (ground beet #080, 31.0 pounds), (4) 202450 (Choice boneless brisket #062, 17.6 pounds), and (5) 202012 (chuck boneless pot roast, 16.0 pounds). Similarly, the top five convenience UPCs in terms of average purchases per 1,000 customers are (1) 208989 (Armour Chicken Fry Beef Patties, 6.37 units), (2) 5015551 (Armour Salisbury Steak, 1.21 units), (3) 7337006 (Budget Sirloin Beef, 1.12 units), (4) 7337004 (Budget Gourmet Oriental Beef, 1.00 units), and (5) 7336006 (Budget Gourmet ~ Pepper Steak with Rice, 0.92 units). 21 Graphs corresponding to movement (units) overtime for each of the 147 beef items are available from the authors upon request. The graphs serve to summarize the variability in item movement on a week-to-week basis. With few exceptions, movement of beef items vary tremendously per week. Additionally, descriptive statistics of budget shares for the 147 beef products are exhibited in Tables 4-6. Budget shares represent the proportion of beef sales attributable to individual products. The top 10 lean fresh beef products, on the basis o1 average budget shares are (1) 201047 (Lean Line Gourmet Ground, 27.5 per- cent), (2) 201023 (Lean Line Extra Lean Boneless Stew Meat, 9.1 percent), (3) 201029 (Lean Line Eye Round Roast, 8.1 percent), (4) 201031 (Lean Line Sirloin Tip Fillets, 5.9 percent), (5) 201059 (Lean Line Flank Steaks, 5.7 percent), (6) 201063 (Lean Line Beef Cube Steaks, 4.9 percent), (7) 201033 (Lean Line Sandwich Steaks, 4.5 percent), (8) 201045 (Lean Line Shish Kabob, 3.7 percent), (9) 201032 (Lean Line Ranch Broils, 3.7 percent), and (10) 201028 (Lean Line Eye Round Steaks, 3.5 percent). Collectively, these 10 products account for almost 77 percent of the sales of lean beef products. On the basis of average budget shares, the top 10 fresh nonlean beef products are (1) 202601 (lean ground beef chuck #079, 16.1 percent), (2) 202600 (fresh ground beef #078, 8.9 percent), (3) 202602 (extra lean ground beef #080, 8.5 percent), (4) 202103 (beef rib eye steak #037, 7.13 percent), (5) 202213 (top sirloin steak boneless #032, 5.5 percent), (6) 202210 (beef loin T-bone steak #029, 4.7 percent), (7) 202209 (boneless strip steak #028, 4.6 percent), (8) 202012 (beet chuck boneless pot roast #054, 3.7 percent), (9) 202308 (beef round steak boneless #007, 3.0 percent), and (10) 202603 (ground beef gourmet #081, 2.6 percent). Col- lectively, these 10 products account for approximately 65 percent of the sales of fresh nonlean beef products. Within the class of convenience beef products, on the basis o1 Table 6, roughly 58 percent of dollar sales is attributable to steak items; 19 percent is attributable to entree items; 16 percent, to ground beef items; 6 per- cent, to roast beef items; and lessthan 1 percent, to beef rib items. Individually, the top 10 convenience items on the basis of average budget shares are (1 ) 208989 (Armour Chick Fry Beef Patties, 14.3 percent), (2) 5106322 (Le Menu Sirloin Tips, 6.7 percent), (3) 5106328 (Le Menu Yankee Potroast, 4.1 percent), (4) 5106324 (Le Menu Chop Sirloin, 4.0 percent), (5) 1386630 (Stouffer Orien- tal Beef Lean Cuisine, 3.7 percent), (6) 7337006 (Budget Sirioin Beef, 3.4 percent), (7) 5106327 (Le Menu Pepper Steak, 3.2 percent), (8) 7337004 (Budget Gourmet Oriental Beef, 3.1 percent), and (9) 5015916 (Classic Lite Steak Diane Mignonette, 3.0 percent), and (10) 5106313 (Le Menu Beef Stroganoff, 2.8 percent). Collectively, these items account for slightly more than 48 percent 0f the sales of convenience beef products. Within the class of fresh beef products, by carcass section, on the basis of Tables 4 and 5, ground beef constitutes roughly 37 percent of dollar sales; loin products constitute 19 percent; rounds constitute almost 12 percent; ribs constitute nearly 10 percent; chuck products constitute 6 percent; and briskets constitute 4 percent. All other beef cuts constitute 11 percent of dollar sales. importantly, in this retail firm, roughly 6 Average dollar sales per week for convenience and fresh beef products are exhibited in Table 7. Con; venience beef products generated nearly $36,000 sales per week, whereas fresh beef products yiel " almost $600,000 in sales per week. Within the class of convenience products, steak items, ground beef items, and beef entrees were most important in terms of dollar sales. Within the class of fresh beef products, by car- cass section, ground beef and loin products were the top contributors to dollar sales. Finally, lean beef products constituted about $34,000 per week in sales, whereas nonlean beef products constituted almost percent of fresh dollar sales is attributable t0 lean beef $564900 per week i" Sams- items, whereas 94 percent is attributable t0 nonlean beef items. Table 4. Budget shares for fresh lean beef products. UPC Mean Std. Dev. Median Minimum Maximum 201020 .0048 .0021 .0043 .0004 .0099 201023 .0909 .0365 .0800 .0458 .2649 201027 .0171 .0039 .0174 .0078 .0287 201031 .0589 .0115 .0607 .0272 .0799 201036 .0317 .0055 .0322 .0193 .0451 201043 .0031 .0018 .0028 0 .0081 201047 .2757 .0386 .2769 .1809 .3832 201061 .0068 .0026 .0064 .0010 .0136 201021 .0132 .0066 .0119 .0009 .0349 201022 .0199 .0032 .0194 .0142 .0307 201024 .0107 .0054 .0097 .0039 .0339 201025 .0082 .0030 .0074 .0045 .0198 201028 .0351 .0083 .0338 .0214 .0637 201029 .0808 .0238 .0767 .0519 .2159 201032 .0371 .0050 .0378 .0241 .0480 201033 .0448 .0053 .0446 .0325 .0776 201039 .0177 .0073 .0152 .0088 .0460 201040 .021 1 .0135 .0169 .0065 .0803 201044 .01 12 .0065 .0099 .0006 .0353 _ 201045 .0373 .0072 .0373 .0194 .0578 201048 .0140 .0031 .0137 .0074 .0236 201059 .0573 .0144 .0557 .0383 .1769 201062 .0020 .0015 .0018 0 .0065 201063 .0491 .0145 .0530 .0071 .0713 201026 .0023 .0016 .0021 0 .0076 201030 .0129 .0081 .0112 .0042 .0630 201034 .0124 .0024 .0119 .0070 .0187 201042 .0024 .0016 .0022 .0002 .0119 201046 .0067 .0038 .0056 .0010 .0172 201060 .0148 .0062 .0138 .0030 .0373 22 Table 5. Budget shares for fresh nonlean beef products. UPC Mean Std. Dev. Median Minimum Maximum ’2?Li 202100 .0003 .0002 .0002 0 .0009 ~\ 202101 .0013 .0064 .0006 0 .0526 202103 .0727 .0159 .0697 .0499 .1506 202105 .0006 4 .0012 .0003 .0000 .0107 202106 .0005 .0003 .0005 .0000 .0014 202107 .0019 .0033 .0011 .0000 .0252 202109 .0027 .0047 .0013_ .0003 .0411 202016 .0060 .0019 .0056 .0029 .0127 202205 .0063 .0049 .0056 .0023 .0327 202017 .0020 .0005 .0019 .0011 .0032 202206 .0013 .0015 .0014 .0003 .0133 202019 .0262 .0153 .0217 .0079 .0355 202210 .0469 .0034 .0450 .0235 .0733 202213 .0547 .0301 .0465 .0232 .2447 202212 .0060 .0047 .0049 .0005 .0473 202211 .0097 .0020 .0095 .0043 .0150 202214 .0001 .0002 .0001 0 .0016 202215 .0061 .0013 .0061 .0032 .0119 202306 .0022 .0016 .0013 .0003 .0115 202303 .0293 .0259 .0232 .0120 .1475 202309 .0153 .0070 .0146 .0034 .0537 202311 .0013 .0016 .0009 .0001 .0131 202312 .0054 .0021 .0054 .0015 .0172 202313 .0105 .0033 .0094 .0059 .0363 202314 .0050 .0023 .0046 .0019 .0154 202315 .0005 .0002 .0005 .0001 .0010 202316 .0001 .0005 .0001 .0000 .0003 *I\:i 202317 .0006 .0002 .0005 .0002 .0014 4 202313 .0250 .0113 .0211 .0113 .0603 202319 .0139 .0096 .0099 .0036 .0447 202320 5 .0004 .0001 .0004 .0002 .0003 202321 .0033 .0033 .0022 .0003 .0233 202322 .0004 .0002 .0004 .0000 .0011 202323 .0005 .0003 .0005 .0000 .0013 202324 .0005 .0002 . .0004 .0001 .0011 202209 .0463 .0243 .0330 .0226 .1416 202400 .0039 .0021 .0034 .0013 .0112 202450 .0221 .0344 .0092 .0042 .1334 202451 .0222 .0112 .0135 .0100 .0719 202500 .0203 .0066 .0193 .0093 .0533 202501 .0033 .0019 .0040 .0002 .0090 202503 .0216 .0031 - .0216 .0149 .0333 202504 .0006 .0003 .0006 .0001 .0013 202505 .0122 .0175 .0066 .0030 .1130 202506 .0014 .0005 .0015 .0003 .0029 202507 .0097 .0014 .0097 .0064 .0137 202503 .0061 .0021 .0057 .0032 .0152 202550 .0037 .0003 .0036 .0023 .0070 202600 .0333 .0213 .0360 .0470 .1731 202601 .1609 .0165 .1611 .1223 .2136 202602 .0349 .0145 .0330 .0570 .1432 202603 .0264 .0032 .0259 .0201 .0343 202605 .0011 .0011 .0006 .0000 .0036 202607 .0124 .0013 .0125 .0091 .0154 3353531 .0003 .0006 .0000 0 .0027 202603 .0133 .0024 .0133 .0036 .0231 202609 .0032 .0030 .0013 .0007 .0115 Ag:~ 202203 .0010 .0030 .0003 0 .0246 (continued) 23 8858507 202325 8858508 201658 202006 202005 202009 202008 202014 202015 202007 202012 Table 5 (continued). .0000 .0050 .0006 .0088 .0063 .0009 .0003 Minimum UPC 1350011 2551923 2551927 2551951 1350032 5015400 5015409 5015410 5015412 5015551 5015910 1350059 5015915 5105313 2552032 5015413 51 05322 51 05324 51 05327 4452503 5015550 7335005 %7337004 1350510 7335003 1352010 1352011 1355520 1355530 205959 1355531 1551470 2550049 3557154 4452511 5105325 7112055 7112157 7335009 759010 Mean .0209 .0105 .0243 .0206 .0155 .0253 .0091 .0092 .0077 .0227 .0127 .0100 .0302 .0284 .0049 .0063 .0673 .0405 .0317 .0065 .0114 .0289 .0312 .0086 .0276 .0086 .0125 .0192 .0372 .1438 .0108 .0031 .0235 .0036 .0068 .0413 .0065 .0108 .0135 .0137 .0097 .0136 .0244 .0081 .0071 .0236 .0129 .0126 .0117 .0356 .0129 .0066 .0122 .0158 .0077 .0109 .0318 .0221 .0139 .0093 .0155 .0198 .0230 .0066 .0232 .0111 .0161 .0139 .0215 .2079 .0198 .0049 .0084 .0043 .0103 .0188 ence beef pfOdUClS. 24 .0221 .0004 .0238 .0226 .0160 .0296 .0012 .0111 .0106 .0292 .0236 OO .0601 .0323 .0293 .0002 .0232 .0256 .0085 .0367 .0202 .0372 .0492 O0 .0250 .0002 .0375 OOO 0049 L__'____ OOOOOOOOOOOO .0015 .0117 OO .0234 .0142 .0123 0O .0086 OOOOOOO .0102 0O .0023 OO .0149 OOOO .0117 .0002 .0207 .0002 .0076 .0020 .0196 .0152 9 .0551 .0721 .1260 .0387 .0289 .1179 .0627 .0805 .0607 .1978 .0877 .0254 .0755 .0913 .0336 .0678 .1905 .1386 .1165 .0424 .0559 .1529 .1561 .0503 .0987 .0546 .0708 .1038 .1563 .6735 .1307 .0263 .0518 .0164 .0461 .1373 .0267 .0556 .0722 .1081 (continued) m 3 Table 6 (continued). UPC Mean Std. Dev. Median Minimum Maximum 1380627 .0076 .0063 .0077 .0291 5015414 .0046 .0068 0 0 .0231 7336007 .0271 .0205 .0201 .0087 .1544 7338005 .0169 .0148 .0220 0 .0780 1382023 .0153 .0201 0 0 .0779 5015923 .0265 .01 90 .0294 0 .0760 7337006 .0344 .0237 .0294 0 .1528 Table 7. Average dollar sales per week tor convenience and fresh beef products. Convenience Beet Products Category Average Dollar Sales Per Week Convenience beef products $35,729 Steak products 19,351 Beef entrees 6,226 Ground beef products 7,863 Beef ribs 260 Roast beef 2,027 Fresh Beef Products , Category Average Dollar Sales Per Week Fresh beef products $597,897 By carcass section Brisket 26,638 Loin 113,531 Rib 59,946 Round 71,038 Ground 222,934 Chuck 39,920 AOB 66,887 Lean 34,206 Nonlean 563,691 Desgriptive $tati5fi¢$ almost 14 pounds per week. In comparison, the average (Commodity Graups) purchase per 1,000 customers tor nonlean products is Descriptive statistics of prices and purchases per 1,000 customers for aggregate beef commodity groups are exhibited in Table 8. The average price of lean beef items in the aggregate is $3.47 per pound; in com- parison, the price of nonlean beef, on the average, is $2.42 per pound, roughly 70 percent of the price of lean beef. Thus, the price premium for lean beef is on the order oi 40 percent in this retail tirm. Except for loin, the price of lean products exceeds the price of nonlean products. In particular, the price for lean brisket is about 1.4 times that oi nonlean brisket; for rib the price premium is 80 percent; for round, 30 percent; for ground, 50 percent; and for chuck, 20 percent. In the aggregate for this retail firm, the average purchase per 1,000 customers for lean products is 25 about 336 pounds. The average purchase of oon- venience products, per 1 ,000 customers, is roughly 23 units. The principal beef product in terms oi purchases per 1,000 customers is ground beef (nearly 170 pounds), and the least important product is rib (almost 20 pounds). Purchases per 1,000 customers for the remaining aggregate groups are on the order of 25 to 40 pounds. For convenience products, the key products in terms of product movement are steak, ground beef, and entrees. The least important convenience items in terms of product movement are roasts and ribs. Finally, Figures 12-33 are graphs corresponding to purchases over time for each of the beet commodity groups. With few exceptions, purchases of the ag- gregate beef products vary tremendously on a weekly basis. . Table 8. Descriptive statistics of prices and purchases per 1,000 customers for aggregate beef commodity’ * groups. 1 Variable Mean Median Std. Dev. Min Max N Prices L LEAN PFLEAN 347.04 349.03 10.33 231.74 332.02 1 13 PFLBRISK 249.33 249.00 10.31 229.00 239.00 113 PFLRIB 772.09 759.00 57.27 359.00 339.00 113 PFLLOIN 429.52 432.20 20.30 330.33 474.40 1 13 PFLAOB 373.33 337.13 23.73 235.53 405.73 113 PFLROUND 293.33 399.74 13.13 310.27 421.31 113 PFLGRND 277.32 230.02 1 1.93 230.49 300.42 1 13 PFLCHUCK 312.33 312.29 15.13 » 279.73 340.30 113 NONLEAN PFNLEAN 242.25 249.33 23.13 139.49 295.41 1 13 PFNLBRSK 173.34 130.34 27.52 99.42 211.10 113 PFNLRIB 413.37 417.55 37.01 240.45 504.40 113 PFNLLOIN 441.20 432.93 33.27 279.12 570.02 113 PFNLAOB 233.23 273.03 33.73 134.90 315.50 113 PFNLRND 303.37 _ 313.92 45.24 177.03 234.71 113 PRNLGRND 137.74 194.30 23.05 132.29 221.15 113 PFNLCHCK 232.99 277.30 47.59 125.43 325.53 113 CONVEN . PCON 245.03 249.47 23.33 133.45 233.33 150 PCSTEAK 259.53 233.23 17.93 175.20 291.20 _ 150 PCGBEEF 170.11 a 159.00 25.74 97.34 199.00 15o PCROAST 320.79 302.93 41.03 201.29 373.00 150 . PCENTREE 253.55 251.41 35.31 145.51 379.00 150 a PCRIB 373.19 339 13.02 299.00 339.00 32 Purchases per 1,000 Customers LEAN FLEAN 13.93 13.94 1.74 7.14 13.34 113 FLBRISK 0.34 0.29 0.13 0.10 0.90 113 FLCHUCK 0.33 0.32 0.12 0.03 0.35 113 FLGRND 5.03 5.17 0.73 2.33 3.93 113 FLLOIN 1.34 1.33 0.32 0.92 2.50 113 FLRIB .03 0.07 0.04 0.00 0.21 113 FLROUND 2.23 2.10 0.72 0.93 3.33 113 FLAOB 4.30 4.10 0.93 2.03 . 10.00 113 NONLEAN FNLEAN 333.34 313.07 72.29 132.31 A 523.43, 113 FNLBRSK 25.50 12.15 37.52 4.30 213.92 1 13 FNLCHUCK 25.37 13.93 24.39 3.14 123.37 113 FNLGRND 133.93 152.37 A 33.13 37.21 319.73 113 FNLLOIN 36.93 32.87 13.82 15.06 90.01 113 FNLRIB 19.48 18.63 6.49 9.52 57.67 1 13 FNLROUND 33.73 24.59 25.45 12.41 130.35 1 13 FNLAOB 31.30 25.95 15.72 15.90 113.37 113 CONVEN CSTEAK 1 1.20 10.37 3.04 4.1 1 20.70 150 CENTREE 3.31 2.77 2.47 0.20 a 12.34 150 CGBEEF 3.93 1.94 1 1.03 0.43 43.07 150 CROAST 0.97 0.34 0.47 0.39 3.47 150 i; CRIB 0.25 0.24 0.19 0.00 1.25 32 1 " 26 Pounds ‘do Pounds 6000 f ~\ 5000 ~ 10000 . 4000 “ (\/ 8000 3000 - 6000 * 2000 " 4000 ' 2000 - 1000 — 0 lllllIll:llllllll:lll|llll:llllllllgllllllllallllllllyllllllllllllllllillllllllillllllllvlllllli:llllllll%ill O “Hull:lllllllwllllllllp|ll|Hwllllllllglnl“ll?“H“|%|l|lllllgllilllllllliillllillllllllglllllillglll Week Week Figure 12. Purchases of fresh-lean beef. Figure 15. Purchases of lean ground beef. Pounds Pounds O 600 " 1500 " 400 ' 1000 §\ ' 200 - i] 50o P ‘ O ||||||||g||||||||Pull“|:||||l|“%l||||l||%ul|“Hglulnupu||‘||%||||||":1|||||||:|||||||l:||||||||?|| o llllllllglilllll%lllllIlllllllllll:lllll|l|:|llll|ll:llllllll:Illllllqllllllllrlllllllllllllllllllllllllllll Week Week Figure 13. Purchases of lean brisket. 759"" 15- Pufdlli" 97 "B" bi"- Pounds ‘ Pounds 160 140 400 * 120 30o _ 10o 8O 200 i‘ 6O 40 100 ‘ W 2o — ' W‘ O ||||||||:|1||11||:111nlullumniulumlmluuiunuuiuunuiil|mn:1unluluuulnlaxllii:ilnu 0 lllll|ll:llllllll:llllllll:llllllll:llllllll:llIllllIlllllllll:lll1llll:l|llllll:llllllllpllllllwjillllllllll h Week Week '5' Figure 14. Purchases of lean chuck. Figure 17- Pufchlsfl 0i i980 rib- 27 Pounds Pounds 200000 T 4000 ~ 150000 - 0000 — 100000 - 2000 - 50000 1000 a O _L l l _L l l l 1 l l l l O llllllll: Illllll0“llllll:llllllll}lllllIll:Illlllll:llllllllgllllllll:lllllIll:Illlllllgllllllllullllllllllll Week f Week Flgure 18. Purchases of lean round. Figure 21. Purchases of nonlean brisket. Pounds Pounds 120000 0000 _ 100000 - 5000 30000 - 4000 00000 - :55 XMLJIMJIWMI A l1ll|lllll|Illllllllllllllllllllllllllllllllllllllllllllllillllll|l|l|||llll||l||||||l||l||l||l||llllll|l I I I I l I I Illlllllllllll lllllllllIlllllllllllllllllllllll“Hlllllll|l|l||||llllllllllllllllllllll|lllllllllllllllllllll o I I ' I 0 I ' I I I I I l I I I I Week Week Flgure 19. Purchases of lean all other beef. Flgure 22. Purchases of nonlean chuck. Pounds 300000 Pounds 400000 250000" 300000 ~ 200000 _ 150000" 200000 100000 MA 100000- 50000" 0 l"l"“:'“"lll:"ll""rIlll"will"lllrlllllllrllllIll:llllllllrllln||:lInuurnunlrnllnnnln O lllH:ll|l|ll|:llll|ll|0ll|lllll:Illllll[PlllllllélllIllll:llllllll:llllllll:llllllllgllllllll:llllllllllll Week » Week Flgure 20. Purchases of fresh nonlean beef. Flgure 23. Purchases of nonlean ground. * V. 28 60000 I») 40000 30000 20000 i0000* Pounds VV88k Figure 24. Purchases of nonlean loin. Pounds 40000 30000- 20000- *\ 10000- 0 illllllllllllllllllllllllllllllilllllllllllliIlllllllllllllllllllllIlllllllllllllllllilllllllllllllllllllllllll I I I I I I I T I I I I 120000 100000 80000 60000 40000 20000 ‘a VV88k Figure 25. Purchases of nonlean rib. Pounds 1- 11111111111701.1110 llllllIIIIllIll!llllllllllllllllllllllllllllllllllllIlllIlllllllIlllllIlllIllllllllllIIllllllllllllllllllllllll I I I I I I I I I I I I Figure 26. Purchases of nonlean round. 100000 80000 60000 40000 20000 Pounds IlllllIlllllIllllIlllllllllllllllllllllllillllllllllllllillIllllIllllllllllllIlllllllllllllllllfilllllllllll I I I I I I I I I I I I Figure 27. Purchases of nonlean all other beef. POUDGS 60000 40000 30000 20000 10000 VVGGK Figure 28. Purchases of convenience products. POUHGS 20000 15000 I 10000 5000 VV6€k Figure 29. Purchases of convenience steak products. 29 pOUfidS 6000* QIZWJWWJL, |||||||||||||||||||||||||||||||||||| n Figure 30. Purchases of convenience beet entrees. Pounds 2500 1 2000* 1500* 1000* 500 Figure 31. Purchases of convenience roast beef products. Econometric Analysis The purpose of econometric analysis in this study is to develop models to explain variation in product move- ment. The functional form chosen for the demand relationships is open to empiricism. The study rests on the use of the linear functional form. The interpretation of parameter estimates as elasticities is convenient with the double logarithmic functional form. Because of potential zero observations, especially for the advertise- ment variables, this form, however, was not employed. Emphasis in the empirical results is on price and adver- tisement elasticities. Price elasticities refer to percent- age changes in purchases caused by unit percentage changes in prices; similarly, advertising elasticities refer to percentage changes in purchases caused by unit percentage changes in advertising. Elasticities are often of primary interest not only to agricultural economists but also to food retailers. Price elasticities allow retailers to deal with shortage or surplus situations to minimize price volatility. Advertising elasticities reveal the sen- sitivity of purchases to advertisement efforts. 30 Pounds 30000 26000 20000 15000 10000 5000 lllll n Figure 32. Purchases of convenience ground beef products. Pounds O 800* 600* 400* 200 WGGK Figure 33. Purchases of convenience beef ribs. Under the assumption that supply is perfectly elastic in this local market, a seemingly unrelated regression (SUR) procedure is workable. Random exogenous variates such as general level of economic activity, competitors’ actions, prices of nonmeat items within the retail firm, or the lack of certain influences may affect purchases of the beef products apart from the specified predetermined variables. Consequently, the distur- bance terms of the equations may be contem- poraneously correlated. Given that the exogenous variables are not the same in each relationship, gains in estimation efficiency can be expected with the SUR procedure relative to the use of ordinary least squares (Fomby et al., 1984). In our study, the empirical results of the representative individual fresh and convenience beef products rest on the use of the SUR procedure. However, ordinary least squares is used to estimate the broad fresh and convenience beef groups because the set of regressors are the same for these equations within a particular category. VP‘ lllllllllllllllllllllllllllllllllllllllillllllllllll A] O I T | T I I I i i I I Empirical Results \ This section concerns the econometric demand available from the authors upon request. Space limita- analyses for the top 10 beef products in the lean, tions prohibit reporting econometric results ior all beet nonlean, and convenience categories (according to products. The econometric models correspond to budget shares) as well as the aggregate beef groups. demand relationships at the retail level of the marketing Table 9 lists the respective products and groups. chain forthisiirm. The dependent variable inthe respec- Analyses iorthe remaining individual beef products are tive demand relationships is units of movement per Table 9. Top 10 beet products by category and list of aggregate beet groups. UPC Description Budget Share Top 10 lean beef products (according to budget share) 201 O47 Lean Line Gourmet Round 0.2757 201023 Lean Line Extra Lean Boneless Stew Meat 0.0909 201029 Lean Line Eye Round Roast 0.0808 201031 Lean Line Sirloin Tip Fillets 0.0589 201059 Lean Line Flank Steaks 0.0573 201063 Lean Line Beef Cube Steaks 0.0491 201033 Lean Line Sandwich Steaks 0.0448 201045 Lean Line Shish Kabob 0.0373 201032 Lean Line Ranch Broils 0.0371 201028 Lean Line Eye Round Steaks 0.0351 Top 10 nonlean beef products (according to budget shares) 202601 Lean Ground Beef Chuck #079 0.1609 a 202600 Fresh Ground Beef #078 0.0888 (L 202602 Extra Lean Ground Beef #080 0.0849 ‘ 202103 Beef Rib Eye Steak #037 0.0727 202213 Top Sirloin Steak Boneless #032 0.0547 202210 Beef Loin T-Bone Steak #029 0.0469 202209 Boneless Strip Steak #028 0.0463 202012 Beef Chuck Boneless Pot Roast #054 0.0373 202308 Beef Round Steak Boneless #007 0.0298 202603 Ground Beef Gourmet #081 0.0264 Top 10 convenience beef products (according to budget share) 208989 Armour Chick Fry Beef Patties 0.1438 5106322 Le Menu Sirloin Tips 0.0673 5106328 Le Menu Yankee Potroast 0.0413 5106324 Le Menu Chop Sirloin 0.0405 1386630 Stouffer Oriental Beef Lean Cuisine 0.0372 7337006 Budget Sirloin Beef 0.0344 5106327 Le Menu Pepper Steak 0.0317 7337004 Budget Gourmet Oriental Beef 0.0312 5015916 Classic Lt Steak Diane Mignonette 0.0302 5106313 Le Menu Beef Stroganoif 0.0284 Aggregate Beet Groups Lean beef products - Nonlean beef products Convenience beef products Lean Brisket Nonlean Brisket Convenience Steak Lean Chuck Nonlean Chuck Convenience Beef Entrees Lean Ground Nonlean Ground Convenience Beef Ribs Lean Loin Nonlean Loin Convenience Ground Beef Lean Rib Nonlean Rib Convenience Roast Beef n Lean Round Nonlean Round .____‘ Lean All Other Beef Nonlean All Other Beef 31 1,000 customers. The respective exogenous (inde- pendent) variables are (1) own-price, (2) prices of com- peting products, (3) advertisement variables, (4) seasonality (monthly dummy variables), and (5) a dummy variable for holidays. Simply put, the purpose of the econometric analysis is to identify and assess fac- tors affecting purchases per 1,000 customers. Em- phasis is on price elasticities and on advertisement elasticities. Forthe respective econometricanalyses, it is neces- sary to operationalize the generic specifications given by equation 2 (see section titled "Conceptual Framework for the Analysis"). For example, in the econometric model for lean beef products, individual UPC 201047 (Lean Line Gourmet Ground Round), cross-price variables correspond to competing lean cuts, namely, chuck, brisket, loin, rib, round, and all other beef. Prices of nonlean beef products, con- venience beef products, and meat products other than beef (pork, poultry, and fish) are also incorporated in the model. Similarly, in the econometric model for nonlean beet products, for example individual UPC 202210 (beef loin T-bone steak #029), cross-price variables cor- respond to competing nonlean cuts, namely, brisket, chuck, ground, rib, round, and all other beef. As well, prices of lean beef products, convenience beef products, and meat products otherthan beef are regres- sors in the model. Advertising variables corresponding to carcass section cuts are also incorporated in the models tor lean beef products and nonlean beef products. Finally, in the econometric model for con- venience beef products, for example individual UPC 51 06327 (Le Menu Pepper Steak), cross-price variables correspond to competing products, namely, co venience entrees, beef ribs, ground beef, and roa beef. Prices of lean and nonlean beef products as well as meat products other than beef are also included. However, no advertisement variables are included in the models for convenience beef products. Convenience beef products were not advertised over the period in question. Fresh Beef and Convenience Beef Products (Individual UPCs) Results of the econometric analyses for individual beef products are documented in Appendix C, and a summary of the econometric analyses is given in Tables 10-12. These tables correspond to analyses performed for lean beef products (Table 10), nonlean beef products (Table 11), and convenience beef products (Table 12). Goodness-of-Fit and Serial Correlation The models adequately capture significant amounts ol variation in purchases per 1,000 customers. The system R2 measure is the statistic used to represent the amount of variation explained by the model because a seemingly unrelated regression (SUR) procedure is used. The closerto 1 , the better the lit of the model. For the representative lean beef products, the system R2 is 0.8418; for nonlean beef products, the system R2 is 0.8987; finally, for convenience products, the system R2 is 0.8236. For the relatively large amount ol variation to Table 10. Summary of econometric analysis: individual lean beet products. Own-Price“ Own-Advert.‘ UPC Elasticity Elasticity DW Test” Seasonality° Holiday“ 201 O47 -1.067‘ NS 2.449 4.945‘ -3.850‘ 201023 -2.485‘ NS 2.093 10.717‘ -3.826‘ 201029 -1 0.531 ‘ 0.043‘ 2.126 2.297‘ -3.094‘ 201031 NS -0.034‘ 2.292 1.578 -1.713‘ 201059 -1 .242‘ -0.029‘ 2.288 1.620‘ -3.429‘ 201063 -1.598‘ NS 1.967 1.724‘ -3.895‘ 201033 -4.746‘ NS 2.044 4.543‘ -4.996‘ 201045 -3.974‘ -0.022‘ 2.268 3.350‘ -0.207 201032 NS -0.024‘ 2.193 1.620‘ -2.728‘ 201028 NS 0.062‘ 2.278 1.825‘ -2.671‘ System R2 = 0.8418 a At the sample means. Durbin-Watson test statistic for serial correlation. g F-statistic. , t-statistic. from zero). Statistically significant at the 0.05 level (for the elasticity measures, denotes regression coefficient statistically different NS denotes regression coefficient not statistically different from zero. S \ » i 32 Table 1 1. Summary of econometric analysis: individual nonlean beef products. i Own-Price‘ Own-Advert.‘ b »' UPC Elasticity Elasticity DW Test seasonality” Holiday“ 202601 -1 .589‘ 0.015‘ 2.452 0.975 -2.026" 202600 -1.599* 0.087‘ 2.084 1.138 1.658 202802 -1 .963‘ 0.015‘ 1.995 1.641 * -2.0o5' 202103 -3.905* 0.033‘ 1.997 2.981 ' 2.128‘ 202213 -3.286* 0.152‘ 2.052 1.898‘ -0.124 202210 -2‘600 0.028‘ 1.973 2.084‘ -0.1 13 202209 -7.51 1 ‘ NS 1.984 4.218‘ 0.205 202012 -3.506‘ 0.016‘ 2.164 1.945‘ -0.579 202308 -5.658* 0.244‘ 2.382 1.380 -o.o3o 202603 -1 .681 ' NS 2.570 1.859‘ 8.928‘ System R2 - 0.8987 a At the sample means. Durbin-Watson test statistic for serial correlation. (c! F-statistic. . t-statistic. from zero). Statistically significant at the 0.05 level (for the elasticity measures, denotes regression coefficient statistically different NS denotes regression coefficient not statistically different from zero. Table 12. Summary ot econometric analysis: individual convenience beef products. Own-Price‘ 4h UPC Elasticity ow Testb Seasonality° Holiday“ *~~ 208989 -2.513' 1.713 0.850 -2.416‘ 5106322 -7.697* 2.567 2.993‘ -4.540‘ 5106328 -8.227" 2.604 5.148’ -4.270' 5106324 -1 5.389‘ 2.287 1.842‘ -1.535 1386630 -4.232* 1 .786 4.193* -3.634' 7337006 -1 1.312‘ 2.555 3.046‘ 1.490 5106327 -9.232' 2.585 3.504’ -3.702' 7337004 -1 1.767‘ 2.582 3.016‘ -1.566 5015916 ‘ -2.612’ 2.155 2.913‘ -2.378" 5106313 -9.834* 2.125 5.219‘ -3.625" System R2 = 0.8236 a At the sample means. b Durbin-Watson test statistic for serial correlation. g F-statistic. _ t-statistic. from zero). Statistically significant at the 0.05 level (for the elasticity measures, denotes regression coefficient statistically different NS denotes regression coefficient not statistically different from zero. be explained on a week-to-week basis, the goodness- of-fit is generally very satisfactory. Additional evidence of reasonable results comes fromthe Durbin-Watson (DW) test statistic. This statistic provides evidence of the existence (or nonexistence) of serial correlation, a phenomenon often observed with time-series data in the evaluation of econometric 1.‘ .28 33 models. All DW test statistics indicate the absence of serial correlation at the 0.05 level of significance. Own-Price Elasticities All own-price elasticities are negative, corresponding to an inverse relationship between purchases (move- ment) and price. Further, except for three lean beef products, all are statistically different from zero. Moreover, the respective own-price elasticities are in Cross-Price Elasticities the elastic range. For the lea'n beef items, the mag- nitudes range from -1.067 to -10.513; for nonlean beef products, -1.589 to -7.511; and for convenience beef items, -2.513 to -15.389. This finding agrees with pre- vious studies (Funk et al., 1977, pp. 536-537; Marion and Walker, 1978, p. 672). ln general, the own-price elasticities for convenience beef products are larger than the own-price elasticities for lean and nonlean beef products. Statistically significant cross-price elasticities cor- ‘- responding to individual UPCs are exhibited in Tables 13-15. Cross-price elasticities may be either positive, indicating gross substitutability, or negative, indicating gross complementarity. For lean beef products, 21 of the 60 cross-cut price elasticities are significantly dif- ferent from zero; of these, 15 are positive and 6 are negative. Additionally, only 3 of the 50 cross-product price elasticities are significantly different from zero; of these, 1 is positive and 2 are negative. The signs of Table 13. Statistically significant“ cross-price elasticities” for individual lean beef products. Cross-Cut Price Elasticity° Cross-Product Price Elasticity‘ UPC BrisketlChuckl Grnd l Loin l Rib lRoundl AOB NLeanl Conv l Pork l Poult l Fish 201047 -1.594 NA 0.755 1.895 201023 3.979 1.346 0.020 NA 201029 -3.719 2.359 3.454 4.545 2.101 NA -1.031 201031 -2.131 NA -0.288 201059 -0.843 -2.858 NA 201063 1.134 NA 201033 -1.618 0.723 2.360 NA 201045 1.705 _NA 201032 NA 201028 1.203 2.612 NA 0.520 a At the 0.05 level oi significance. At the sample means. ° Cross-cut price elasticity indicates the cross-price elasticity of a particular UPC with respect to a particular lean beef cut (i.e., brisket, chuck, ground loin, rib, round, or all other beef). Cross-product price elasticity indicates the cross-price elasticity of a particular UPC with respect to nonlean beef, convenience beef, pork, poultry, or fish. NA Not applicable. Table 14. Statistically significant’ cross-price elasticities“ for individual nonlean beef products. Cross-Cut Price Elasticity° Cross-Product Price Elasticity“ UPC BrisketlChuckl Grnd l Loin l Rib lRoundl AOB Lean l Conv l Pork l Poult l Fish 202601 NA -0.807 202600 NA -1.493 0.973 202602 NA 0.328 1 .881 ~0.378 202103 0.364 NA 0.235 202213 NA 1.235 202210 0.576 NA 202209 NA ' 1.168 1.463 5.761 4.484 0.826 -0.996 202012 1.023 NA -1.957 1.508 -2.101 7.268 202308 1.762 NA 16.045 -2.262 202603 NA -0.668 a At the 0.05 level of significance. b At the sample means. ° Cross-cut price elasticity indicates the cross-price elasticity of a particular UPC with respect to a particular nonlean beef cut (i.e., brisket, chuck, ground, loin, rib, round, or all other beef). d Cross-product price elasticity indicates the cross-price elasticity of a particular UPC with respect to lean beef, convenience beef, pork, poultry, or fish. NA Not applicable. 34 d1 able 15. Statistically significant’ cross-price elasticities” for individual convenience products. Cross-Cut Price Elasticity” Cross-Product Price Elasticity“ \ UPC Steak ] Ribs |Ground| Roast [Entree NLean i Lean I Pork l Poult ] Fish 206939 -7.212 NA‘ 1.44s 5106322 NA -1.204 1.346 5106323 -1.137 NA 0.386 2.091 5106324 NA -5.171 1336630 -2.250 NA 0.91 a 7337006 NA 1 .372 5106327 NA 1.906 7337004 NA 5015916 NA 5106313 4.49s NA a At the 0.05 level of significance. b At the sample means. ° Cross-cut price elasticity indicates the cross-price elasticity of a particular UPC with respect to a particular convenience beef item (i.e., steak, ribs, ground, roast, or entree). Cross-product price elasticity indicates the cross-price elasticity of a particular UPC with respect to nonlean beef, lean beef, pork, poultry, or fish. NA Not applicable. statistically significant cross-cut elasticities are predominantly positive. The crosscut price elasticities range from -3.719 to 4.545. Generally, cross-product "Rprices exert no discernible influence on purchases of 1.. lean beef. For individual nonlean beef products, only 15 of the 60 cross-cut price elasticities are significantly different from zero. Of the 10 nonlean beef items, all are sensitive t0 at least one cross-cut price. Typically, prices of bris- ket, rib, round, and all other beef affect purchases of the individual nonlean beef purchases. However, prices of chuck, ground, and loin generally do not affect pur- chases of the individual nonlean beef products. Further, 10 of the 50 cross-product price elasticities are sig- nificantly different from zero; of these, 6 are positive and 4 are negative. The price of lean beef positively influen- ces purchases of UPCs 202602, 202209, 202012, and 202308. The price of convenience beef, on the other hand, negatively influences purchases of UPCs 202209 and 202308. The price of poultry positively affects pur- chases of UPCs 202103 and 202209. The price of fish, however, negatively affects purchases of UPCs 202602 and 202209. For individual convenience beef products, only 7 of the 40 cross-cut price elasticities are significantly dit- ierent from zero. The cross-cut price elasticities range from -7.212 to 0.386. Particularly, the price of beef ribs negatively influences 6 of the 10 individual products. Additionally, only 6 of the 50 cross-product price elas- ticities are significantly different from zero. Generally, fish and lean beef are gross substitutes for the con- venience beef items. The prices of pork, poultry, and 35 nonlean beef typically are not statistically significant influences on purchases of convenience beef. Own-Advertisement Elasticities In this study, 2 of the 10 lean products and 8 of the 10 nonlean products have positive and statistically sig- nificant own-advertisement elasticities (Tables 10, 11, and 16). Contrary to expectations, however, four lean products have negative and statistically significant own- advertisement elasticities. For individual lean products, the magnitudes of the own-advertisement elasticities range from -0.034 to 0.062, whereas for individual non- lean products, these elasticities range from 0.015 to 0.224. The magnitude of the own-advertisement elas- ticities is much smaller than the magnitude of the own- price or cross-price elasticities. Cross-Advertisement Elasticities Statistically significant cross-advertisement elas- ticities for lean and nonlean beef products are exhibited in Table 16. Cross-advertisement effects for most of the individual fresh beef products are marginal. Only 14 cross-cut advertisement elasticities are significantly dif- ferent from zero for lean beef, whereas only 5 are significantly different from zero for nonlean beef. Cross- cut advertisements are a statistically significant in- fluence primarily in the purchases of UPC 201031 and UPC 201059. The magnitude of the cross-cut adverlise- ' ment elasticities are smaller than the magnitude of the own-advertisement elasticities. Additionally, cross- product advertisement elasticities are negative and statistically significant for only two products, UPCs Table 16. Statistically significant‘ advertisement elasticities” for individual fresh beef products. Cross-Cut Advertisement Elasticity’ Cross-Product Advertisement Elasticity UPC Brisket Chuck Ground LiOl'l Rib Round Other Meat AOB Products° Lean Beef 201 O47 201023 201029 201031 201059 201063 201033 201 045 201032 201028 Nonlean Beef 202601 202600 202602 202103 202213 202210 202209 202012 202308 202603 0.016 0.018 0.016 0.013 -0.020 0.018 0.017 0.01s“ 0.087“ 0.15“ 0.021 0.032 0.10s“ 0.013 0.015 0.152“ 0.02s“ -0.041 -0.010 0.043“ -0.034 -0.009 -0.014 0.002“ -0.010 0.033“ 0.036 0.224“ 0.029 0.029“ -0.139 -0.341 a At the 0.05 level of significance. b At sample means. ° Fish, lamb, pork, poultry, and veal. Denotes own-advertisement elasticities. particular beef cut. meat products (nonbeei). ° Cross-cut advertisement elasticity indicates the cross-advertisement elasticity of a particular UPC with respect to a ‘Cross-product advertisement elasticity indicates the cross-advertisement elasticity of a particular UPC with respect to other 202103 and 202213. Where statistical significances occur, the magnitude of the cross-product advertise- ment elasticities generally exceeds the magnitude of the own-advertisement or cross-cut advertisement elas- ticities. Holidays and Seasonality Influences of the holiday variable and seasonality are shown in Tables 10-12. For individual lean and con- venience beef products, the holiday variable is negative. Typically, the holiday variable is also statistically dif- ferent from zero. Thus, fewer purchases of lean beef and convenience beef products occur during holidays relative to nonholidays. For nonlean beef products, the holiday variable is not statistically different from zero for 6 of the 1O individual products. Except for UPCs 202601, 202602, 202103, and 202603, purchases of 36 nonlean beef during holidays are not statistically dif- ferent from those purchases during nonholidays. Seasonality, on the other hand, generally significant- ly influences purchases of beef products. Except for UPCS 201031, 202600, 202601, 202308, and 208989, all the individual lean, nonlean, and convenience beef products are subject to seasonal influences. Fresh Beef and Convenience Beet Products (Aggregate Groups) Appendix C documents the results of the econo- metric analyses for broad groups of beef products. A summary of the econometric analyses is given in Table 17. The focus in this section is noton individual UPCs but on groups of UPCs. The groups are (1) convenience rib, (2) convenience steak, (3) convenience entrees, (4) Table 17. Summary of econometric analysis: aggregate beef groups. Own-Price’ Own-Advert.‘ Elasticity Elasticity ADJRSO" ow Text‘ Seasonalityd Holiday’ Lean Brisket NS o.o7a* 0501* 2.150 2.242‘ 2.053‘ Chuck NS NS 0.571 ‘ 2.295 0.838 -1.633 Ground -1.185‘ NS 0.551 ‘ 2.423 4.713‘ -3.795‘ Lion -1.300‘ NS 0.486‘ 2.381 1.551 -3.216‘ Rib NS 0.049‘ 0.425‘ 1.833 2.411‘ -1.246 Round -5.694‘ 0.040‘ 0.697‘ 2.349 2.190‘ -2.882‘ AOB -2.666‘ NS 0.644‘ 2.164 4.037‘ -4.454‘ Nonlean Brisket -5.732‘ 0.172’ 0.776‘ 2.1 17 1.240 1.033 Chuck -2.902‘ 0.097‘ 0.904’ 2.143 1.828‘ -0.287 Ground -1.209‘ 0.040‘ 0.753’ 2.338 0.827 -0.036 Lion -1.897‘ 0.060‘ 0.820‘ 2.372 1.400 -1.507 Rib -2.146’ 0.059‘ 0.609‘ 2.473 0.314 0.337 Round -3.756‘ 0.109‘ 0.876‘ 2.513 0.972 -0.189 AOB -2.895‘ 0.053‘ 0.814‘ 2.102 2.138‘ -0.958 Convenience Steak -2.088‘ NA 0.763‘ 2.257 2.604‘ -4.956‘ Entree -3.127" NA 0.600‘ 2.075 1.514 -1.696‘ Ground -3.022‘ NA 0.746‘ 2.539 2.595‘ -4.613‘ Roast -4.692‘ NA 0.780‘ 1.730 3.593‘ -1.754‘ Rib -1 9.925‘ NA 0.828‘ 2.039 5.758‘ -1.250 a At the sam le means. Adjusted ° Durbin-Watson test for serial correlation. d F-statistic. f t-statistic. Statistically significant at the 0.05 level. e NS denotes regression coefficient not statistically different from zero. NA Not applicable. convenience ground beef, (5) convenience roast beef, (chuck); finally, the Rfi statistic for convenience beef (6) brisket (lean, nonlean),(7) chuck(lean, nonlean),(8) products ranges from 0.600 (entrees) to 0.828 (rib). ground (lean, nonlean), (9) loin (lean, nonlean), (10) rib Consequently, the econometric analyses are highly (lean, nonlean), (11) round (lean, nonlean), and (12) all satisfactory on the basis of goodness-of-fit. Additional other beef (lean, nonlean). Ordinary least squares is evidence of reasonable results comes fromthe DWtest used inthe estimation process forthe aggregate groups statistic. All DW test statistics indicate the absence of because the set of exogenous variables, or regressors, serial correlation at the 0.05 level of significance. are the same for all the equations in each of the three categories (lean, nonlean, convenience). Thus, no own-Prim? Elasiilimes 96508 i" 9ffi¢i§Fl¢Y 0i "19 Pafamelef 985018198 are Ff-fal- Except for lean brisket, lean chuck, and lean rib, the lled "Om "SW9 The Seemlngl)’ Unrelated FGQTQSSIO" own-price elasticities for the broad groups are negative (SUR) PFOCQdUFQ- and statistically different from zero. For lean beef, the rrrrrrrrrrrrr-rrr-rrr r-rrrr serrr-r crrrrelrrrrrr ?5"."2§3"<$Z...°;'.?;°;‘#3l'§Z.l22fii.;;3?.;Z.‘.§Zé%'?;2.‘f}£l? In all instances, the models for the respective ag- to -5.732 (brisket); and for convenience products, from gregate groups capture significant amounts oi variation -2.088 (steak) to -19.925 (rib). i" purchases perteee euetemereThe adjusted Rzlfiz) Similar to the findings for individual products, the measure isthe statistic used to represent the amount of response to price Changes is elastic. Lean round and variation explained by the model. The closer to 1, the | H m b f rt- I I T t h better the fit of the model. The ‘ti? statistic for lean beef ,f,i,';,‘;‘,_§,,,§f 223$, ZQQSCE 15,1335, pmeuels 'e“9ee"°m eAee (rib) ‘e e-eswlreurld)? fer rib, nonlean round, and nonlean all other beef are "emea" bee‘ the range ‘e "em e-eee (m) ‘e e904 sensitive to changes in own-price as well. Finally, con- 37 venience rib, roast, ground beef, entree, and steak products are also highly sensitive to changes in own- price, all other things held constant. Cross-Price Elasticities Tables 18-20 show statistically significant cross-price elasticities corresponding to aggregate beef groups. In general, purchases of lean beef products do not respond to changes in the price of nonlean beef products (Table 18). However, except for brisket, ground, loin, and rib cuts, purchases of nonlean beef are sensitive to changes in the price of lean beef. Conse- quently, the price of nonlean beef is generally not a key determinant of purchases of lean beef, but the price of lean beef is a prime determinant of nonlean chuck, round, and all other beef. with a single exception, the price of convenience products has no statistically sig- nificant influence on purchases of lean and nonlean beef products. Likewise, except for convenience roast products, the prices of lean and nonlean beef do not significantly affect purchases of convenience beef products. On the whole, only 6 of the 57 cross-product price elasticities relevant to pork, poultry, and fish are statis- tically different from zero. Pork is a gross complement to lean brisket. Poultry is a gross complement to both lean loin and convenience ground beef. Fish is also a gross complement to both lean and nonlean brisket, but fish is a gross substitute for convenience entree. For fresh beef, cross-cut prices have a relatively minor influence on purchase patterns. Of the 42 cross- cut price elasticities for lean (nonlean) beef, 11 (6) are significantly different from zero (Tables 18 and 19). The price of lean chuck positively influences purchases of lean brisket, but the price of lean loin negatively influen- ces purchases of this product. For lean products, chuck and brisket are substitutes; loin is a complement to (substitute for) brisket and chuck (round); ground beef is a complement to loin, rib, and all other beef; rib is a substitute for ground beef and all other beef; chuck and round are substitutes; and brisket is a complement to ground beef. For nonlean beef products, rib and all other beef (round) are complements (substitute) to chuck; rib is a complement to both ground beef and all other beef; and round is a substitute for loin. For convenience products, cross-cut prices also play a relatively minor role. Of the 20 cross-cut elasticities for convenience beef products, 4 are statistically different from zero (Table 20). The price of rib exerts a negative influence on purchases of entrees and ground beef. Likewise, the price of steak negatively affects purchases of both entrees and ground. Own-Advertisement Elasticities Own-advertisement elasticities for fresh beef groups are exhibited in Tables 17 and 21. ln this study, for the broad groups in question, the own-advertisement elas- ticities are without exception positive and mostly statis- tically significant. The own-advertisement elasticities have more influence on purchases of nonlean beef products, for which all estimates are statistically sig- nificant, than on purchases of lean beef products. For lean beef, the only significant elasticities correspond to brisket, rib, and round. The elasticities for lean beef range from 0.040 to 0.073. For nonlean beef, the range is from 0.040 to 0.172. The magnitude of the own- adverlisement elasticities is much smallerthan the mag- nitude of the own-price elasticities. Cross-Advertisement Elasticities Statistically significant cross-advertisement elas- ticities corresponding to aggregate beef groups are exhibited in Table 21 . The cross-cut advertisement elas- ticities are marginal. For lean (nonlean) beef products, Table 18. Statistically significant’ price elasticities!’ for aggregate lean beef products. Cross-Cut Price Elastlcity° Cross-Product Price Elasticity“ b At the sample means. convenience beef, or nonlean beef. c°mm°d"y BrlsketlChuckl Grnd | Loin l Rib [Round] AOB Pork | Pauli] Fish [conv lNLean 31191911 3.399 -2.397 -1.19o -o.972 Chuck -2.514 Ground -1.185 -1.205 1.799 Loin 4.929 4.900 -0.197 Rib 4.929 Round 2.513 2.725 45.394 41.937 AOB 4 .549 1.02s 2.99s a At the 0.05 level of significance. ° Cross-cut price elasticity indicates the cross-price elasticity of a particular beef cut with respect to another lean beef cut. Cross-product price elasticity indicates the cross-price elasticity of a particular beef cut with respect to pork, poultry, fish, 38 Table 19. Statistically significant” price elasticities” for aggregate nonleanbeef products. Cross-Cut Price Elasticity° Cross-Product Price Elasticity“ Pork l Poult l Fish I Conv l Lean C°'“'“°""Y Brisket|Chuck| Grnd l Loin | Rib [Retinal AOB Brisket -5.732 i Chuck -2.902 -1 .614 Ground -1.209 -0.602 Loin 4.897 Rib -2.146 Round AOB -2.164 -2.672 0.970 -2.005 5.878 0.398 -3.756 6.992 -2.895 3.467 a At the 0.05 level of significance. At the sample means. ° Cross-cut price elasticity indicates the cross-price elasticity of a particular beef cut with respect to another nonlean beef cut. d Cross-product price elasticity indicates the cross-price elasticity of a particular beef cut with respect to pork, poultry, fish, convenience beef, or lean beef. Table 20. Statistically significant‘ price elasticities” for aggregate convenience beef products. Cross-Cut Price Elasticity° Cross-Product Price Elasticity“ c°mm°dw Rlb | Steak l Entree leround| Roast Pork [Poultry] Fish | Lean [NLean Rib 49.925 Steak -2.088 Entree 3.728 -1 .789 -3.127 1 .475 Ground -3.844 -1.144 -3.022 -0.741 Roast -4.692 -1 .136 a At the 0.05 level of significance. At the sample means. item. lean beef, or nonlean beef. ° Cross-cut price elasticity indicates the cross-price elasticity of a particular beef cut with respect to another convenience beef d Cross-product price elasticity indicates the cross-price elasticity of a particular beef cut with respect to pork, poultry, fish, only 9 (2) of 42 cross-cut elasticities are significantly different from zero. Advertisement elasticities for fish, pork, poultry, lamb, and veal on purchases of fresh beef products are, however, not statistically significant. Holidays and Seasonality As exhibited in Table 17, purchases of nonlean beef during holidays are not significantly different from non- holiday purchases. This result, however, is not evident for either lean beef or convenience beef products. In particular, purchases of lean ground, loin, round, and all other beef (brisket) are significantly lower (higher) during holidays than during nonholidays. As well, pur- 39 chases of convenience steak, ground, roast, and entree products are significantly lower during holidays relative to nonholidays, all other things held constant. Seasonal purchase patterns are evident for all broad convenience beet groups except entrees. Similarly, purchase patterns are evident for all broad lean beef groups except lean chuck and lean loin. However, only nonlean chuck and nonlean all other beef among the broad nonlean beef groups are subject to seasonality in purchases. The cornerstone of this analysis is the specification and estimation of econometric models to analyze pur- chases of beef products on a per 1 ,000 customer basis. The purpose is to identify and assess key factors that allow producers, processors, and distributors to analyze trends in retail markets, improve planning, and provide better service to consumers. The models adequately capture significant variation in purchase patterns and importantly are not subject to serial correlation problems. Key variables include own- price, prices of competing products, and own-advertise- ment effects. Retailers may utilize the models to assess promotional activity, to forecast purchases, and to deter- mine optimal space allocation. Because development of effective marketing programs is a primary concern of retail food chain executives, the analyses can be used to make pricing and advertising decisions. In particular, purchase patterns of the individual beef products in Table 21. Statistically significant’ advertisement elasticities” for fresh beef products by carcass section. Cross-Product , Advertisement Cross-Cut Advertisement Elasticity’ Elasticity Commod" Other Meat y Brisket Chuck Ground Lion Rib Round AOB Products° Lean Beef Brisket 0.07s“ Chuck 0.064 -0.059 Ground 0.013 -0.010 Loin 0.014 0.014 -0.019 -0.025 Rib 0.049“ Round 0.040“ AOB 0.015 Nonlean Beef Brisket 0.172“ 0.10s Chuck 0.097“ Ground 0.040“ -0.009 Loin 0.060“ Rib 0.059“ Round 0.109“ AOB 0.053“ a At the 0.05 level of significance. b At sample means. ° Fish, lamb, pork, poultry, and veal. Denotes own-advertisement elasticities. ° Cross-cut advertisement elasticity indicates the cross-advertisement elasticity of a particular beef cut (lean and nonlean) with respect to another beef out. ‘Cross-product advertisement elasticity indicates the cross-advertisement elasticity of a particular beef cut (lean and nonlean) with respect to other meat products (nonbeef). Conclusions and Implications for Further Research 40 question are highly sensitive to own-price changes and moderately sensitive to the effects of advertising. All other things held constant, given elastic demands for individual beef products, incentive exists for this firm to lower average prices for selected cuts to maximize total revenue. A strategy to increase advertisement exposure to boost demand for beef cuts may also be worthwhile. However, it is not possible to discern whether a strategy to reduce prices is preferable to a strategy to increase advertising exposure or vice versa. Such a determina- tion depends upon the costs of the respective strategies. Despite the apparent success in analyzing retail demand relationships with scanner data, concern lies with generalizing the results to regional or national levels. Scanner data from supermarkets in a particular location represent a "controlled" experimental situation. The community-specific results may not allow defen- sible, broad regional or nationwide influences. Because N. of this potential limitation, the results of local analyses (such as this study) should not be used on a stand-alone basis but as supporting evidence in conjunction with a research approach designed to conduct analyses with scanner data on a regional or national basis. Given that scanner data either on a local, regional, or national basis are available only from the private sector, given the potentially enormous cost considera- tions of either money or physical resources, and given the volume and integrity of scanner data, perhaps the single most important recommendation is for analysts and marketers to lobby heavily for the effective acquisi- tion and organization of scanner data. Although analysts typically do not have the comparative ad- vantage in data collection, they do have the comparative advantage in analysis. At least two ways exist to present arguments for acquiring and organizing scanner data. First, given the budget cutbacks of the federal government, which definitely influence data collection, it may be appropriate for public agencies (presumably either the Bureau of Labor Statistics or the United States Department of Agriculture) to negotiate with private firms (e.g., Infor- mation Resources, lnc.) to acquire scanner data. Costs for scanner data are not trivial, but neither are costs for various consumer surveys or panels. Furthennore, neither could an individual researcher efficiently collect or organize the volume of information nor could an individual researcher afford the information. Of course, the costs and benefits of this type of data collection require consideration. Second, if individual researchers banded together and combined efforts in collaboration with national retail food chains and/or commodity groups (e.g., the Beef Industry Council [BlC]), research with scanner data on a national or at least regional level would be cost effective. We appreciate the Beef Industry Council of the Na- tional Live Stock and Meat Board forfu nding this project. Special recognition is due to Sheila Courington, Greg Findley, and Burdette Breidenstein. As well, we thank Leonard Haverkamp for encouragement and support of this research. Moreover, we thank Neville Clarke, Robert Merrifield, Daniel Padberg, and John Nichols, all of the Texas Agricultural Experiment Station, for support of this project. We give special commendatio-ns to Nila Acknowledgments 41 lf neither of these two proposals for data acquisition is feasible, the individual researcher must focus on the local retail firm, which presumably has multiple stores. ln this instance, at least in the short-run, analysts can conduct research across the country, interacting with each firm at each location. Such a process, however, will require a unified effort for the acquisition, organiza- tion, and analysis of the data, so that, in some fashion, the results can be generalized over several regions. There must also be agreement on which commodities to analyze. Furthermore, agreement on which variables to incorporate in econometric models and which time frame to choose for analysis is essential. Obviously, these questions are not necessarily trivial. Though much recent empirical and theoretical work exists on demand and market analyses, reliable es- timates of demand parameters for individual beef com- modities are few. Much data are now available to food retailers because of scanning technology. These scan- ner data have tremendous potential for use in the analysis of consumer demand for specific products. Translating these data into information for manage- ment, advertising, and pricing decisions, however, remains a major concern. Scanner data indeed may result in the most detailed and definitive source of retail food industry statistics available to researchers and marketing executives. Use of scanner data can expand demand analyses. Scanner data promise fresh insights in market research. Although the realization of benefits from the use of scanner data is in the embryonic stage of development, in the next decade, analysts will concentrate on scanner data assembly, management, and analysis. Conceiv- ably, with proper management, scanner data may well be the ultimate data source of demand and market analyses at the retail level. This particular pilot study reveals the potential utility of scanner data in market research on beef products. Reece for programming expertise, data management, and computational assistance- Stacy Zemanek and Natalie South are recognized fortheir diligence in typing this manuscript. Finally, we acknowledge review com- ments from colleagues at Texas A&M University, name- ly Daniel Padberg, Gary Williams, and Teofilo Ozuna. Blattberg, R.C., and K.J. Wisniewski. 1986. Analyzing intra-category, inter-brand price competition: a theory of price tier competition. Working Paper, University of Chicago, January. Branson, R.E., et al. 1987. Data sources for demand analysis, in Food demand analysis: problem, issues and empirical evidence. R. Raunikar and C.L. Huang (eds.), lowa State Press. pp. 33-53. Branson, R.E., H.R. Cross, J.W. Savell, G.C. Smith, and R.A. Edwards. 1986. Marketing implications from the national consumer beef study. Western Journal of Agricultural Economics, 11, 1:82-91. Burke Marketing Research. 1987. The consumer climate for meat. Report to the American Meat Institute and the National Live Stock and Meat Board. Capps, 0., Jr. 1989a. Added convenience as a factor in at-home animal products demand, in The economics of meat demand. pp. 284-96. Capps, 0., Jr. 1989b. Utilizing scanner data to estimate retail demand functions for meat products. American Journal of Agricultural Economics, 71, 3:750-60. Capps, 0., Jr., D.S. Moen, and R.E. Branson. 1988. Consumer characteristics associated with the selection to try lean meat products. Agribusiness, 4, 6:549-557. Capps, 0., Jr., and J.M. Pearson. 1986. Analysis of convenience and nonconvenience food expendi- tures by U.S. households with projections to the year 2000, in Food demand analysis: implications for future consumption. O. Capps, Jr., and B. Senauer (eds.), Virginia Polytechnic Institute and State University, Blacksburg, Virginia, pp. 233-50. Capps, 0., Jr., J.R. Tedford, and J. Havlicek, Jr. 1985. Household demand for convenience and noncon- venience foods. American Journal of Agricultural Economics, 67:862-859. Carmen, H.F., and E.E. Figueroa. 1986. An analysis of factors associated with weekly food store sales variation. Agribusiness, 2:375-90. Chevalier, M. 1985; Increase in sales due to in-store display. Journal of Marketing Research, 12:426- 431. Cox, C., and R. Foster. 1985. What's ahead forthe U.S. food processing industry? Discussion. American Journal of Agricultural Economics, 67:1155-7. Literature Cited 42 Cox, K.K. 1964. The responsiveness of food sales to shell space changes in supermarkets. Journal of Marketing Research, 2:63-7. Cox, T.L., and M. Wohlgenant. 1986. Prices and quality effects in cross-sectional demand analysis. American Journal of Agricultural Economics, 68, 5:908-19. Curhan, R.C. 1973. Shelf space allocation and profit maximization in mass retailing. Journal of Market- ing, 37:54-60. . 1974. The effects of merchandising and temporary promotional activities on the sales of fresh f m its and vegetables in supermarkets. Journal of Marketing Research, 111286-94. Decisions Center, Inc. 1987. Giant lean awareness and usage. Research Report to the Center for Meat Marketing Research, American Meat Institute, Febmary. Doyle, P., and B.Z. Gidengil. 1977. A review of in-store experiments. Journal of Retailing, 53:47-62. Fomby, T.B., R.C. Hill, and S.R. Johnson. 1984. Ad- vanced econometric methods. Springer-Verlag. Funk, T.F., K.D. Meilke, and H.B. Huff. 1977. Effects of retail pricing and advertising on fresh beef sales. American Journal of Agricultural Economics, 59:533-7. Guadagni, P.M., and J.D.C. Little. 1983. A Logit model of brand Choice calibrated on scanner data. Marketing Science, 2(Summer):203-38. Holdren, B.R. 1960. The structure of a retail market and the market behavior of retail units. Prentice-Hall, Inc. ' Hoofnagle, W.S. 1965. Experimental designs in measuring the effectiveness of promotion. Journal of Marketing Research, 2:154-62. Houthakker, H.S. 1952. Compensated change in quan- tities and qualities consumed. Review of Economic Studies, 191155-64. Johnson, S.R., Z.A. Hassan, and R. Green. 1984. Demand systems estimation: methods and applica- tions. Iowa State Press. Jourdan, D.K. 1981. Elasticity estimates for individual retail beef cuts using electronic scanner data. M.S. Thesis, Texas. A&M University, College Station, August. L .y) Lemieux, C.M., and M.K. Wohlgenant. 1988. Consumer acceptance and demand for lean pork. Unpublished paper, North Carolina State University, Raleigh. Lesser, W.G., and J. Smith. 1986. The accuracy of supermarket scanning data: an initial investigation. Journal of Food Distribution Research, 17 (February):69-74. Marion, B.W., and F.E. Walker. 1978. Short-run predic- tive models for retail meat sales. American Journal of Agricultural Economics, 60:667-73. McLaughlin, E., and W. Lesser. 1986. Experimental price variability and consumer response: tracking potato sales with scanners. Cornell University, Agricultural Economics Staff Paper No. 86-28, Sep- tember. Menkhaus, D.J., G.D. Whipple, S.J. Torok, and R.A. Field. 1988. Developing a marketing strategy for branded, low fat, fresh beef. Agribusiness, 4:91- 103. i Moriarty, M.M. 1985. Retail promotional effects on intra- and interbrand sales performance. Journal of Retailing, 6:27-47. Morris, B. 1985. How much will people pay to save a few minutes of cooking? Plenty." Wall Street Jour- Q nal, August 25. *9. 4 Pasour E.C., and R.A. Schrimper. 1965. The effect of length of mn on measured demand elasticities. Journal of Farm Economics, 47:774-88. Progressive Grocer. 1989. Scanning update. October: 55. Schroeter, John. 1988. Analysis of behavior scan data and beef consumption responses to promotion and advertising. The Center for Agricultural and Rural Development, Iowa State University, July. Sexauer, B. The role of habits and stocks in consumer expenditure. Quarterly Journal of, Economics, 91 :127-42. Shugan, S.M. 1987. Estimating brand positioning maps using supermarket scanning data. Journal of Marketing Research, 24(February):1-18. Skaggs, R.K., D.J. Menkhaus, S.J. Torok, and R.A. Field. 1987. Test marketing of branded, low fat, fresh beef. Agribusiness, 3,3:257-72. i Stafford, T.H., and J.W. Wills. 1979. Consumer demand increasing for convenience in food products. Na- tional Food Review, (Winter):15-17. The Food Institute. 1986. Red meat and poultry: in- dustries in transition 1970-1985. June. \ . \‘ 43 Tomek, W.G. 1985. Limits on price analysis. American Journal of Agricultural Economics, 67( Decem- ber):905-15. Wall Street Journal. 1985. Beef‘s drop in appeal pushes some packers to try new products. August 28. Wilkerson, J.G., J.B. Mason, and C.H. Paksoy. 1982. Assessing the impact of short-term supermarket strategy variables. Journal of Marketing Research, 19:72-86. Wittink, D.R., M.J. Addona, W.J. Hawkes, and J.C. Porter. 1988. The estimation, validation, and use of promotional effects based on scanner data. Work- ing Paper, Cornell University, January. Wohlgenant, M.K. 1985. Estimating cross elasticities of demand for beef. Western Journal of Agricultural Economics, 10(2) :322-329. Wohlgenant, M.K., and W.F. Hahn. 1982. Dynamic adjustment in monthly consumer demands for meats. American Journal of Agricultural Econo- mics, 64:553-7. Yankelovich, Skelly, and White, lnc. 1983. The con- sumer climate for red meat. Report to the American Meat Institute and the National Live Stock and Meat Board. . 1985. The consumer climate for red meat. Report to the American Meat Institute and the Na- tional Live Stock and Meat Board. OBS CAIN-l OBS O m tOODNOJM-AODN-l U) COQNOJUI-bCON-B OBS -A OKOQNOIUI-bOON-i Appendix A - List of Individual UPCs Brisket UPC Description 2010390000 - Lean Line Trimmed Briskets 2024500000 Choice Boneless Brisket #062 2024510000 Choice Trimmed Brisket #067 Chuck UPC Description 2010600000 Lean Line Chuck Tenders 2010610000 Lean Line Chuck Tender Steaks 2020050000 Choice Chuck Tender #047 2020060000 Choice Chuck Tender Steak #048 2020070000 Gravy Steak (Cut From Chuck) #049 2020080000 Beef Chuck Shoulder Swiss Steak #050 2020090000 Beet Chuck Steak Center Boneless #051 2020120000 Beef Chuck Boneless Pot Roast #054 2020160000 Beet Chuck Eye Steak #060 Ground Beef UPC Description 2010470000 Lean Line Gourmet Ground Round 2010480000 Lean Line Gourmet Beef Patties 2026000000 Fresh Ground Beef #078 2026010000 Lean Ground Beet Chuck #079 2026020000 Extra Lean Ground Beef #080 2026030000 Ground Round Gourmet #081 2026050000 Ground Beef Patties #083 2026080000 Extra Lean Beef Patties #086 2026090000 Family Pak Ground Beef Loin UPC Description 2010200000 Lean Line Sirloin Strips 2010220000 Lean Line Fondue 2010240000 Lean Line Tailless T-bone Steaks 2010250000 Lean Line Top Sirloin Steaks 2010260000 Lean Line Tenderloin Roast 2010270000 Lean Line Tenderloin Steak 2010310000 Lean Line Sirloin Tip Fillets 2010340000 Lean Line Beet Stroganofi 2010440000 Lean Line Sirloin Tip Roast 2022030000 Choice Boneless Strip 12/14 #023 44 11 12 13 14 15 16 17 18 19 2O 21 22 23 OBS ii dO ‘ Statistically significant at the 0.05 level. Variable UC202012 UC202603 UC202602 UC202601 UC202600 INTERCEPT -15.8146 24.4183‘ 52.5746 313.0037‘ 90.7104 (-.297) (3.732) (1.402) (4.348) (.981) OWN-PRICE -.2333‘ -.0505‘ -.2827‘ -.5557‘ -.5269‘ (-6.654) (-5.827) (-9.177) (-11.756) (-7.955) PFLEAN .3336‘ .0108 .1708‘ -.2055 .3238 (2.341) (.615) (1.726) (-1.048) (1.340) PCON -.0441 -.0021 -.0388 .0321 -.1444 (-.891) (-.363) (-1.101) (.489) (-1.627) PFNLBRSK .0937‘ .0005 .0393 .0130 .1178 . (2.077) (.096) (1.258) (.214) (1.440) PFNLGRND -.0582 - - - - (-1 .104) - - - - PFNLLOIN .0051 .0003 .0102 .0053 -.0231 (.255) (.121) (.739) (.187) (-.692) PFNLAOB -.1253‘ -.0066 .0048 -.0634 .1797‘ (-2.251) (-.993) (.129) (-.859) (1.824) PFNLRND .0788‘ .0008 .0039‘ -.0029 .0631 (2.809) (.247) (1.761) (-.076) (1.255) PFNLCHCK - .0074 .0219 .0454 -.0581 - (1.615) (.833) (.866) (-8.53) PFNLRIB -.0743‘ -.0126‘ -.0378 -.1297‘ -.1756‘ (-2.126) (-3.263 (-1.634) (-2.864) (-2.925) ADVAOM .0004 .0002 -.0026 -.0003 -.0029 (.174) (.825) (-1.439) (-.081) (-.617) 62 h _ Variable uc2o2o12 uc202303 uc202302 uc202301 uc202300 k (\A03R|s1< .0145 .0007 .0050 .0121 .0105 -\ (1.150) (.435) (.573) (.710) (.459) ADRIB -.0243 -.0033 -.0213 -.0553* -.0432 (-1.023) (-1.150) (-1.237) (-1.703) (-1.112) ADAOB .0104 -.0015 .0143 -.0101 .0247 (.532) (-.743) (1.245) (-.443) (.314) ADROUND .0071 A .0092 A .0127 (.323) (.011) (1.553) (.007) (.330) ADGBEEF -.0121 .0005 .0073* .0133* 0707* (-1.339) (.375) (1.790) (1.930) (5.951) ADCHUCK .0341 * .0015 .0020 .0230* -.0135 (3.735) (1.239) (.307) (2.192) (-.795) ADLOIN -.0035 .0003 .0039 .0107 -.0195 (-.353) (.721) (1.309) (.733) (-1.092) M1 -1.5210 .3012 3.2901 5.4039 -7.2033 (-.343) (1.171) (1.102) (.923) -1.001) M2 -.5429 -.3513 -2.2233 5.5553 -13.1302* (-.114) (-.333) (.335) (.331) (1.792) M3 -1.0923 -.7337 -2.3753 -3.2319 -9.1042 (-.247) (-1.547) (-.797) (-.545) (-1.334) M4 4.2343 -.3932 2.7223 .1317 -3.7039 (.977) (-.770) (.913) (.022) (-1.230) M5 11.5333* -.4537 .3012 -2.5553 -13.3214* (2.304) (.334) (.100) (-.423) (-2.349) M6 .5005 -1.0359* -4.0334 -2.9311 -10.4325 (.102) (-1.347) (1.249) (.447) (1.330) M7 3.7975* -.7419 -.3270 -.1201 -.0934 (1.705) (1.232) (.240) (.017) (.144) M8 4.5037 -.3393 1.449 1.3333 -9.1394 (.392) (1.147) (.427) (.270) (1.432) M9 1.3123 -.5405 .5312 -4.3451 A (.404) (-1.115) (.191 ) (.727) - M10 1.5712 .0304 3.3211 2.2332 -3.5334 (.343) (.153) (1.194) (.339) (.323) M11 5.1913 .2237 2.0733 4.3032 1.5043 (1.233) (.459) (.724) (.349) (.244) H -1.3405 -1.3203* -3.3539* -7.5130* 3.2043 (.579) (3.923) (2.005) (2.023) (1.353) PORK -.0147 -.0022 -.0422 -.0241 .0193 (.403) (.513) (-1.353) (.493) (.299) POULTRY .0401 -.0024 .0177 -.0222 -.0099 (1.313) (.339) (.333) (.543) (-.131) FISH .0133 -.0003 -.0239* -.0133 -.0533 (.333) (.292) (1.323) (.359) (1.533) 0w 2.134 2.570 1.995 2.452 2.034 "A" denotes less than 0.0001. * Statistically significant at the 0.05 level. 63 Convenience Beef Products Individual UPCs Parameter Estimates (t-Values) Variable C5106313 Q§7337004 C1386630 C5015916 C55106322 u INTERCEPT 3.6566‘ 7.2894 5.8582‘ 1 .5492‘ 6.0479‘ (7.397) (1.529) (4.857) (2.200) (6.013) OWN-PRICE -.0079‘ -.0604‘ -.0139‘ -.0027‘ -.0157‘ (-14.162) (-10.769) (-9.159) (-3.117) (-11.562) PCSTEAK .0001 ~ -.0006 -.0009 - - (.346) (-.412) (-.941) - - PCGBEEF .0005 -.0042 -.0009 .0003 -.0005 (1.345) (-.950) (-.916) (.593) (-.627) PCROAST -.0003 .0040 -.0001 -.0004 -.0003 (-1.107) (1.155) (.150) (-.924) (-.621) PCRIB -.0012‘ .0059 -.0049‘ -A -.0024‘ (-2.137) (.944) (3.311) (.075) (-2.179) PCENTREE - - - .0001 .0006 - - - (-.254) (1.190) PORK .0002 .0025 -A .0005 .0004 (.611) (.665) (-.080) (1.103) (.646) POULTRY -.0002 .0010 A .0003 -.0005 (-.645) (.336) (.115) (.795) (-1.009) FISH A .0027 .0014‘ -.0003 .0004 (.321) (1.563) (3.411) (-1.247) (1.333) M1 .0219 .4666 .2627‘ .1299‘ .0534 (.745) (1.470) (3.362) (2.790) (.907) M2 -.0222 .1740 .2017‘ .1133‘ -.0335 (-.795) (.577) (2.723) (2.520) (-.590) u M3 -.0707‘ .7193‘ .1697‘ .0687‘ -.0304 (-2.630) (2.519) (2.322) (1.702) (-.563) M4 -.0692‘ .3422 .2588‘ .0902‘ .0373 (-2.730) (1.257) (3.702) (2.264) (.743) a M5 -.0639 1 .1660* 2679* .1575* .0766 (-1.663) (2.652) (2.600) (2.606) (1.002) M6 -.0834‘ -.0538 .4225‘ .0851 -.0253 (-2.588) (-.155) (4.614) (1.590) (-.396) 5 M7 -.0848‘ .1431 .1754‘ -.0197 -.0566 (-2.415) (.379) (1.659) (-.365) (-.619) M8 -.0801‘ -.1544 .1437 .0588 -.0519 (-2.164) (.366) (1.432) (.991) (-.711) M9 -.0628‘ -.0716 .2315‘ .0623 -.0531 (-1.856) (-.194) (2.541) (1.173) (-.805) M10 -.1172‘ -.2638 .0754 .1737‘ -.1796‘ (2.996) (.727) (.629) (3.166) (-2.211) M1 1 .0732‘ -.4288 .0551 .1433‘ .1394‘ (2.312) (-1.210) (.653) (2.713) (2.124) H -.0751‘ -.3521 -.2006‘ -.0739‘ -.1789‘ (-3.625) (-1 .566) (-3.634) (-2.378) (-4.540) PFLEAN .0004 -.0045 -A -A .0040‘ (.394) (-.417) (-.011) (-.064) (2.008) PFNLEAN -.0003 .0034 -.0006 -.0004 -.0005 (-.841) (.930) (-.686) (-. 773) (-.818) DW 2.125 2.582 1.786 2.155 2.567 Note: System R2 = 0.6236. "A" denotes less than 0.0001. * Statistically significant at the 0.05 level. 64 h Variable 05106324 c5106323 uc203939 c5106327 c7337006 f INTERCEPT 11.9671 * 3.3411* 14.5104* 3.2230* 7.9936 ‘" . (4.676) (6.569) (3.312) (5.503) (1.502) OWN-PRICE -.0279* -.0103* -.0162* -.0096* -.0673* (-10.952) (-13.769) (-4.903) (-11.607) (11364) PCSTEAK - -.0003 -.0062 - - - (-.336) (-1.591) - - PCGBEEF -.0003 -.0006 - -.0003 -.0041 (-.130) (-1.233) - (.673) (-.343) PCROAST .0023 - .0037 -.0001 .0050 (1.364) - (1.063) (-.463) (1.314) PCRIB -.0079* -.0014* -.0219* -.0001 .0047 (-2.335) (-2.134) (-4.115) (-.201) (.631 ) PCENTREE -.0006 0007* -.0037 .0004 .0017 (-.344) (2.291) (-1.223) (1.256) (1.311) PORK -.0003 .0001 -.0029 .0004 .0035 (-.143) (.293) (-.333) (1.210) (.346) POULTRY .0022 A -.0023 -.0001 .0015 (1.431) (.019) (-.919) (-.417) (.452) FISH .0013 .0002 .0033* .0001 .0037* (1.341) (.314) (1.906) (.542) (1.362) M1 -.0753 .0423 .3343 .0393 .4549 (-.443) (1.212) (1.019) (1.153) (1.279) M2 -.1032 -.0529 .4600 -.0531* .0345 (-.627) (1.550) (1.469) (-1.743) (.250) M3 -.0732 -.0715* .0252 -.0456 .7236* (.4951) (2.271) (.033) (1.502) (2.273) M4 .2203 -.0520* .0653 -.0391 .2937 (1.532) (1.722) (.240) (-1.334) (.931) M5 .3137 .0161 .1397 -.0117 1.2544* (1.391) (.351) (.349) (.261) (2.697) M6 .2462 -.0311* .1003 -.0540 -.1214 (1.292) (2.117) (.295) (1.453) (-.314) M7 .2030 -.0662 -.0715 -.0267 .0523 (1.011) (-1.592) (.197) (.666) (.125) M3 .1433 -0677 -.0294 -.0465 -.2794 (.634) (-1.543) (.075) (-1.094) (.629) M9 .2939 -.0391 .3399 -.0253 -.1902 (1.475) (.990) (.932) (.673) (.464) M10 -.2313 -.1336* .4404 -.1103* -.3309 (1.400) (2.329) (1.173) (2.314) (-.314) M1 1 -.0377 .0773* .0373 .0744* -.5313 (-.453) (2.044) (.101) (1.921) (1.463) H -.1745 0.0932* -.5405* -.0349* -.3726 (-1.535) (4.270) (-2.416) (3.702) (-1.490) PFLEAN -.0039 .0023* -.0021 .0021* -.0062 (.702) (2.370) (.190) (1.730) (.512) PFNLEAN .0013 -.0002 -.0016 -.0003 .0035 (.976) (.493) (-.454) (.366) (.350) 0w 2.237 2.604 1.713 2.535 2.555 "A" denotes less than 0.0001. ‘ Statistically significant at the 0.05 level. m \ ‘Q4. Fresh Lean Beef Products: Aggregate Commodities Parameter Estimates (t-Values) Variable FLBRISK FLCHUCK FLGRND FLLOIN INTERCEPT 0.6765 1.6656 6.1916 4.4676 (0.516) (1.596) (1.019) (1.569) PFLBRISK -0.0049 -0.00004 -0.0246* 0.0059 (-1.559) (-0.015) (-1.666) (0.660) PFNLEAN -0.0006 0.0009 0.0064 0.0007 (-0.667) (-1 .456) (1 .066) (0.509) PCON -0.0007 0.0005 0.0062 0.0009 (-0.960) (0.622) (0.976) (-0.572) PFLRIB 0.0009 -0.0006 0.0119* 0.0017 (1.660) (-0.464) (6.994) (1.226) PFLLOIN -0.0019* 0.0020* 0.0076 -0.0046* (-1.647) (-2.199) (1.576) (-2.196) PFLAOB 0.00006 -0.0004 -0.0011 -0.0017 (0.042) (-0.607) (0.154) (-0.517) PFLROUND 0.0007 0.00006 -0.0040 0.0006 (0.710) (0.102) (-0.924) (0.145) PFLGRND 0.0016 -0.0014 -0.0224* 0.0096‘ (0.569) (-0.665) (-2.052) (-1.619) PFLCHUCK 0.0067* -0.0007 0.0006 -0.0006 (6.041) (0.665) (0.141 ) (0.226) ADVAOM 0.00002 0.00006 -0.0006 0.00007 (0.420) (0.950) (-1.607) (-0.226) ADBRISK 0.0006* 0.00006 -0.0005 0.0006* (4.571) (0.292) (0.905) (1.951) ADRIB 0.00006 0.0001 -0.0042* 0.0006 (-0.174) (0.612) (-2.524) (0.667) ADAOB 0.00002 0.0002 0.0010 -0.0006* (0.104) (-1.054) (-1.292) (-2.219) ADROUND 0.0002* 0.0006* -0.0007 0.0005* (-1.666) (-4.176) (-1.566) (-2.646) ADGBEEF 0.00007 0.0001 0.0004 0.0004* (0.776) (1.169) (1.079) (1.956) ADCHUCK 0.00002 0.0004* 0.0004 0.0006 (0.175) (4.560) (0.752) (1.166) ADLOIN 0.0002 0.00007 0.0016* 0.0001 (1.264) (0.666) (2.426) (0.694) PORK 0.0014* 0.0005 -0.0006 -0.0005 (-2.619) (0.961) (0.250) (0.424) POULTRY -0.0005 -0.0006 0.0026 -0.0016* (-1.002) (-0.716) (-1.642) (-1.775) FISH -0.006* 0.0004 -0.0006 0.0006 (-1.654) (1.620) (0.560) (0.455) M1 0.1464* 0.0765 1 .2099* 0.2900* (2.406) (1.441) (4.271) (2.165) M2 0.1661* 0.0457 1.1406* 0.1645 (2.456) (0.756) (6.660) (1.117) M6 0.1469* 0.0964* 0.6720* 0.0464 (2.465) (1.729) (6.104) (0.652) M4 0.1969* 0.1074* 1.6661* 0.0527 (2.661) (1.776) (4.251) (0.656) 66 N Variable FLBRISK FLCHUCK FLGRND FLLOIN M’. M5 0.2782* 0.0789 0.8922* -0.0407 ‘~' (8.859) (1.182) (2.748) (-0.287) M8 0.1980’ 0.1070 -0.5788 -0.2895 (2.287) (1.899) (-1.454) (-1.448) M7 0.2088* 0.0471 -0.4821 0.8881* (2.407) (0.818) (-1.087) (-1.815) M8 0.2891* 0.0878 -0.5407 0.2845 (8.028) (0.475) (-1.824) (-1.228) M9 0.2528* 0.1175 0.4085 -0.1227 (2.752) (1.447) (0.988) (0.820) M10 0.2078" 0.0858 0.2897 0.0878 (2.749) (1.288) (0.885) (0.288) M11 0.0772 0.0818 1.0248* 0.0497 (1.211) (1.452) (2.498) (0.882) H 0.0850* 0.0597 -0.7215* -0.2888* (2.058) (-1.888) (-8.795) (-8.218) 0w 2.150 2.295 2.428 2.881 ADJ R2 0.5018 0.5712 0.5508 0.4857 * Statistically significant at the 0.05 level. Variable FLRIB FLROUND FLAOB /'\ INTERCEPT 0.8189 2.5829 9.8775* (1.477) (0.451) (1.705) PFLBRISK 0.0008 0.0188 0.0108 (0.595) (-1.021) (0.772) PFNLEAN 0.00001 0.0018 0.00005 (0.088) (0.542) (0.015) PCON 0.00009 0.0078* -0.0018 (0.879) (-2.547) (0.421) PFLRIB 0.0002 0.0018 0.0054* (0.788) (0.587) (1.915) PFLLOIN 0.0002 0.0144* 0.0080 (0.887) (8.875) (0.887) PFLAOB 0.0007 0.0048 0.0285‘ (-1.408) (0.780) (-4.882) PFLROUND 0.00007 0.0881* 0.0052 (0.228) (-8.254) (1 .249) PFLGRND 0.0018‘ . 0.0079 0.0229" (-1.898) (0.787) (-2.202) PFLCHUCK -0.0002 0.0184* 0.0088 (0.454) (8.588) (0.888) ADVAOM 0.000008 0.00008 0.00002 (0.287) (0,179) (0.109) ADBRISK 0.00001 0.000004 0.0002 (0.881) (-0.008) (0.411) ADRIB 0.0008‘ 0.0004 0.0017 (2.857) (-0.289) (-1.1 18) ADAOB 0.00009 0.0004 a 8.1892 (-1 .589) (0.552) (0.000) k. 67 Variable FLRIB FLROUND 51.403 ADROUND -0.00003 0.0013* -0.0001 (-0.944) (4.133) (0.259) ADGBEEF 0.00001 0.0001 0.0003 ‘ (0.451) (0.313) (1.592) ADCHUCK -1.2143 -0.0004 0.0002 (-0.004) (-0.359) (0.525) ADLOIN 0.000009 0.0003 0.0012* (0.231) (0.341) (2.290) PORK -0.000002 -0.0002 -0.0011 (-0.011) (-0.070) (-0.479) POULTRY -0.0002 0.0001 -0.0023 (-1.595) (0.071) (1.273) FISH 0.0001 0.0005 0.0021 (1.079) (0.345) (1.273) M1 0.0191 0.3534* 1.3992* (-0.973) (3.272) (5.132) M2 0.0344 0.3773* 0.3337* (-1.533) (2.334) (2.951) M3 -0.0309* 0.5015* 0.3093 (-3.143) (1.932) (1.153) M4 -0.0391* 0.4743 0.4399 (-3.137) (1.332) (1.533) M5 -0.0333* 0.3520 0.3444 (-2.979) (1.171) (1.111) M3 0.0534* 0.3313 0.2347 - (-2.129) (0.903) (0.393) Q M7 -0.0712* 0.2177 0.4904 (-2.300) (0.593) (1.294) M3 0.0445 0.1233 0.2313 (1.532) (0.341) (0.371) M9 0.0495* 0.0070 0.0254 (1.700) (0.013) (0.033) M10 -0.0711* 0.5495* 0.3432 (2.973) (1.714) (1.047) M11 0.0755* 0.1931 0.4715* a (3.740) (0.725) (1 .339) a H 0.0133 -0.5032* 0.3072* g (-1 .243) (2.332) (4.454) 0w 1.333 2.349 2.134 ADJ R2 0.4243 0.3939 0.3433 ‘ Statistically significant at the 0.05 level. Fresh Non-Lean Beef Products: Aggregate Commodities Parameter Estimates (t-Values) Variable FNLBRSK FNLCHUCK FNLGRND FNLLOIN INTERCEPT 234.0119 4.5323 353.3517* 134.1743‘ i (1.353) (0.071) (2.230) (2.395) i; PFNLBRSK 0.3733* 0.0317 0.1213 0.0213 (3.303) (1.512) (0.912) (0.553) i PFLEAN 0.0131 0.4243* 0.3210 -0.1179 (0.043) (2.149) (0.743) (0.930) J 68 Variable FNLBRSK FNLCHUCK FNLGRND FNLLOIN - 4.?“ PCON -0.0955 -0.0339 -0.0231 0.0054 (-0.371) (-0.533) (-0.191) (0.123) PFNLRIB -0.0037 -0.0935‘ 0.2332‘ -0.0202 (-0.035) (-2.235) (-2.197) (-0.343) PFNLLOIN 0.0707 -0.0233 -0.0223 -0.1471‘ (1.123) (-0.337) (-0.349) (-7.752) PFNLAOB -0.1907 0.1333‘ 0.0523 0.0122 (-1.219) (-2.373) (0.323) (0.253) PFNLRND -0.0037 0.0799‘ 0.0597 0.0451‘ (-0.033) (2.344) (0.713) (1.332) PFNLGRND 0.0395 -0.0334 10433‘ 0.0273 (0.579) (-0.517) (-3.553) (0.533) PFNLCHCK 0.1103 0.2733‘ 0.0147 -0.0035 (0.993) (-5.953) (0.123) (0.192) ADVAOM 0.0104 0.0021 -0.0043 0.0037 (-1.333) (0.375) (0.554) (-1.300) ADBRISK 0.1103‘ 0.0179 0.0133 0.0044 (3.050) (1.173) (0.503) (0.402) ADRIB 0.0137 0.0350 0.1204‘ 0.0037 (0.193) (-1.193) (-1.375) (0.313) ADAOB 0.0334‘ 0.0151 0.0041 0.0020 (-1.791) (0.733) (0.031) (0.135) ADROUND 0.0173 0.0052 0.0137 0.0025 (0.397) (0.505) (0.354) (0.335) .»~ ADGBEEF 0.0092 0.0059 0.1033‘ 0.0013 (0.439) (0.371) (4.903) (0.193) ADCHUCK 0.0030 0.0491‘ 0.0230 0.0032 (0.217) (4.253) (0.311) (-0.335) ADLOIN 0.0137 0.0127 0.0141 0.0403‘ (0.539) (-1.035) (0.437) (4.535) PORK 0.1101 0.0034 0.0533 -0.0257 (-1.055) (0.193) (0.523) (0.315) POULTRY 0.0131 0.0351 0.0433 0.0145 (0.207) (0.955) (-0.533) (0.549) FISH 0.1423‘ 0.0033 0.915 0.0023 (-2.339) (0.331) (-1.341) (0.143) M1 20.5323 0.2703 21.3494 3.3313 (1.333) (0.051) (1.343) (1.013) M2 15.5493 4.3942 12.3515 -2.1205 (1.113) (0.753) (0.332) (0.504) M3 17.2479 2.3339 9.3349 4.3703 (1.334) (0.533) (0.741) (1.113) M4 20.9544 3.0750 14.7732 7.3337‘ (1.352) (1.520) (1.130) (2.004) M5 33.9334‘ 17.4303‘ 9.3354 3.9331‘ (2.315) (3.214) (0.722) (1.733) M3 17.3093 3.3214 11.1033 3.0033‘ (1.205) (1.134) (0.750) (1.342) M7 12.3942 15.5377‘ 20.0322 3.3315‘ (0.315) (2.440) (1.279) (1.334) M8 14.5279 11.0532‘ 13.4700 3.3773‘ f\ (0.979) (1.730) (1.203) (1.333) k. 69 Q Variable FNLBRSK FNLCHUCK FNLGRND FNLLOIN M9 6.9065 7.6210 24.6997* 6.5206 (0.517) (1.664) (1.616) M10 4.6402 6.2676 22.6654* 6.6264 (0.665) (1.164) (1.655) M11 5.2407 6.5965‘ 26.0449* 6.1966 (0.461) (1.667) (0.666) H 6.4762 -0.9660 -6.7691 (1.022) (-0.267) (-1.507) 0w 2.117 2.146 2.672 ADJ R2 0.7764 0.9041 0.6202 * Statistically significant at the 0.05 level. Variable FNLRIB FNLROUND FNLAOB INTERCEPT 50.9646 66.4656 29.6572 (1.491) (-0.651) (0.519) PFNLBRSK -0.0056 -0.0207 0.0667 (-0.200) (-0.615) (0.799) PFLEAN -0.0066 0.6717* 0.6127* (-0.069) (6.144) (1.967) PCON 0.0026 -0.0655 -0.0041 (0.061) (-1.179) (-0.076) PFNLRIB -0.0992* -0.0462 -0.1615* (-4.620) (-0.660) -4.177) PFNLLOIN 0.0097 0.0096 0.0196 (0.690) (0.292) (0.666) PFNLAOB 0.01 14 -0.0526 -0.6696* (0.627) (-0.666) (-5.776) PFNLRND -0.0064 -0.4107* 0.0424 (-0.464) (-9.911) (1.691) PFNLGRND 0.0020 0.0496 0.0104 (0.057) (0.666) (0.179) PFNLCHCK -0.0142 0.0656 0.0066 (-0.571) (1.501) (0.162) ADVAOM -0.0005 -0.0024 -0.001 1 (-0.266) (-0.606) (-0.672) ADBRISK -0.0022 -0.0141 -0.0046 (-0.270) (-0.765) (-0.617) ADRIB 0.0941* -0.0094 -0.0262 (6.052) (-0.264) (-0.999) ADAOB -0.0064 -0.0059 0.0640* (-0.566) (-0.265) (1 .669) ADROUND -0.0065 0.0669* 0.0091 (-0.640) (5.072) (0.964) ADGBEEF 0.0020 -0.0071 -0.0056 (0.465) (-0.659) (-0.675) ADCHUCK -0.0066 0.0117 -0.0006 (-0.562) (0.665) (-0.056) ADLOIN -0.0006 -0.0160 0.0011 (-0.096) (-1.072) (0.106) 7O Variable FNLRIB FNLROUND FNLAOB PORK 0.0079 0.0557 0.0001 (0.341) (1.047) (0.003) POULTRY -0.0054 -0.0133 0.0224 (-0.323) (-0.303) (0.533) FISH 0.0153 -0.0043 0.0275 (1.272) (-0.174) (1.353) M1 -0.1235 2.5300 0.3271 (-0.044) (0.409) (0.175) M2 3.1071 -0.9973 -2.9429 (1.001) (-0.141) (-0.553) M3 2.0323 5.2095 -2.5537 (0.705) (0.943) (-0.550) M4 1.4439 -5.9377 1.7524 (0.512) (-0.919) (0.370) M5 2.7544 3.5372 14.4115‘ (0.955) (1.313) (2.957) M5 1.7332 1.9475 2.1534 (0.557) (0.255) (0.400) M7 0.5729 2.4545 4.9314 (0.159) (0.317) (0.354) M3 1.5007 1.3791 5.0579 (0.434) (0.249) (0.910) M9 1.0372 4.7223 1.0203 (0.349) (0.595) (0.204) M10 1.7371 5.2215 3.2551 (0.539) (0.774) (0.555) M11 1.9303 4.3520 2.9075 (0.730) (0.734) (0.537) H 0.5157 -0.7333 -2.9501 (0.337) (-0.139) (-0.953) 0w 2.473 2.513 2.102 ADJ R2 0.5095 0.3759 0.3141 * Statistically significant at the 0.05 level. Convenience Beet Products: Aggregate Commodities Parameter Estimates (t-Values) Variable CSTEAK CENTREE CGBEEF CROAST CRIB INTERCEPT 52.6643‘ 13.5603’ 13.3974‘ 8.9140* 6.0745’ (3.843) (1.837) (2.630) (3.570) (6.915) PCSTEAK -0.932* -0.0181* -0.0081* -0.0014 0.0011 (-7.606) (-2.742) (-1.783) (-0.637) (1.398) PCGBEEF 0.0040 0.0027 -0.0350* . 0.0023 -0.0002 (0.320) (0.401) (-7.557) (1.001) (-0.272) PC ROAST -0.0105 0.0006 0.0039. -0.0207* -0.0006 (-1.025) (0.114) (1.304) (-11.090) (-0.969) PCENTREE -0.0041 -0.0347* 0.0004 -0.0003 -0.0010 (-0.388) (-6.1 11) (0.096) (-0.187) (-1.543) PCRIB -0.0196 —0.0266* -0.0191* -0.0018 -0.0131’ (-1.083) (-2.730) (-2.851) (-0.557) (-11.272) ‘Statistically significant at the 0.05 level. 72 Variable CSTEAK CENTREE CGBEEF CROAST CRIB PORK 0.0101 0.0020 -0.0000 0.0010 0.0000 (1.201) (0.512) (-0.091) (0.004) (0.492) POULTRY -0.0100 -0.0029 -0.0000* -0.0004 -A (-1.205) (-0.020) (-2.000) (-0.200) (-0.097) FISH -0.0017 0.0079* 0.0011 -0.0000 -A (0.027) _ (2.005) (0.554) (-0.914) (-0.040) M1 0.1090* -0.0172 -0.1125 0.2040 0.0141 (0.240) (-0.015) (-0.010) (1.042) (0.200) M2 1.4007 0.1010 -0.0191 -0.2227 -0.1411* (1.520) (0.019) (-0.055) (-1.002) (2.045) M0 1.0407 -0.5092 -0.5042* -0.5047* -0.1201* (1.590) (-1.1 19) (-1.059) (-0.277) (-2.272) M4 1.2209 -0.5709 0.0050 -0.1207 -0.1210* (1.520) (-1.027) (0.019) (-0.002) (-2.009) M5 1.0470 -0.7705 0.2014 -0.0009 -0.0709 (0.040) (-1.100) (0.409) (-0.040) (-0.097) M0 0.7052 ~ -1.2500* -0.0500* -0.1915 -0.0047 (0.749) (-2.270) (-1.700) (-1.029) (-0.900) M7 -0.5050 -1.2900* -0.4000 -0.0420* -0.1902* (-0.459) (-2.170) (-1.109) (-1.707) (-2.700) M0 -0.1270 -1.1557* -0.5770 -0.2145 -0.2100* (-0.109) (-1.009) (-1.001) (-1.009) (-2.922) M9 1.2200 -0.5702 -0.0400 -0.0970 -0.0014 (1.109) (-1.010) (-0.090) (-0.511) (0.407) M10 0.0204* -0.4027 0.7755* 0.1090 0.0002 (2.902) (-0.071) (1.072) (0.004) (1.120) M11 2.2702* 0.1524 -0.1700 0.4559‘ 0.1000* (2.059) (0.250) (-0.429) (2.202) (1.920) H -0.1410* -0.5792* -1.0072* -0.2020* -0.0500 (-4.950) (-1 .090) (-4.010) (-1 .754) (-1 .250) PFLEAN -0.0105 0.0200 0.0100 -0.0000 -0.0019 (-0.002) (1.507) (1.420) (-0.105) (-0.909) PFNLEAN -0.0144 -0.0051 -0.0009* -0.0021 -0.0002 (-1.077) (-0.921) (-2.000) (-1.104) (-0.245) 0w 2.257 2.075 2.509 1.700 2.009 .ADJ R2 0.7027 0.0000 0.7400 0.7004 0.0205 "A" denotes less than 0.0001. [Blank Page in Original Bulletin] [Blank Page in Original Bulletin] Mention of a trademark or a proprietary product does not constitute a guarantee or warranty of the product by The Texas _ Agricultural Experiment Station and does not imply its approval to the exclusion of other products that also may be suitable. *- All programs and information of The Texas Agricultural Experiment Station are available to everyone without regard to race, color, religion, sex, age, handicap, or national origin. O.8M—4/91