sexal ‘uoums a6a||oQ . mums/RS KMSJGAQUfi ywgv sexal eu_|_ - 10109115 ‘JGHH ‘v p1eMp3 - uoums wawuadxg |em1|nog16v sexai musnan] DNISSC-IOOHd-(IOOJ svxcal 3H1 QNIQNVHNH HQJI QHIJJNHJRIQJJQ 3lN3WflOQQ 3lVl$ SVXHJ. 9661 w Nvr MIQEIBAINH INYV SVXHL _ AHVHEH 7551 zmanc-molq eta-a . , v - y!’ 611.35g? AHOLlSOdHG HLViS SW31 i "$53331 [Blank Page in Orignfl Bulletin] ‘ s, Opportunities for Enhancing the Texas Food-Processing Industry Marie Wildenthal Oral Capps, Jr. H. L. Goodwin, Jr. Teofilo Ozuna, Jr. Gary W. Williams, and John P. Nichols* *Respectively, assistant research economist, Transportation Economics Program, Texas A&M Uni- versity; professor, Department of Agricultural Economics, Texas A&M University; associate profes- sor, Department of Agricultural Economics, Texas A&M University; assistant professor, Depart- ment of Agricultural Economics, Texas A&M University; professor, Department of Agricultural Eco- nomics, Texas A&M University; and professor, Department of Agricultural Economics, Texas A&M University. Key words: food processing, input-output analysis. CONTENTS Page Executive Summary ......................................................................................... .. iii Acknowledgment .............................................................................................. .. iii Tables ............................................................................................................... .. iv Introduction ...................................................................................................... .. 1 Statement of the Problem ............................................................................ .. 1 Objectives .................................................................................................... .. 1 Forces Affecting the Food-Processing Industry ................................................. .. 2 Technological Factors .................................................................................. .. 2 Institutional Factors .................................................................................... .. 2 Economic Factors ........................................................................................ .. 2 Location Factors ................................. ....................................................... .. 2 The Texas Food—Processing Industry ................................................................ .. 6 Survey of the Texas Food-Processing Industry ........................................... .. 6 Industry-by-Industry Evaluation .............................................................. .. 13 Input-Output Analysis .................................................................................... .. 15 Transaction Table ...................................................................................... .. 15 Multipliers ................................................................................................. .. 18 Concluding Comments .................................................................................... .. 21 Summary .............................................................................. ................... .. 2 1 Industry-by-Industry Evaluation .............................................................. .. 23 Applications of this Research .................................................................... .. 24 Suggestions for Future Research .............................................................. .. 25 Conclusions ............................................................................................... .. 26 Literature Cited ............................................................................................... .. 26 EXECUTIVE SUMMARY Economic development concerns state governments because it directly af- fects residents’ standard of living. Therefore, state governments are interested in strategies for strengthening their economies. Measures of strength include out- put, income, and employment. States with substantial agricultural production can implement a strategr of maintaining or increasing markets for that produc- tion, positively affecting all the aforementioned measures. In some states, the focus has been to “add value” to raw agricultural commodities. This bulletin can aid governmental agencies to make funding decisions about competing opportunities for adding value in various food-processing industries. Texas’ food-processing industry is the focus of this research. A survey was con- ducted to determine the cost components of Texas food processors, the percent- ages of purchases from Texas, and the reasons for these purchases. The McMenamin-Haring (1974) procedure was used to update the Texas In- put-Output Model to investigate output, income, and employment impacts of increased processing of agricultural commodities. Multipliers were gleaned from the input-output analysis. The input-output results were combined with infor- mation about food-processing industry trends obtained from secondary sources and from the survey. As a result, potential benefits associated with the gain or loss of selected Texas food-processing plants were identified. ACKNOWLEDGMENT The authors wish to acknowledge the Texas Agricultural Experiment Station for financial assistance in working on this project. iii TABLES Page Table 1. Coefficients of localization for food-processing industries in the United States ...................................................................................... .. 5 Table 2. Distribution of total cost by cost components for responding Texas food-processing plants, 1989 ................................ .. 8 Table 3. Average percentage of purchases from Texas suppliers by responding Texas food-processing plants, 1989 ................................ .. 9 Table 4. Major reasons for buying materials as cited by Texas food-processing plants, 1989 (186 responding plants) ......... .. 10 Table 5. Distribution of respondents’ sales of Texas food products by market, 1989 .................................................................................... .. 12 Table 6. Percentage of respondents’ sales of Texas food products to Texas destinations, 1989 ................................................................. .. 13 Table 7. Sectors of the updated 1986 Texas Input-Output Model ................ .. 16 Table 8. Total dollar change in output per $1 direct change in final demand by sectors of the Texas economy, 1986 ..................... .. l8 Table 9. Total dollar change in income per $1 direct change in income by sectors of the Texas economy, 1986 ............................... .. 19 Table 10. Total change in person-years of employment per one person-year direct change in employment by sectors of the Texas economy, 1986 ....................................................................... ..20 Table 1 l. Effects of $1 million change in final demand on output, income, and employment for each sector of the Texas economy, 1986 .................................................. .._ ................... .. 21 iv INTRODUCTION State governments are understandably con- cerned with the condition of their economies — a state’s economic environment determines the potential standard of living for residents. There- fore, state governments are becoming increas- ingly interested in strengthening their econo- mies, specifically in terms of output, income, and employment. States that receive a substantial proportion of their revenue from agricultural production have the additional consideration of maintaining or in- creasing markets for that production. In some of these states, the focus has been on “adding value” to raw agricultural commodities (Capps et al. 1988). In 1989, Kraybill and Johnson identified agricultural value-added programs in 22 states. The goals of these programs were twofold: (1) to find new employment opportunities in rural ar- eas and (2) to create additional demand for agri- cultural products. In general, value-added oppor- tunities are important because they increase re- turns for production, thus enabling firms to grow. As new sources of economic activities for a region or state, these opportunities are also sources of increased jobs, employment, and government rev- enues (Nichols 1988). Investigation of value-added agricultural op- portunities has proceeded along three paths. The first examines historical records of state assis- tance to determine the feasibility and preferred implementation of prospective value-added ac- tivities (Kraybill and Johnson 1989; Moore 1987; Greene 1988; Texas Department of Agriculture 1986). The second recommends ways to develop a value-added research agenda, including con- sideration of its goals, potential inefficiencies, and potential effectiveness. The role of govern- ment institutions in value-added research is also addressed (Otto and Williams 1989; O’Rourke 1989; Ferris 1989; Holt 1989; Paarlberg 1989; Christy and Connor 1989; Meyers 1989; Kraybill and Johnson 1989; Tilley 1989; Giertz and Crihfield 1990). The third comprises quantita- tive models of the output, income, and employ- ment impacts of the introduction of food-pro- cessing plants to a region (Sporleder and Hushak 1991; Jones and Mustafa 1972). Statement of the Problem Each of the aforementioned approaches pro- vides important insights for defining the re- search problem described in this bulletin. State governments have limited resources to devote to research on value-added food products. Sug- gestions for developing a value-added research agenda contain recommendations about types of industries on which to focus and ways to as- sist new businesses; often, assessing the im- pacts of the recommendations is not done. Past quantitative research has analyzed output, in- come, and employment impacts of food indus- tries on regional economies. However, the data for these analyses often represent past trends, and current trends in the food-processing in- dustry are not considered. In addition, primary data pertaining to reasons processors purchase in-state versus out-of-state inputs are rarely col- lected. Increased purchases of in-state inputs would increase economic activity within the state. The purpose of this bulletin is to integrate primary and secondary data describing the food- processing industry with quantitative estima- tions of output, income, and employment im- pacts of potential food-processing plants. This information would greatly aid state governments in allocating their research funds for value- added activities. Texas provides an interesting example for value-added analysis. Reeh (1990) identified sev- eral Texas food industries in which the state’s share of national raw commodity production is greater than the state's share of the comparable processed product. These products are meat products, preserved fruits and vegetables, grain mill products, bakery products, and confection- ery products. This information indicates that value-added opportunities for increased food processing exist in Texas. In determining which opportunities can strengthen the Texas economy, decision makers could benefit greatly from data organized to facilitate impact assess- ment of an additional food-processing plant on Texas output, income, and employment and _ from data to evaluate these impacts in view of the Texas food-processing environment. Objectives The primary objective of this bulletin is to facilitate decision making in fund allocation for value-added opportunities among competing food-processing industries. This overall objec- tive will be accomplished by achieving the fol- lowing sub-objectives: (1) Identify food-processing trends. (2) Identify the cost components of food-pro- cessing industries and determine food processors’ preferences for in- or out-of- state commodities as well as the reasons for these preferences. (3) Investigate the output, income, and em- ployment impacts of food-processing plants. FoRcEs AFFECTING THE FOOD-PROCESSING INDUSTRY Authors generally agree that the three main forces affecting the food-processing industry are technological, institutional, and economic (Christy and Connor 1989; Capps et al. 1988). These three forces dynamically interact, which hinders independent discussion of these forces. Technological Factors Food-processing plants are designed and built for current technological processes. Tech- nologically older plants may not be able to com- pete with new plants because of either a lack of cost competitiveness or an inability to manu- facture new products. Recently, there have been many technological innovations in the food-pro- cessing industry (Sanderson and Schweigert 1988; Logan et al. 1988; Babb and Chambers 1988). Examples of these innovations include aseptic packaging, irradiation, and biotechno- logical advances. Increased market concentration fosters these technological innovations, which increases rivalry in new product development. Increased market concentration also broadens the finan- cial base of firms through both increased re- tained earnings and an increased ability to bor- row (Christy and Connor 1989). In this innova- tive environment, opportunities may exist for a technologically new value-added plant to replace a technologically obsolete plant. Institutional Factors While technology influences the competitive need for new plants, institutional factors influ- ence the ability to create new plants and the incentive for their creation. National legislation affects the economic environment. For example, fiscal and monetary policies influence interest rates, which alter the affordability of capital (Christy and Connor 1989). Immigration laws affect the increasingly immigrant food-process- ing labor supply (Connor et al. 1985). Foreign, fiscal, and monetary policies influence the ex- istence of export markets (Connor 1988; McCorkle 1988; McCorkle et al. 1988). Environ- mental and food safety regulations affect the cost of compliance with these regulations (Archibald and Dahl 1988; McCorkle et al. 1988; Connor et al. 1985). Taxation levels influence the amount of revenue necessary to make a profit (McCorkle 1988). All these factors also affect the cost and availability of food-processing inputs; farm and other manufacturing operations are influenced by these regulations, as well as regu- lations specific to their industries (McCorkle et al. 1988). When choosing location within the country, however, the food processor can choose the type of state and local legislation to which it is sub- ject. Applicable regulations relate to workmen’s compensation insurance, taxation, pollution, and many other aspects affecting the cost of doing business. However, regions can choose to provide financial assistance in terms of timing and mode of investments, reducing regulatory burdens, and supporting a plant’s research and development efforts (Connor 1988). State gov- ernments need to be sensitive to these factors if they are trying to encourage economic growth through business expansion. Economic Factors A food-processing firm determining whether and how it should expand considers the pro- duction costs of competing locations (Connor et al. 1985), the competitive structure and organi- zation of the industry, and the demand for the proposed product (both in the United States and abroad). A state government seeking to attract a new plant to the state should be aware of these factors as well. Production Costs Food-processing requires inputs of labor, raw food, energy, and capital. A food processor com- petes with other manufacturers for labor, en- ergr, and capital. The availability of all inputs at competitive prices, especially during peak processing seasons, is of primary importance when choosing a food-processing plant location. Christy and Connor (1989) noted that hav- ing made estimates of future demand growth and major cost calculations, a food-processing man- ager is likely to find that each of several final candidate locations is about equally cost effec- tive. In this case, the business climate (tax and regulatory considerations discussed in the pre- vious section, for example) or managerial pref- erences lead to the final decision. From a local government's perspective, a per- ceived benefit of value-added activity is that new plants will purchase some of their inputs from local suppliers, thereby boosting the local economy. Factors involved in the decision to purchase inputs locally include input availabil- ity, cost, quality, transportation costs, a firm’s national contract, and sole suppliers of certain inputs. Secondary information on these factors is not readily available, and a region might have to survey its food processors to determine how these factors influence the input purchasing decisions of its food processors. Industry Structure Industry structure can affect production costs as well as the availability of markets. Connor et al. (1985) list many aspects of mar- ket structure, including concentration, condi- tions of entry, product differentiation, market shares, advertising and other selling costs, economies of scale, multiplant economies of scale, sales diversification, and conglomeration. Building a new value-added processing plant requires considering all elements of industry structure. Barriers to entry are the initial ob- stacles to overcome. Costs involved in advertis- ing, product differentiation, building the opti- mal size plant, and so forth must be calculated to ensure that projected revenue can cover pro- jected costs. When an existing firm expands, optimal plant and multiplant economies of scale are most relevant. By investigating these fac- tors, a state government can get a better idea of whether a type of food-processing industry mer- its further consideration when choosing the one or those in which to invest. Demand for Food As stated earlier, a key factor in the success of a new food-processing plant is having a prod- uct that people want to purchase. Economic theory relates that demand for any product de- pends on income, relative prices of commodi- ties, and preferences. To distinguish among food industries, preferences are the focus of this dis- cussion. Preferences as revealed in trends in processed food consumption are examined ini- tially. The demographic composition of the popu- lation and the preferences of different demo- graphic groups are also important and are dis- cussed in following text. Finally, foreign prefer- ences are examined and export opportunities are described. ~ Domestic Consumption Trends Red meats, including prepared red meats (ham, bacon, sausages), seem to be losing mar- ket share to other high-protein food such as poultry, fish, and tofu (Connor and Wills 1988; Logan et al. 1988). Except for the relatively rapid growth of cheese (including imitation cheeses), most dairy products are expected to experience low to moderate growth (Connor and Wills 1988; Babb and Chambers 1988). Most processed fruits and vegetables will probably experience below-average growth; the exceptions are the fast-growing frozen fruit and vegetable indus- tries, in which average growth is expected (Connor and Wills 1988; Carman and French 1988). The high growth in rice milling of recent years was due to expansion of overseas mar- kets; these projections may be too high if less- developed countries achieve the self-sufficiency they seek (Connor and Wills 1988; Farris et al. 1988). Increased domestic substitution of fructose for sucrose, as well as export demand, bodes well for the wet-corn milling industry, unless nonnutritive sweeteners make substantial in- roads. Relatively strong growth for the animal and vegetable oil industries is expected result- ing from demand by export markets and the food service industry (Connor and Wills 1988). Among beverages, the recent rapid growth in beer, wine, and soft drinks is likely to decline from the high 1958-1977 rates but to remain above average (Connor and Wills 1988; Farris et al. 1988). Many of the industries making highly convenient and sweet or high-calorie foods (e.g., canned specialties, frozen foods, candy, breakfast cere- als) seem likely to experience average to below- average growth. Popular new items (e.g., frozen quiche for microwave ovens) can experience ini- tial bursts of extraordinarily rapid growth, which later decline to average or below-average levels (Connor and Wills 1988). Demographic Trends Demand is directly related to changes in to- tal population (Meyers 1989; Christy and Connor 1989). However, the demographic composition influences the impact of the population change on processed food consumption. Demographic factors emphasized by Capps et al. (1988) in- clude age, household size, labor force, and eth- nic composition. Kinsey and Heien (1988) also considered health consciousness to be a factor. These five demographic factors are discussed here. Aye The U.S. population as a whole is aging (Babb and Chambers 1988), which affects the type of foods consumed. Kinsey and Heien (1988) cited several studies by Salathe (1979) that in- dicated that older people will consume more cheese, poultry, coffee, and tea. Gallo et al. (1979) indicated that consumption of some ce- real products, milk, soft drinks, red meat, pre- pared foods, and food consumed away from home (FAFH) has declined for elderly people. Buse and Fleishner (1982) contended that eld- erly people living alone spend a larger propor- tion of their budgets on fresh fruits and veg- etables, cereals, bakery goods, and poultry than does the average consumer. Household Size Connor et al. (1985) considered the decline in household size to significantly affect the change in demographic composition of the U.S. population. Smaller households cannot realize the same economies of scale in food consump- tion as do larger households. Smaller house- holds spend more per capita on bakery prod- ucts, cheese, soft drinks, fish, and food away from home (FAF H) (Sexauer and Mann 1979; Connor et al. 1985). Connor et al. ( 1985) also noted that larger households consume relatively more dairy products, cereals and breads, pork, poultry, eggs, sugars and sweets, coffee, soup, and prepared foods. Labor Force Participation The husband and wife both work outside the home in 40% of all family units (Connor et al. 1985). The major impact of increased participa- tion of women in- the labor force has been higher demand for convenience foods (Capps et al. 1985; Redman 1980). The consumption of con- venience foods and FAFH will increase as more women become employed full time. This scenario implies that high-quality, easily prepared food will be increasingly demanded. Ethnic Food Consumption Sanderson and Schweigert (1988) noted an increasing interest in ethnic foods in the United States. This interest is reinforced by the growth of minority populations in this country (Capps et al. 1988). Internationalization of the food in- dustry will be strengthened by increasing num- bers of travelers among countries, increased glo- bal communications, and by the international business activities of most major food compa- nies. These companies contribute to the demand and market for new food products from foreign lands (Sanderson and Schweigert 1988). Specific foods affected by this trend include spices, resulting from the increased popularity of Mexican and other ethnic foods. Prepared foods are also bolstered as many ethnic foods are sold partly or fully prepared. Frozen Mexi- can and Italian foods were predicted to grow at a rate of 15% between 1982 and 1992 in terms of dollar sales (Kinsey and Heien 1988). Recent increases in the demand for Mexican-style tor- tillas and snack foods have increased the con- sumption of special corn meals and flours (Farris et al. 1988). Health Consciousness A significant trend in food consumption has grown out of the improved information on diet and health (Kinsey and Heien 1988). Processes that prolong storage and shelf life of fresh foods will take on more importance as consumers de- mand more fresh and / or natural foods. Trends can be anticipated in specific food products. Future changes in the demand for dry- milled corn products may occur if corn bran and corn germ flour are successfully marketed to health conscious Americans (Farris et al. 1988). Consumer concerns about the cholesterol con- tent of red meats and dairy products possibly being related to heart disease appear to have negatively affected consumption of these prod- ucts (Brown and Schrader 1990; Capps and Schmitz 1991; Logan et al. 1988; Babb and Chambers 1988). Processed pork products have the additional disadvantage of being associated with nitrosamines and excessive salt (Logan et al. 1988). Poultry consumption may be increas- ing faster than that of red meat. This may be due to concerns about the cholesterol content of red meat (Logan et al. 1988). The interest in natural foods and more active lifestyles may positively affect dairy consumption (Babb and Chambers 1988). Export Opportunities Popular opinion suggests that the United States should export processed foods and thus capture more of the value-added activity through overseas sales. Many authors, however, believe that the United States should concentrate on exporting primarily raw agricultural commodi- ties instead of processed agricultural commodi- ties (Connor et al. 1985; Kinsey and Heien 1988; McCorkle 1988; Paarlberg 1989; Connor 1988). Kinsey and Heien (1988) as well as Connor et al. (1985) noted that the United States has a comparative advantage in exporting agricultural commodities. Despite efforts of United States manufacturers to add value to basic products for export abroad, export growth continues to be weighted toward raw agricultural commodi- ties (McCorkle 1988; Christy and Connor 1989). Although it may be true that the United States should export primarily raw agricultural products, a small export market does exist for processed agricultural products. Connor et al. (1985) noted that the demand for high-value products will increase in developed and middle- income countries because of tourism and influ- ence of American eating patterns. Sanderson and Schweigert (1988) added that the global communications system is reinforcing this de- mand, as are the numerous food companies participating in international business activities. In addition to processed food, developing coun- tries will demand more semiprocessed products for use in local food-processing. Location Factors At the beginning of this section, the success of a new food-processing plant was noted to be a function of consumer's demand for the prod- uct and the state’s competitiveness in its pro- duction. Here, the competitiveness of a plant in rival locations is examined by identifying char- acteristics of an optimal plant location. Factors involved inf the location of a plant include in- dustry competitive structure, transportation infrastructure, adequacy of labor supply, re- moteness from market, and business climate, namely, the impact of community regulations that affect the ability of the firm to carry out business activities. For example, to conform to the perfect competition model, an industry must seek to exist in the least-cost location to remain in business. The coefficient of localization is a quantita- tive measure that provides a basis for a prelimi- nary decision about which industries to explore further for possible relocation. It measures the relative regional concentration of a specified type of food manufacturer compared with a given base such as population or income (lsard 1960; Capps et al. 1988). If foodprocessing were dis- tributed equally over the population, the coeffi- cient would equal zero. If it were concentrated in one region, it would equal one. Using Isard’s procedure, the coefficients of localization in Table 1 were calculated with employment data for each state and the District of Columbia as listed in the Census of Manufactures (U.S. De- partment of Commerce 1987). Food industries oriented to local consumer demand have coefficients close to zero. There are several reasons for orientation to consumer de- mand: the manufacturing process may result in an increase in bulk or weight of the inputs, a high degree of product perishability may exist, or the market may be specialized and concen- trated. Examples of demand-oriented food in- dustries include packaged fluid milk, ice cream, bread, rolls, cakes, flour, potato chips, shoe- string potatoes, and bottled soft drinks. Isard (1960) noted that industries with coefficients close to zero are relatively nonconcentrated re- gionally and are candidates for moving into an area seeking economic growth. The coeffi- cients of localization listed in Table 1 are closer to zero than they are to one, so these industries are all candidates for introduction into Texas. Table 1. Coefficients of localization for food-processing. Indus- tries in the United States. Industry Coefficient of localization‘ Meat 0.35 Dairy 0.25 Fruit and vegetables 0.32 Grain milling 0.30 Bakery products 0.16 Sugars and confections 0.34 Fats and oils 0.27 Beverages 0.23 Miscellaneous 0.24 ‘Coefficients were calculated as per Isard (1960) using Census of Manufactures (U.S. Department of Commerce 1987) employment data for each state and the District of Columbia. Food industries located near input supplies have coefficients close to one. This proximity to input supplies may be desirable because of great _weight or bulk loss during the manufacturing process, perishability of the raw material, or transportation rates. Examples of industries include meat packing, meat processing, fruit and vegetable freezing, rice milling, corn refining, and sugar beet processing. In Table 1 , however, the coefficients calculated for these industries are not close to one. This discrepancy could be due to the aggregated nature of the data. In ad- dition, more sophisticated techniques may be required in future studies that determine planned plant location. Food industries for which location decisions are not dominated by cost of delivery to cus- tomers or access to nearby input supplies are labeled “footloose” or “mobile.” Examples of these industries include breakfast cereals, prepared flour mixes, confectionery, frozen foods, yogurt, processed and imitation cheese, and shorten- ing and cooking oils. Note that even with a strong attraction to demand or supply, a plant may not locate where expected for reasons mentioned at the beginning of this section. THE TEXAS FOOD-PROCESSING INDUSTRY In the previous section, trends in the na- tional food-processing industry were examined. However, specific information about the Texas food-processing industry is necessary to deter- mine the potential for enhanced food-process- ing in Texas. Specific information about cost components of thefood-processing industries, food processors’ preferences for in- or out-of- state commodities, and reasons for these pref- erences is required. This information is gener- ally not published, but can be obtained from certain governmental agencies or by surveying food processors. Preferences of food processors and rationales for particular input sources can be determined only by survey methods. This section discusses the Texas food-processing in- dustry by describing the survey methods used in this study and presenting the survey results. n Survey of the Texas Food-Processing Industry The Sample The information sought in this survey was obtained from Texas food processors. A list of these manufacturers is available in the 1990 Directory of Texas Manufacturers compiled by the Bureau of Business Research at the Uni- versity of Texas at Austin. For this research, the Texas food-processing population consisted of the 1,175 plants listed in the 1990 Directory of Texas Manufacturers (Bureau of Business Re- search 1990) that manufactured products clas- sified under Standard Industrial Classification (SIC) code 20, food and kindred products. In addition, the sample consisted of the entire population because each plant was surveyed. The 1 990 Directory of Texas Manufacturers (Bu- reau of Business Research 1990) lists as many as five different products produced by each plant but does not give the percentage of the total sales of each product. Therefore, the plants listed in the 1990 Directory of Texas Manufacturers (Bu- reau of Business Research 1990) were classi- fied according to the first product listed on the assumption that this product would be produced in the largest volume. Wildenthal (1992) com- pared the characteristics of the plants listed in the Census of Manufactures (U.S. Department of Commerce 1987) with those listed in the 1990 Directory of Texas Manufacturers (Bureau of Business Research 1990) and found that the percentages of plants in each category for both sales and employment are highly similar, thus supporting the use of the 1990 Directory of Texas Manufacturers (Bureau of Business Research 1990) list of food-processing plants to obtain the Texas food-processing population. Questionnaire Design The questionnaire for this research was de- signed to collect the following information about Texas food-processing: (1) description of the cost components of food-processing industries; (2) reasons for input purchases from Texas versus those from other states; (3) the distribution of processed food products among various mar- kets; and (4) the proportion of the output sold in Texas. Because of cost and survey complex- ity, personal and telephone surveys were not considered as methods of questionnaire admin- istration. Therefore, the survey was designed as U; '5 a mail survey. The chief executive officer, as opposed to the purchasing or sales agent, of each plant was the recipient because questions re- late to both sales and purchases. Surveys from the University of Washington, the Nebraska Department of Labor, the Univer- sity of Illinois, the Texas A&M University Sys- tem, and the University of Wisconsin were re- viewed during questionnaire development. The survey asked for plant sales, purchases, employ- ment, capital expenditure information; a listing of major raw materials and inputs and whether and why they were purchased inside or outside of Texas; and investment in entrepreneurial ac- tivities. The question content, response format, and question sequence were patterned after those of the questionnaires discussed in the previous paragraph. The questionnaire was re- vised after pretests with a local ice cream manu- facturer, baker, and meat processor. Questionnaire Administration The printed eight-page questionnaire was mailed with a nonpersonalized cover letter hav- ing a social utility appeal and signed by Texas A&M University professors recognized by many in the industries being surveyed. The initial questionnaire was mailed under bulk postage, but subsequent mailings were with first-class postage. Confidentiality and anonymity of ques- tionnaire respondents were assured. A deadline was given for a response, and the respondents were asked to indicate whether they wished to receive a copy of the questionnaire results. Three follow-up questionnaires were mailed, and all questionnaire recipients who had not responded were then called. The complexity, time require- ments from recipients, and length of the ques- tionnaire were more than would generally be considered acceptable for a mail survey, how- ever. As a result, the response rate was lower than otherwise would be desirable. Data Analysis The data gathered from the survey were ana- lyzed in three different ways: comparing the sur- vey response rate to that of previous studies; comparing the respondents with the population; and calculating descriptive statistics from the questionnaire results. Survey Respondents Of all the 1,175 plants in the Census of Manufactures (U.S. Department of Commerce 1987), only 999 were found to be operational food-processing plants. Responses were received from 186 plants, a 19% response rate. This re- sponse rate is not atypical (Mautz and Neumann 1990; Dommeyer 1985). Comparison of the Population and Ques- tionnaire Respondents Comparison of the population and question- naire respondents is important because of the possibility of reaching incorrect conclusions from nonrepresentative respondents (Tull and Hawkins 1987). The comparison of the popula- tion of Texas food processors and the question- naire respondents was conducted on the basis of simple descriptive statistics for food indus- try, size, and geographic distribution as well as a qualitative choice model to determine the rep- resentativeness of the sample. Descriptive Statistics of the Population and the Sample The Texas food-processing population for this research is defined as the plants listed in the 1 990 Directory of Texas Manufacturers (Bu- reau of Business Research 1990), whose prod- ucts can be classified in SIC 2O (food and kin- dred products). Wildenthal ( 1992) showed in de- tail that the percentage of responding plants in each Texas Input-Output Classification (TIOC) is similar to the distribution of population plants for both food-processing industries and for num- ber of employees. Because of the large percent- age of Texas plants that did not release their sales information to the 1 990 Directory of Texas Manufacturers (Bureau of Business Research 1990), the representativeness of the question- naire respondents in terms of sales volume is not clear. The percentage of responding plants in the Far West/ Panhandle and Central regions is much like the percentage of plants in the population of those regions, while the percent- age of responding plants in the East and South regions is slightly higher than the percentage of plants in the population of those regions. Qualitative Choice Model A sample has response bias if some people being surveyed are more likely to respond than others. A statistical way of examining the re- sponse bias of the sample is through the use of a qualitative choice model. The aspect of in- terest is the decision of a food processor to re- spond to a questionnaire. Therefore, this decision is the dependent variable in the model. In addi- tion, the choice made is associated with charac- teristics of the food processors. The characteris- tics of interest are food industry, number of em- ployees, sales volume, and geographic location. The qualitative choice analysis, its assump- tions, and specific procedures are explained in detail by Wildenthal (1992). Plant characteris- tics of employment, location, and sales did not have a statistically significant impact on Whether a food processor completed the survey. Of the food-processing industries, beverage, meat-pro- cessing, and grain mill plants were more likely to complete the survey than were “other food” processors, while bakery, poultry, preserved food, feed, and dairy processors were less likely to complete the survey. Perhaps in future stud- ies, bakery, poultry, preserved food, feed, and dairy processors may need to be oversampled to obtain a more representative sample. Questionnaire Results As stated in the questionnaire description, the desired information can be classified into four different categories: (1) descriptions of the cost components of food-processing industries; (2) reasons for input purchases from inside Texas or outside Texas; (3) distribution of Texas food products between various markets; and (4) proportion of the output sold in Texas. Cost The cost components for the responding Texas food-processing plants are shown in Table 2. These cost breakdowns provide an estimate of the impact of a food-processing plant being introduced into a community. The importance of each category of input can be assessed for typical plant cost components. In this way, if an input category is found inferior or superior to out-of-state inputs, a better assessment of the costs and benefits of changing or promoting the input can be determined. For the purposes of this discussion, inputs are classified into four categories. Inputs with SIC codes below SIC 20 are classified as raw materials. These inputs include raw agricultural products and minerals. Inputs in SIC 20, food and kindred products, are classified as pro- cessed food inputs. Those with SIC codes above SIC 20 are classified as nonfood inputs. These inputs include packaging materials, paper prod- ucts, food preservatives, vitamins, and other materials. The final category is “other materi- als” purchased, such as cleaning supplies. Of particular interest to an agricultural com- munity is the percentage of resources spent on raw products and the percentage of these raw products purchased locally. Table 2 reveals some anomalies. For example, beverage plants do not appear to purchase energr or water or to have depreciation expenses. Although problems exist with the collected data, some generaliza- tions about raw material purchases can be made. Most expenditures of the responding Table 2. Distribution of total cost by cost components for responding Texas food-processing plants, 1989. Food industry Prsv. Other Material Meat Dairy Feed Bread food food Beverages Ail (Average percentage of total cost) Fiaw ingredients 3.7 59.8 14.9 9.2 29.0 0.8 10.0 15.1 Processed food products 37.1 5.4 23.7 40.3 19.4 78.3 15.0 56.4 Processed nonfood products 4.8 13.6 6.7 6.5 18.9 4.3 22.5 7.8 Other materials 4.6 1.7 38.6 7.0 6.7 0.2 2.5 2.0 Wages/salaries 28.6 7.6 7.0 17.8 10.2 6.8 12.5 7.7 Overhead 13.4 2.9 1.5 8.5 3.9 4.2 25.0 4.1 Depreciation 1.8 2.1 1.8 3.3 4.7 1.3 0.0 1.9 Other production cost . 2.3 4.3 3.7 6.0 4.2 1.9 12.5 2.7 Energy 2.8 2.0 1.8 1.4 2.4 2.1 0.0 2.1 Water 0.9 0.6 0.3 < 0.1 0.6 < 0.1 0.0 0.2 Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Source: Calculations by the author from survey results. plants are weighted more toward materials (raw materials, processed food inputs, processed nonfood inputs, and other materials) than to- ward nonmaterial inputs (wages / salaries, over- head, depreciation, other production costs, en- ergy, and water). The exceptions are meat and beverage plants, for which expenditures are dis- tributed almost equally between both costs. The material expenditures of dairy and preserved— food-processing plants are weighted more heavily toward raw ingredients, while those of meat, bakery, and other food-processing plants are spent more on processed food inputs. Bev- erage plants purchase more processed nonfood inputs than other inputs, and feed manufactur- ers spend more on other materials. Percentage of Materials Purchased in Texas Of interest in this research is the percentage of money spent for out-of-state inputs. This money might instead be redirected toward Texas input purchases, thus strengthening the Texas economy. If a plant is being encouraged to locate in Texas, the availability of inputs is also a con- sideration. The percentage of raw ingredients, pro- cessed food, nonfood, and other materials pur- chased in Texas by responding Texas food pro- cessors is shown in Table 3. These figures indi- cate which input categories have the greatest po- tential for increased purchases in Texas. This in- formation can help explain reasons the respond- ing food-processing plants give for their purchase of inputs from in- or out-of-state sources. Note that neither grain mill nor poultry plants re- sponded to this section of the questionnaire, so comments from these industries are not reported. Raw ingredients. Responding dairy, feed, and beverage processors purchase nearly all their raw ingredients from Texas suppliers. Respond- ing meat processors do not purchase raw in- gredients from Texas because the responding meat processors do not slaughter animals but purchase the meat that they further process. This processed meat is classified in this research under processed food inputs. Bakery, preserved food, and other food processors purchase one-quarter to one-half of their inputs from Texas suppliers. Processed food inputs. Responding dairy, feed, and beverage manufacturers buy approxi- mately 80% of their processed food inputs from Texas suppliers, while meat processors purchase half from Texas sources. Bakeries, preserved food, and other food processors purchase less than one-fourth of their processed food inputs from Texas suppliers. Processed nonfood inputs. Responding bak- eries, preserved food processors, beverage bot- tlers, and meat packers purchase at least half of their processed nonfood inputs from Texas suppliers, while feed, other food, and dairy pro- cessors buy one-quarter to one-third of their processed nonfood inputs from Texas sources. Other materials. Approximately half of the responding dairy, feed, bakery, and preserved food processors’ inputs are provided by Texas suppliers. None of the beverage bottlers’ other materials comes from Texas sources, while one- third of the other food processors’ and one-fifth of the meat processors’ other materials are bought from Texas suppliers. Reasons for Purchasing Inputs In Texas or Outside Texas In the previous section, specific food-pro- cessing industries are identified as having greater opportunities to increase their purchases of Texas products than do others. Table 4 iden- tifies reasons food processors preferred Texas or out-of-state inputs. V Table 3. Average percentage of purchases from Texas suppliers by responding Texas food-processing plants, 1989. Food industry‘ Prsv. Other Materials Meat Dairy Feed Bread food food Beverages _ (Average percentage of purchases in Texas) Flaw ingredients 0.0 100.0 90.7 45.9 30.1 26.2 100.0 Processed food products 54.0 79.6 82.6 8.1 22.6 0.3 83.3 Processed nonfood products 51.5 31.6 25.2 95.8 90.1 29.7 77.8 Other materials 18.9 57.0 54.7 48.5 67.6 31.8 0.0 Source: Computations by the author from survey results. ‘Grain mill and poultry respondents did not provide this information. Table 4. Major reasons tor buying materials as cited by Texas food-processing plants, 1989 (186 responding plants)! Number of Percentage of Number oi plants Percentage of plants Reason times cited total reasons cited citing this reason citing this reason Availability ln-state 406 32.2 140 75.3 Out-of-state 21s 36.0 91 43.9 Cost In-state 365 28.9 129 69.4 Out-of-state 148 24.8 72 38.7 Quality ln-state 1 99 1 5.8 80 43.0 Out-of-state 1 18 1 9.7 57 30.6 Transportation ln-state 202 16.0 80 43.0 Out-of-state 22 3.7 1 1 5.9 Sole Source ln-state 42 3.3 33 1 7.7 Out-of-state 52 8.7 38 20.4 National Contract ln-state 19 1.5 10 5.4 Out-of-state 27 4.5 12 6.5 Made by Affiliate In-state 12 1.0 11 5.9 Out-of-state 8 1 .3 6 3.2 Other In-state 16 1.3 12 _ 6.5 Out-of-state 8 1 .3 6 3.2 Total ln-state 1261 1 00.0 1 86 Out-of-state 598 100.0 186 ‘ Each plant could check each reason for any of its primary six materials, labor, or depreciable fixed assets. Thus, each reason could be checked as many as eight times per plant. The second column represents the total number of times that the reason was checked divided by the total number of reasons checked. This information assists in determining whether an aspect of Texas production that, when promoted, might increase sales of state commodities. Areas in which Texas can improve to be more competitive can be identified as well. However, note that one noninvestigated area is the possibility that some processors may not purchase out-of-state inputs because of state loyalty. In elaborations of these preferences, note that grain mill and poultry plants answered questions in this section but did not answer questions in the previous section on costs. Availability. As a reason for buying their in- puts from Texas suppliers, 140 out of 186 re- sponding plants cited availability (Table 4). Availability was cited by 20% to 75% of the re- sponding plants in each food-processing indus- try with only two exceptions. Only 1 out of the 1O 23 responding meat processors cited availabil- ity of raw materials as a factor in purchasing these inputs from Texas suppliers. Note that few of the responding meat processors slaughter ani- mals, so availability of raw materials is not as relevant for this industry as it is for other indus- tries. In addition, only 2 out of the 19 responding bakeries cited availability of processed nonfood purchases as a factor in purchasing these inputs from Texas suppliers. Many meat, bakery, dairy, and feed processors cited availability as a reason for purchasing processed food inputs. Similarly, several poultry, bakery, dairy, and feed manufac- turers noted availability as a reason for their pur- chases of local raw materials. Several poultry and beverage processors listed availability as a rea- son for purchasing processed nonfood inputs in Texas. Except for poultry, more than half of all the inputs discussed in this section are purchased from Texas sources. Availability of inputs was cited as a reason for buying out-of-state inputs by approximately half of the responding food processors (Table 4). Responding grain mills did not give this reason. Availability was important for processed food bought out-of-state by meat packers, for raw materials purchased out-of-state by preserved food processors, and for processed nonfood products purchased by beverage bottlers. The reasons plants listed for out-of-state purchases can be generalized into three differ- ent cases of input availability. The first case is that of the input in question being available in Texas, regardless of whether it is produced out- of-state. Without considering quality, cost, and other comparisons with out-of-state inputs, Texas input producers just need to make their products available. The second case is that of the input being available in Texas but a proces- sor maintaining an out-of-state supplier. Ex- amples of this case include a citrus processor who maintains an out-of-state source as assur- ance against a freeze and a fish processor who maintains an out-of-state source so he can pro- cess fish during the Texas off-season. These are simply Wise business practices that should con- tinue. The third case is an input not being avail- able in Texas and available only from out-of- state sources. Examples of these inputs include pork, turkey, packaged seasonings, lollipop sup- plies, beverage supplies, processed soybeans (because of the small amount of soybeans pro- duced in Texas), almonds, peppers, and toma- toes, which are not grown in Texas. The demand for these products warrants further investiga- tion to determine whether the Texas market can sustain a supplier and whether the supply of these products, such as soybeans, almonds, tomatoes, and peppers, can be maintained. Cost. Cost was cited by almost 70% of the questionnaire respondents as a reason they buy Texas inputs (Table 4). When cost is cited as a reason for purchase of Texas inputs, quality and transportation can be factors involved in that cost. These factors are considered later in this report. The specific cost-influenced Texas pur- chases by poultry, feed, bakery, and beverage plants should be further investigated so they can be promoted in future efforts to seek plant relo- cations or start-ups in Texas. The food-process- ing industries that indicated cost is a factor in 11 their decision to purchase out-of-state inputs did not list specific Ways in which cost is a fac- tor in their purchase of out-of-state inputs. With only three exceptions, a food processor from each food industry cited cost as a reason for purchase of either raw materials, processed food inputs, or processed nonfood inputs from out- of-state sources. Any of these reasons can be investigated, but the feed, bakery, and other food industries consistently had more than one pro- cessor citing this reason for. each type of input, and these industries could be studied first. Dairy processors cite cost as a reason for the out-of- state purchase of processed food and nonfood inputs, and these reasons could be part of an initial study as well. Quality. Forty-three percent of the respond- ing plants cited quality as a reason for buying Texas inputs (Table 4), which could be influ- enced by loyalty to the state. For out-of-state input quality, feed, bakery, preserved food, and “other food” processors exhibit the main con- cern for quality of each type of input, while meat processors cite concern for the quality of pro- cessed food inputs. These areas Warrant further investigation for Ways Texas can improve its products. Aspects of Texas input quality valued by beverage, meat, dairy, and bakery processors should be pursued for purposes of promoting Texas inputs. Transportation. Transportation availability and cost was also cited by 43% of the respond- ing plants (Table 3). Poultry processors were the only processors who did not cite transportation as a consideration for any input purchase. Texas freight rates appear to be favorable for some items and unfavorable for others. Reasons given by Texas dairy, grain, and feed processors for favoring Texas inputs for transportation reasons deserve further investigation. Few of the food processors have problems with Texas transpor- tation. However, transportation considerations of feed mills warrant further study. Other fac- tors to consider include transportation of pro- cessed food for meat and preserved food pro- cessors, transportation of raw materials for dairy and “other food” processors, and transportation of processed nonfood for “other food” processors. Sole source. Approximately one-fifth of all re- sponding food processors cited sole source of inputs as a reason for buying either Texas or out-of-state inputs (Table 4). The existence of a sole source is favorable when the source is in and unfavorable otherwise. Much like national contracts, there is no benefit to promoting the existence of an affiliate, and affiliate input trans- actions are difficult for Texas entrepreneurs to attract. Similar to national contracts and sole sources, it is advantageous that few plants pur- chase out-of-state inputs made by affiliates. Other reasons. Seven percent of the respond- ing food processors listed other reasons for buy- ing Texas inputs (Table 4). Three comments are listed in the previous discussion. Other com- ments deal mainly with convenience issues. Texas and unfavorable when the source is out- side Texas. The existence of a Texas sole source for an input could be promoted to new food-pro- cessing plants. When the only source of an input is outside Texas, more work must be done to en- tice the source to move to Texas. Fortunately, few plants have sole sources outside Texas. National contract. National contracts for Texas inputs were cited by only 5% of the re- sponding food processors. National contracts restrict input purchases to out-of-state suppli- ers for only 6.5% of the responding food proces- sors (Table 4). National contracts are similar to sole sources in that they are favorable if the na- tional contract is with a Texas supplier and un- favorable if the contract is not with a Texas sup- plier. For example, a beverage plant noted that its decisions were controlled by the corporate office. In situations such as this, there is no ben- efit to be gained from promoting the existence of national contracts for Texas inputs. The ex- istence of national contracts is a difficult situa- tion for Texas input entrepreneurs to overcome. Fortunately, few plants are purchasing inputs out-of-state for this reason. Made by afliliate. Approximately 7% of the food processors purchased Texas inputs that were made by affiliates (Table 4). Only 3% of the responding food-processing plants pur- chased out-of-state inputs because they were made by an affiliate (Table 4). As with sole sources and national contracts, inputs made by affiliates are favorable if the affiliate is in Texas Markets for Texas Food Products The importance of markets lies in the dis- tinction between final demand markets versus processing markets. When products are sold to final demand markets, value can no longer be added to them. When they are sold to process- ing markets, more value is added to the prod- ucts. In this way, more wages and salaries, in- terest, taxes, and the other value-added com- ‘ponents are generated for the economy. The markets of interest are listed in Table 5. According to the input-output definition, final demand markets are export markets and state, local, and federal governments. Most of the markets listed under “other markets” are final demand markets, as well. Only one market, the processing market, performs further processing of processed food products. The remaining mar- kets, namely the retail firms, hotels, restau- rants, institutions, other wholesalers, brokers, Table 5. Distribution of respondents’ sales of Texas food products by market, 1989. Markets Food industry _ Prsv. Other Meat Poultry Dairy Grain Feed Bread food food Beverages (Percentage of sales) Retail firms 35.0 32.9 53.0 5.9 36.7 7.4 34.3 9.0 50.0 State/local govt. 1.5 0.3 1.8 < 0.1 1.2 0.0 0.3 0.1 0.0 Federal govt. 1.5 0.0 1.1 0.4 0.0 0.2 0.4 0.1 < 0.1 Exports 2.2 0.0 1.1 83.6 0.2 3.9 0.1 1.4 0.0 Hotel/restaurant 18.6 6.6 5.1 0.7 0.0 2.3 0.5 2.4 < 0.1 Institutions 3.8 0.4 5.9 1.2 < 0.1 0.9 1.9 0.1 < 0.1 Other wholesalers 21.5 2.2 0.2 6.9 7.2 39.0 34.2 21.6 < 0.1 Brokers 0.0 < 0.1 0.0 0.2 < 0.1 5.3 5.9 0.4 < 0.1 Processors 9.7 5.7 7.9 0.9 0.4 0.0 7.0 19.6 0.0 Co. distribution facility l 5.8 48.9 23.2 0.1 9.0 1.9 13.7 28.6 49.9 Other markets 0.4 3.0 0.7 0.1 45.2 39.1 1.7 13.4 0.1 All markets 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Source: Computations by the author from survey results. 12 Q, and company distribution facilities, add value in terms of time and place to the products. The grain mill sector is the only sector for which most of the production goes mainly to fi- nal demand (84.0%). The meat, dairy, and bak- ery industries sell roughly 5% of their produc- tion t0 final demand markets. The remaining sectors, however, sell 2% or less of their pro- duction to final demand markets. Therefore, given equal capabilities of food-processing in- dustries to adapt to the community, food-pro- cessing industries other than the grain mill in- dustry should be chosen for investment. Further processing of Texas food products oc- curs for less than 20% of the output in each in- dustry. Almost 20% of the “other food” products are further processed. None of the bakery or bev- erage products is further processed. However, bak- ery processors use grain mill products as inputs, so their production generates further economic activity within the food-processing sector. Less than 10% of the remaining industry products are further processed. Given these circumstances, the “other food” products would be favored over other food-processing industries because of the in- creased employment and income from further pro- cessing of the “other food” products. Another important aspect of markets is how much production is sold to out-of-state process- ing firms. Table 6 shows the converse of this situation, namely, how much of the production is sold to Texas buyers. The value of this knowl- edge relates to the desire for further value to be added to commodities in Texas. For example, all the animal feed sold to processors is sold to Texas processors, but for obvious reasons, very little animal feed is sold for further processing. Rela- tive to other food industries, slightly more of the grain and preserved foods are sold to further pro- cessors, and most of this production is sold to Texas processors as well. Between 5 and 20% of the remaining products is sold to processors, and between 70 and 99% of this production is sold to out-of-state processors. These figures indicate an opportunity exists to further process the produc- tion of these latter industries, namely the meat, poultry, dairy, and “other food” industries. Fur- ther study would determine what expected dollar value of commodities is under consideration and the type of processing required. Contrary to the previously stated desire to increase the amount of production that is further processed is a desire to sell more production to final demand markets out-of-state. Because of the small quantity of production sold to final de- mand markets and the large percentage of sales of these sectors to Texas firms, most of the final demand sales are sold to Texas buyers. Industry-by-Industry Evaluation These findings from the survey are now evaluated on an industry-by-industry basis. This summary will help determine the plausibility of investing in industries identified in the next sec- tion as providing promising income or employ- ment opportunities in Texas. Table 6. Percentage of respondents’ sales of Texas food products to Texas destinations, 1989. Food industry Prsv. Other v Markets Meat Poultry Dairy Grain Feed Bread food food Beverages (Percentage of sales) Retail firms 54.6 100.0 97.5 50.8 85.5 82.1 60.6 26.5 15.1 State/local govt. 24.5 100.0 _ 100.0 38.0 81.4 0.0 40.5 83.5 0.0 Federal govt. 19.4 0.0 100.0 0.0 0.0 100.0 46.0 6.6 100.0 Exports 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Hotel/restaurant 61 .3 100.0 99.7 42.9 0.0 93.6 90.5 9.3 100.0 Institutions , 34.1 100.0 95.6 83.2 100.0 100.0 70.7 47.2 80.2 Other wholesalers ' 63.1 50.1 80.0 4.5 63.3 60.6 62.4 7.4 51.2 Brokers 0.0 0.0 0.0 20.0 100.0 20.0 10.0 52.0 100.0 Processors 20.7 1.0 6.9 80.0 100.0 0.0 61.6 30.5 0.0 Co. distribution facility 100.0 0.0 59.5 75.0 99.4 100.0 47.4 75.3 0.1 Other markets 90.6 70.0 100.0 92.5 ' 96.8 30.5 61.7 1.6 100.0 Source: Calculations by the author from survey results. 13 Meat Products The respondents were not representative of animal slaughterers, so these meat processors could only be surveyed to obtain further infor- mation about the raw material preferences 0f meat processors. The meat processors who did respond to the questionnaire obtained meat from slaughterers and further processed and/or packaged it. From the cost information, another aspect appearing worthy of investigation is the reason meat processors buy other materials out- of-state. The major reasons for meat processors’ pur- chases of processed food inputs in and out of Texas were availability of these inputs. Proces- sors noted limited availability of pork, turkey, and packaged seasonings, indicating that fur- ther study of these markets in Texas might be worthwhile. Quality of Texas food and nonfood inputs, as well as non-Texas food inputs, are also important for Texas meat processors. Less than 5% of Texas meat processors’ sales went to final demand markets. Ten percent went to processors, 80% of which were located out- of-state. Small Texas processors might consider investigating what type of processing occurs out- of-state and determining whether enough pro- cessing exists to support a plant. The remain- ing 85% of the sales went to markets that add value to meat in terms of time and place, such as retail firms, restaurants, and wholesalers. Poultry Products Availability and cost of Texas raw materials, as well as the availability of processed Texas nonfood inputs, were important to two of the four responding Texas poultry processors. Two responding processors purchased out-of-state processed nonfood inputs because of cost. These purchases could be further investigated to see whether Texas input suppliers can become more competitive by adjusting some aspect of production. Less than 1% of the reported Texas poultry production was sold to final demand markets. Eight percent was further processed, almost solely by non-Texas processors. Given the mag- nitude of sales by poultry plants, processors in Texas may have an opportunity to capture some of this market for further processing. This pos- sibility merits more investigation. More than 90% of the sales went to firms that add time l4 and location value to the processed poultry, pri- marily by retail firms and their own company's distribution facilities. Dairy Products Availability and transportation of Texas raw materials were important reasons for the re- sponding dairy plants’ purchases of Texas raw materials, possibly because of the bulk and per- ishability of raw milk. This hypothesis, however, needs further investigation. Availability and cost of Texas processed food and nonfood inputs were also important considerations for their in-state purchases. Five percent of the responding dairy proces- sors’ sales were made to final demand sectors. Eight percent were made to processors, 7% of whom were located in Texas. This further pro- cessing merits more investigation, as dairy sales are fairly large and a large percentage of the value of processed dairy products is being shipped out-of-state. Approximately 85% of dairy products were shipped to firms that add location and time value, especially retail firms and company distribution facilities. Grain Mill Products Two out of the five responding grain mill pro- cessors bought Texas raw materials and pro- cessed nonfood materials for reasons of avail- ability and transportation. Eighty-four percent of sales by responding grain mill processors went to final demand markets, the highest percent- age of sales going to “final demand of any food industry. One percent went into further process- ing, 80% of which is processed in Texas. Fifteen percent went to markets, such as retail firms and other wholesalers, that added time and lo- cation value. Animal Feed Products Responding animal feed processors indicated that availability, cost, and transportation were reasons they purchased raw materials and pro- cessed nonfood inputs from Texas suppliers. One manufacturer noted the convenience of buying Texas corn, oats, and milo. Forty-seven percent of animal feed sales went to final de- mand markets, and less than 1% went to pro- cessing. The remaining 52% went to markets that added time and location value to feed, such as retail firms. Bakery Products Responding Texas bakeries cited availabil- ity, cost, and quality most often as reasons for purchasing Texas processed food products. One baker purchased out-of-state inputs only if the input could not be acquired in Texas. Four per- cent of bakery sales went to final demand mar- kets. None went to further processors, and the rest went to firms that added time and location value to the baked goods, such as retail firms and other wholesalers. Note that bakeries pur- chased grain mill products, so increased num- bers of bakery plants increased the economic activity of the food-processing sector through grain mill product purchases. Preserved Foods Responding preserved food processors in Texas preferred Texas raw materials mainly be- cause of their availability and cost. They also bought out-of-state raw materials because of avail- ability. Cost information supports the importance of this information. This situation may be due to reasons that are difficult to remedy, such as poor Texas crops (citrus) and off-production seasons (fish). Less than 1% of production from respond- ing preserved food plants’ was sold to final de- mand markets. Seven percent was further pro- cessed, mainly by Texas processors. The remain- ing 82% of preserved food production was sold to firms that add time and location value to the prod- ucts, such as retail firms, other wholesalers, and company distribution firms. “Other Food” Products Responding processors in the “other food” category purchased all Texas inputs according to availability and cost. However, reasons were given for buying inputs that could not be ob- tained in Texas, and awareness of these reasons might be useful for a Texas input entrepreneur starting up a business. Approximately 2% of “other food” products were sold to final demand markets. 'I‘wenty percent were further processed, 30% of which were further processed in Texas. The rest went to markets that added value in terms of time and place, such as other whole- salers and company distribution facilities. Beverages Responding beverage processors mainly pur- chased Texas processed nonfood inputs because of their availability, cost, and quality. Processed food inputs were purchased from Texas and out-of-state firms because of availability. Spe- cific inputs were listed for entrepreneurs wish- ing to undertake further investigation. Less than 1% of beverages went to final demand and none were further processed. They were sold almost exclusively to retail firms and company distri- bution facilities. INPUT-OUTPUT ANALYSIS Because one of the objectives of this study is to evaluate output, income, and employment implications of increased processing of agricul- tural commodities, the use of an input-output model is called for. Output, income, and em- ployment impacts arising from changes in each sector of the economy are readily calculated from this model. For this study, the 1979 Texas In- put-Output Model was updated to 1986 to make it suitable for answering current questions. To make the 184 sectors manageable, the Texas Input-Output Model was aggregated into 47 sec- tors (listed in Table 7). The model was updated using the technique developed by McMenamin and Haring (1974). The data sources used in the updating process are available from Wildenthal (1992). Fortran programs developed by Ozuna ( 1986) were used to update the model. The complete Texas Input-Output Model can be found in Wildenthal (1992). Transaction Table The sector production disposition is shown in each sector's row of the transactions table, which is presented in Wildenthal (1992). Accord- ing to this information, the two major purchas- ers of agricultural products are the export and domestic food-processing sectors. Exports con- stitute an important market for irrigated cotton (77% of production), dryland cotton (84%), irri- gated feed grains (51%), dryland feed grains (55%), and feedlot production (56%). Domestic 7 processing is a large market for dryland (51%) 15 and irrigated (35%) cotton (cotton ginning), for- estry (80% to lumber and printing), fisheries (55% to preserved food), range production (60% to meat processing), feedlot production (34% to meat processing), dairy production (78% to dairy processing), and poultry and egg production (5% to poultry processing). ‘ Except for the animal feed sector, the food- processing sectors sell little of their output to the agricultural production sectors. The major Table 7. Sectors of the updated 1986 Texas Input-Output Model. Irrigated cotton Irrigated food grains irrigated feed grains Other irrigated crops Dryland cotton Dryland food grains Dryland feed grains Dryland crops and livestock, not elsewhere classified Range livestock production Feedlot livestock production Dairy production Poultry and egg production Cotton ginning Agricultural services Forestry Fisheries Mining Construction Meat products Poultry products Dairy products Grain mill products Animal feeds Bakery products Canned, preserved, pickled, dried, and frozen food Other food and kindred products Beverages Textile and apparel Lumber, paper, printing, and publishing Chemicals and petroleum refining Rubber, leather, plastics, glass, stone, and clay products Metal industries and products Machinery, transportation equipment, and instruments Miscellaneous manufacturing Transportation and communications Utilities Wholesale trade Retail trade - Finance, insurance, and real estate (F.l.R.E) Services Households Property payments Federal government State government Local government Capital formation Exports OIOQNOIUI-POJIO-i purchasers of agricultural processed products are the household and export sectors. The only agricultural-processing sectors that do not sell more than half of their products to households are the grain mill, animal feed, and preserved 16 food sectors. Exports are the major market for grain mill products (74%), while feed and pre- served food each receive 46% of their revenue from exports. Eight percent of meat products, 6% of grain mill products, and 27% of “other food” products receive further processing from food-processing sectors. Animal feed receives 15% of its revenue from the range production sector and 19% from the poultry and egg pro- duction sector. The preserved food sector is unique in having sales of 9% each to the retail: and services sectors. This situation differs greatly from the one outlined in the discussion of the survey of the Texas food-processing industry. This difference is partly due to the structure of the questions addressed in that section. For example, the only means of recording sales to households is to record them as sales to “other markets.” “Other markets” do not receive enough sales to account for the high percentage of sales to households revealed in the transactions table. Sales to com- pany distribution centers could account for part of this if the distribution center is considered to be part of the company and it sells directly to consumers. As for export sales, the grain mill sector results are similar for both the survey and the model because responding grain mills sold 84% of their output to export markets. The exports of feed and preserved food products greatly differ because responding plants sold less than 1% to export markets. For sales to processing sectors, the meat and “other food” sales are similar, butgrain sales of the respond- ing plants are less than 1% of all sales. Respond- ing preserved food processors sell 34% of their sales to retail firms, as opposed to the model’s estimated 9% of sales to retail. The Texas Input-Output Model results indi- cate that the main inputs in the production of agricultural products are provided by the final purchases and nonfood manufactured products sectors. All agricultural production sectors ex- cept the feedlot and poultry and egg production sectors buy at least half of the value of their inputs from final purchase sectors. Most agri- cultural production sectors purchase inputs from several nonfood manufacturing sectors. The three main sources of agricultural pro- cessing inputs are the agricultural product, the household, and the import sectors. The meat- processing sector receives 44% of the value of its inputs from agricultural production sectors, ‘W \ while the poultry-processing sector receives 50% of the value of its inputs from the poultry and egg sector. The bakery products sector receives 30% of the value of its inputs from the house- hold sector. Imports make up 16 to 46% of the value of the inputs purchased by the beverage, grain mill, preserved food, animal feed, and “other food” sectors. The food-processing sec- tors may purchase some inputs from most sec- tors. With this much interdependence among sectors, an increase in the amount of food pro- cessed is likely to have a positive impact on most other sectors of the economy. Ranking Sectors by Value Added Figures for value added by sector indicate the income earned by resources engaged in each activity. As was explained in the input-output analysis discussion, the agricultural and food- processing industries are highly disaggregated, while the other industries are highly aggregated, so it is not surprising to find agricultural pro- duction and processing industries ranked in the last 20 positions of the value-added table found in Wildenthal (1992). As a result of the disag- gregation, the agricultural sectors have smaller total purchases relative to the aggregated indus- tries and have correspondingly smaller value added. Seven of the food-processing industries are ranked individually between 17 and 26 in contribution to the Gross Texas Product (GTP). “Other food” and kindred products and bever- ages constitute slightly less than 0.50% of GTP. Meat products make up 0.27%, while bakery, dairy, preserved food, and grain mill products each contribute between 0.1 1 and 0.16% to GTP. Animal feed is ranked 32 and poultry products are ranked 36, each contributing approximately 0.05% to GTP. All food-processing activities com- bined contributed 1.8% to GTP, which would place the food-processing industry at twelfth place in GTP. The actual impact of food-process- ing is larger, however, due to the interdepen- dence of agricultural processing with the other economic sectors. Agricultural production has a combined con- tribution of 1.64% of GTP, which would rank thirteenth for contributions to GTP. As was ex- plained in the discussion of the transaction table, final purchase sectors provide more than half of the value of inputs for most agricultural "\ production sectors. 17 Intersector Purchases Intersector purchases are purchases by one sector of products from one or more other sec- tors. The more interconnected the economy, the greater the impact of a plant on GTP. Food-pro- cessing industries fall into the middle range, rank- ing 13 through 23. Meat packers purchase almost 2% of output of Texas sectors, while beverage manufacturers purchase 1% from these sectors. Total intersector purchases for the food-process- ing industry are 4.8% of those for the whole Texas economy. One would expect that agricultural pro- cessing would have a high ranking because agri- culture is one of the primary industries in Texas and sufficient time has elapsed for input indus- tries to develop (Mustafa 1971). While discussing the survey of Texas food processors, ways were identified that suppliers of input for food processors could make their products more attractive to responding food pro- cessors. Further investigation into the reasons food processors preferred to purchase inputs in- or out-of-state indicated opportunities for more Texas processed food input purchases by meat, bakery, preserved food, “other food,” and bev- erage manufacturers. Processed nonfood input purchases by meat, dairy, grain mill, feed, “other food,” and beverage manufacturers appear promising for such expansion as well. The small percentage of intersector sales by Texas food processors further supports the potential gains from following through on this investigation. Intersector Sales The intersector sales vary widely, from 18.48% for fire, insurance, and real estate to 0.02% for dryland cotton. This variation is due to the different stages of production of different outputs — some outputs are ready to be sold to final demand sectors, while others require fur- ther processing before final sale. In addition, many sectors sell much of their output outside of the state for further processing. The food-processing industry sells 1.4% of its production to Texas markets not classified as final demand. Texas sales of the food indus- tries vary greatly as well, but each industry con- tributes less than 1% to the sales between Texas sectors. This makes sense because the food-pro- cessing sectors are less likely to have outputs that require further processing than are sectors such as those producing raw agricultural prod- ucts. The Texas food-processing survey indi- cated that food processors sold 5% or less of their output to final demand, except for grain milling, which sold 84% of its output to final demand. Little of the remaining output went to further processors. Most of the remaining out- put went to firms, such as distributors, retail firms, and restaurants, that added value to the output by moving it around. Multipliers Multipliers are the traditional values used in predicting the effects of planned or induced changes in some sector of the economy. The val- ues calculated for Texas in 1986 are used here. Note that all calculations assume increasing de- mand for the product. The real usefulness of these figures will be known once the sustain- able demand for these products is known. In addition, the following results can also be used to estimate the consequences if a plant leaves the area (decreasing demand). Output Multipliers Output multipliers for the 1986 Texas In- put-Output Model are shown in Table 8. These figures represent the total change in output of all sectors in the Texas economy generated by a particular sector’s delivery of an additional $1 worth of output to final demand. The ranking of sector impacts on the economy changes dramati- cally when the indirect effects are considered. Poultry products now rank first, with a $1 in- crease in poultry demand, resulting in a $2.34 increase in Texas output. Eight of the 9 food- processing sectors are ranked in the top 21 out- put-inducing sectors. Meat products are ranked fourth with a multiplier of 2.16, and dairy prod- ucts are next with a multiplier of 2.01. Preserved food has a multiplier of 1.93 and is ranked elev- enth. Beverages are ranked seventeenth with a multiplier of 1.85, and animal feed products are next with a multiplier of 1.80. Grain products are ranked twenty-first with a multiplier of 1 .73, and “other food” products is last with a multi- plier of 1.25. In the discussion of the Texas food-process- ing survey, Texas food processors were shown to sell most of their output to buyers, which added location value to the products, while in the fifth section, Texas food processors were shown to sell most of their output to households 18 Table 8. Total dollar change In output per $1 direct change in ~ final demand by sectors of the Texas economy, 1986. Sector Sector Output number name multipliers Flank (Dollars) 20 Poultry products 2.34 1 12 Poultry and eggs 2.21 2 10 Feedlot livestock production 2.16 3 19 Meat products 2.16 4 21 Dairy products 2.01 5 2 irrigated food grains 1.98 6 3 Irrigated feed grains 1.96 7 25 Preserved food 1 .93 11 11 Dairy 1.91 12 6 Dryland food grains 1.87 13 1 Irrigated cotton 1.86 14 8 Dryland crops and livestock, NEC 1.86 15 27 Beverages 1 .85 17 23 Animal feeds 1 .80 18 4 Other irrigated crops 1.79 19 24 Bakery products 1.77 20 22 Grain mill products 1.73 21 13 Cotton ginning 1.73 22 7 Dryland feed grains 1.73 24 16 Fisheries 1.71 25 5 Dryland cotton 1.67 27 9 Range livestock production 1.59 31 39 F.I.H.E. 1.50 36 26 Other food and kindred products 1.25 4O and exports (final demand). The difference could be due to response error in the survey or to the market choices presented to the food processors in the survey. In either case, potential output might be greater if products were further pro- cessed instead of merely transported or con- sumed, but no indication of such a possibility exists for this output. Income Multipliers The direct income effect is the amount of each dollar increase in the output of each sec- tor that goes to households in the form of wages and salaries, proprietor income, rents, or prof- its. The total effect is the total change in income as a result of a $1 change in final demand. If the desirable policy objective is to increase in- come in an economy by increasing output, the sectors with a high total income effect should be targeted (Table 9, column 4). Efforts would focus on bakery products, and the highest total income effect would be for food- processing sectors, if this were the goal. \ Table 9. Total dollar change in income per $1 direct change in income by sectors of the Texas economy, 1986. \ Sector Sector Direct Total Income number name effect effect multipliers Flank ---------------------------- -- (Dollars) -------------------------------- 10 Feedlot livestock production 0.04 0.50 11.55 1 19 Meat products 0.11 0.51 4.61 2 12 Poultry and eggs 0.10 0.42 4.06 4 25 Preserved food 0.12 0.40 3.39 5 20 Poultry products 0.15 0.44 2.92 7 27 Beverages 0.13 0.37 2.88 8 21 Dairy products 0.19 0.50 2.66 10 22 Grain mill products 0.16 0.39 2.45 11 23 Animal feed 0.17 0.41 2.45 12 2 irrigated food grains 0.25 0.50 2.01 14 11 Dairy 0.32 0.58 1.85 15 3 Irrigated feed grains 0.28 0.52 1.84 17 26 Other food and kindred products 0.10 0.17 1.77 18 6 Dryland food grains 0.30 0.52 1.76 19 24 Bakery products 0.30 0.53 1.74 20 8 Dryland crops and livestock, NEC 0.34 0.59 1.73 21 4 Other irrigated crops 0.32 0.55 1.70 22 1 Irrigated cotton 0.35 0.59 1.66 26 16 Fisheries 0.35 1.56 1.63 27 13 Cotton ginning 0.37 0.60 1.61 28 7 Dryland feed grains 0.35 0.54 1.54 29 39 F.I.R.E. 0.40 0.59 1.48 33 5 Dryland cotton 0.48 0.67 1.40 37 9 Range livestock production 0.49 0.67 1.38 38 Income multipliers represent the total change in income in the economy caused by a $1 change in income for that sector. As seen in Table 9, when indirect effects are taken into ac- count, all agricultural processing sectors rank in the top 20 income-generating activities in the Texas economy (Table 9, column 5). Income mul- tipliers from all sectors range in value from 11.55 for feedlot livestock production to 1.38 for range livestock production. Meat products rank a dis- tant second with 4.61, which is more in line with the rest of the values than that of feedlot live- stock production. Preserved food has an income multiplier of 3.39, while animal feeds, grain mill products, dairy products, beverages, and poul- try products multipliers range between 2.45 and 2.92. Bakery products and “other food” multi- pliers are the lowest with 1.74 and 1.77. Wages and salaries make up much of the costs of a sector with high-income multipliers. If the only policy aim is to increase income in the economy through an initial increase in income in a particular sector, the sectors with the highest income multipliers should be targeted (Table 9, 19 column 5). Meat, with the highest income multi- plier for food-processing sectors, would be the fo- cus of attention in this situation. Note that meat also has the second highest total income effect for food-processing industries and should be tar- geted under either objective. Employment Multipliers Direct employment effects indicate the num- ber of people employed per year per unit of out- put. Total employment effects are computed by considering the repercussions on employment in all sectors as a result of the initial change in I final demand in a sector. If the policy goal is to increase employment solely through the expan- sion of output, the industry with the largest to- tal employment effect receives attention. In terms of food-processing sectors, the target would be the poultry products sector. Employment multipliers measure the total change in employment resulting from a $1 change in employment in a sector. Employment multipliers are calculated by dividing total em- ployment effects by direct employment effects. As shown in Table 10, column 5, employment multipliers for Texas sectors range from 6.5 for dairy to 1.22 for other irrigated crops and ap- parel. Agricultural processing sectors rank from 11 to 33, with a range of values of 3.58 for dairy products to 1.43 for “other food.” Beverages rank 13 with a multiplier 0f 3.12, while meat products are next with 3.08. Grain mill prod- ucts have a multiplier of 2.77, and both pre- served food and animal feed have a multiplier of 2.57. Poultry products have a multiplier of 1.90 and bakery products have 1.87. As with income effects, the policy objective is, critical in the interpretation of the employ- ment effects table. If the objective is to increase employment in the economy by an initial in- crease in employment in a particular sector, the industry with the largest employment multiplier should be sought. The food-processing indus- try meeting this criterion is the dairy products sector. It should be noted, however, that em- ployment multipliers do not account for possible underemployed resources and excess capacity, l r which would dampen the effects projected here ' (Jones and Mustafa 1972). Projected Effects of a $1 Million Change in Final Demand Although it is important to look at the ef- fects of a $1 change in final demand on the eco- nomic aspects of the sectors, a state's basis for policy action may depend on which sector has the greatest total impact on state income, out- put, or employment. In Table 11, the effects of a $1 million change in final demand for each sector’s output are examined. The output col- umn contains the same column of figures as found in the third column of Table 8. The in- come figures are the direct income effects found in the fourth column of Table 9, while the em- ployment figures are the direct employment ef- fects found in the fourth column of Table 10. The industry receiving attention depends on the objective of the policy maker. If food-pro- Table 10. Total change in person-years of employment per one person-year direct change in employment by sectors of the Texas economy, 1986. Sector Sector Direct Total Employment number name effect effect multipliers Rank ----------------------- -- (Person-years) ---------------------------- 11 Dairy 1.82 11.81 6.50 2 9 Range livestock production 1.35 7.54 5.58 4 12 Poultry and eggs 2.73 15.33 5.57 5 6 Dryland food grains 1.80 9.60 5.35 6 5 Dryland cotton 1.80 8.86 4.92 7 1 Irrigated cotton 2.53 10.81 4.27 8 7 Dryland feed grains 2.26 8.83 3.92 9 21 Dairy products 3.07 10.99 3.58 11 10 Feedlot livestock production 2.90 9.86 3.39 12 27 Beverages 4.06 12.67 3.12 13 19 Meat products 3.33 10.27 3.08 14 2 Irrigated food grains 4.09 14.48 3.05 15 16 Fisheries 3.07 9.15 2.98 16 22 Grain mill products 3.52 9.75 2.77 17 3 Irrigated feed grains 4.56 11.97 2.62 18 25 Preserved food 6.82 1 7.56 2.57 1 9 23 Animal feeds 4.55 11 .70 2.57 20 ' 39 F.I.Fi.E. 5.66 10.78 1.90 22 20 Poultry products 10.94 20.79 1.90 23 24 Bakery products 9.13 17.07 1.87 25 8 Dryland crops and livestock, NEC 18.98 27.59 1.45 32 26 Other food and kindred products 5.69 8.16 1.43 33 13 Cotton ginning 18.94 24.91 1.32 35 4 Other irrigated crops 34.41 41.34 1.22 37 2O \ Table 11 . Effects of $1 million change in final demand on output, income, and employment for each sector of the Texas economy, 1986. Sector Total change in: number Name Output Income Employment (Million dollars) (Million dollars) (Person-years) 1 irrigated cotton 1.86 0.59 10.81 2 irrigated food grains 1.98 0.50 12.48 3 irrigated feed grains 1.96 0.52 11.97 4 Other irrigated crops 1.79 0.55 41.84 5 Dryland cotton 1.67 0.67 8.87 6 Dryland food grains 1.87 0.52 9.59 7 Dryland feed grains 1.73 0.54 8.83 8 Dryland crops and livestock 1.86 0.59 27.60 9 Range livestock production 1.59 0.67 7.54 10 Feedlot livestock production 2.16 0.50 9.86 11 Dairy 1.91 0.58 11.81 12 Poultry and eggs 2.21 0.42 15.33 13 Cotton ginning 1.73 0.60 24.92 16 Fisheries 1.71 0.56 9.15 19 Meat products 2.16 0.51 10.27 20 Poultry products 2.34 0.44 20.79 21 Dairy products 2.01 0.50 10.98 22 Grain mill products 1.73 0.39 9.75 23 Animal feeds 1.80 0.41 11.70 24 Bakery products 1.77 0.53 17.07 25 Preserved food 1 .93 0.40 17.56 26 Other food and kindred products 1.25 0.17 8.16 27 Beverages 1 .85 0.37 12.67 39 F.I.R.E. 1.50 0.59 10.78 cessing industries are to be targeted and the only goal is to expand total output or employment, the poultry industry is the sector to investigate. If the sole goal is to expand income, the bakery industry should be the focus of the analysis. Further analysis into the demand for these prod- ucts will provide insight into the potential to achieve the values of these coefficients. CONCLUDING COMMENTS Summary State governments are understandably in- terested in strategies to strengthen their econo- mies in terms of output, income, and employ- ment. In agricultural states, one tactic is to add value to raw agricultural commodities within the state instead of sending commodities outside the state for processing. Literature describing research on adding value to commodities covers topics such as ex- panding or developing markets for value-added products, exploring appropriate roles for the 21 government to fund research in value-added activities, evaluating economic growth potential through value-added opportunities in rural economies, observing potential inefficiencies in value-added activity location, and exploring trends in the food-processing industry. In the analysis reported in this bulletin, quantitative analysis of impacts on output, income, and employment from changes in the final demand for processed food products is combined with information about the state's food-processing environment. The framework for this research involves investigation of the major forces affect- ing the food-processing industry in general and the Texas food-processing industry in specific and provides a brief description of export and location focuses of research on value-added ac- tivities. Quantitative results are then interpreted in light of food-processing and location issues. To further investigate means of sustaining positive trends in output, income, and employ- ment, especially as reported in the previous sta- tistics, a survey was conducted to determine the cost components of Texas food processors, the percentage of materials they purchase in Texas, and their reasons for these purchase decisions. For this investigation, there were not enough responses about the cost components of vari- ous food-processing sectors to make any gener- alizations about which types of inputs are most important to their businesses. However, the plants did report whether they buy inputs from in- or out-of-state sources and why they buy inputs where they do. Food-processing indus- tries that purchase large percentages of their raw materials from out-of-state sources include those classified as preserved food, “other food,” and dairy plants. Preserved food, “other food,” meat, and bakery plants purchase a large per- centage of their processed food inputs from out- of-state sources. Purchasers of large percent- ages of processed nonfood inputs are dairies, bakeries, feed mills, and “other food” processors. Other materials are purchased out-of-state in large percentages by meat packers, beverage bottlers, and “other food” processors. Overall, almost three-fourths of Texas pro- cessors purchase Texas commodities according to availability and cost. The influence of the at- titude of buying strictly within Texas was not investigated. Some processors purchase some of their inputs from an out-of-state source for assurance that they have a source should a problem develop with the Texas input supply, such as a Texas citrus freeze. Responding plants indicated that certain inputs, such as turkey, pork, packaged seasonings, beverage supplies, processed soybeans, almonds, peppers, and to- matoes, are available only from out-of-state sup- pliers. The ability of Texas to support a sup- plier of these inputs or, in the case of almonds, peppers, and tomatoes, to raise some of these commodities merits further investigation. Poul- try processors, feed millers, bakeries, dairies, and beverage bottlers are the major food pro- cessors that prefer Texas inputs because of cost. Input quality and transportation factors were each mentioned by almost half of the plant repre- sentatives. Feed millers, bakeries, meat packers, and preserved and “other food” processors prefer Texas inputs because of input quality. Transpor- tation is a factor in the preference of dairy, grain, and feed processors for purchase of Texas inputs. Less than 20% of the respondents indicated that they had a sole source for an input, purchased inputs under national contracts, or purchased inputs made by an affiliate. 22 Availability and cost were also the most fre-wi quently cited factors for out-of-state input pur- chases. Feed millers, bakeries, and “other food” processors are the main food processors who prefer out-of-state inputs because of cost. Qual- ity was cited by almost one-third of the respon- dents, and beverage bottlers, meat packers, dair- ies, and bakeries were the major food proces- sors who prefer out-of-state inputs. Twenty per- cent of the respondents had a sole out-of-state source. Less than 20% of the responding plants mentioned the purchase of inputs under na- tional contracts or of inputs made by affiliates as reasons for out-of-state purchases. Input preferences of food processors were studied to seek increased opportunities of mar- keting Texas agricultural products by empha- sizing characteristics that processors like about Texas commodities or by discovering problems with Texas commodities and / or advantages of out-of-state commodities. Further investigation into reasons Texas inputs are preferred may reveal specific aspects of inputs that might be promoted to prospective Texas food processors. Reasons for preferring out-of-state inputs may be further investigated to determine whether aspects of Texas inputs can be improved to en- hance the competitiveness of Texas input sup- pliers. Increased purchases of Texas inputs would provide one means of increasing Texas output, income, and employment. Quantitative methods of predicting increases in output, income, and employment caused by changes in the amount of input use would en- able researchers to prioritize which opportuni- ties they pursue. The McMenamin-Haring (1974) procedure was used to update the Texas Input- Output Model transaction table into a 1986 model that emphasized agricultural production and processing. The transaction table was ma- nipulated to give the direct and total require- ments tables. The transactions table and the direct requirements table show that the agri- cultural processing sectors. sell outputs prima- rily to households and export markets, while inputs primarily come from final demand and nonfood manufacturing sectors. Agricultural production sectors sell output primarily to ex- port and domestic food-processing sectors and purchase inputs primarily from the agricultural production, household, and imports sectors. The model is not set up to detect whether exports“ are delivered to final demand or processing sec- . .....;.....¢...._.._. ._.....~....,....i..».. '\ tors, so the destination of exports merits fur- ther investigation. If Texas products are exported to processors, more investigation on whether the products could be further processed in Texas is warranted. As for imported inputs, the ability of Texas manufacturers to supply these inputs should be further investigated as well. The total requirements table is the basis of multiplier analysis, which is used to predict the output, income, and employment impacts of an increase in final demand for processed agricul- tural products. Multipliers represent the total change in output, income, or employment in the economy resulting from a $1 change in final demand for the products of the sector repre- sented by the multiplier. After briefly summa- rizing the impacts of each industry, the food- processing industry that ranked consistently high by output, income, and employment mul- tipliers will be examined. Industry-by-Industry Evaluation This section focuses on applications of the Texas Input-Output Model, which estimates the impact of changes in final demand on aspects of the Texas economy. These results will be evaluated here on an industry-by-industry ba- sis. An important caveat for all results is that the interpretations assume that final demand will sustain the increased employment of pro- duction resources. The capabilities of the plants to establish, given the industry structure, is also assumed. Meat Products The meat-processing sector has great poten- tial if the policy goal is to increase the output of the Texas economy. On the basis of multipliers, meat processing was ranked fourth of all sec- tors and second of the food-processing sectors for its impact on output. Meat processing is also an important sector to consider if the aim is to increase income through an initial increase in income as opposed to generating an increase in income through output increases. The meat sec- tor ranks first in total income multipliers and second in all industries, so it could also be con- sidered in attempts to increase income through output increases. The potential success of the meat industry is not clear, however. In the second section, the \ demand factors that were considered were not generally favorable for the meat industry. Spe- 23 cifically, the preferences of older people, those in small households, and those interested in health issues negatively impact the demand for meat products. Recently, however, some of the hog industry has relocated to the Panhandle to capture economies of size and scope and to avoid diseases (Mazzocco 1991). The real estate in the Panhandle is relatively inexpensive. It remains to be seen how the value-added meat industry performs in Texas. Further research also is needed to explore Texas’ comparative advantage in meat production. Poultry Products The poultry sector has the highest output multiplier of all industries. When food indus- tries are considered as a means of increasing employment through output increases, this in- dustry is also noteworthy. When the size of the sector’s final demand is considered, however, this sector ranks twenty-second overall and sev- enth among food-processing industries. For in- come multipliers, however, it ranks third among food-processing industries. The potential for success of this industry in achieving its goal is quite high. The demand fac- tors in the second section were favorable to the further growth of the poultry-processing indus- try. Specific relevant demand factors include both the preferences of older people and those concerned with their health, as well as trends in consumption patterns. Dairy Products The dairy-processing sector is a prime con- sideration when the objective is to evoke an in- crease in employment through a food-process- ing industry without concurrently increasing output. Dairy processing has the highest em- ployment multiplier of any food-processing sec- tor. Among food-processing sectors, it has the third highest output multiplier and the fifth highest income multiplier. The demand factors discussed in the litera- ture review do not provide a clear idea of the potential success in this industry. Locational factors, namely the perishability of dairy prod- ucts, ensure that a base level of production will exist in Texas. Growth in population centers will increase final demand, which will in turn have output, income, and employment effects, which were estimated with multipliers. Grain Mill Products The grain mill sector does not stand out among the areas considered as an area of im- pact on the Texas economy. It is ranked fourth among food-processing industries for employ- ment multipliers, sixth in income multipliers, and eighth in output multipliers. This lacklus- ter is partly attributable to the limited interac- tion this sector has with other sectors of the Texas economy. Most of its output goes directly to final demand, specifically exports. Animal Feed Products A As with the grain mill industry, the animal feed industry does not appear to offer much potential for bolstering the Texas economy. Among food-processing industries, it ranks sixth in output and employment multipliers and sev- enth in income multipliers. This sector interacts with other Texas sec- tors to a greater extent than the grain mill sec- tor does. Most of its output is sold to firms, add- ing location value to the products. In addition, most of its inputs are purchased from Texas suppliers. An increase in final demand from the ex- panding livestock sector could cause an increase in the amount of grain mill products demanded. The grain mill industry is the smallest food-pro- cessing sector considered. Bakery Products The bakery industry should be considered when the policy objective is to increase income through an “increase in output of a food-process- ing sector. It has the highest total income effect of any food-processing industry. In addition, the de- mand factors discussed in the literature review are favorable for this industry. Otherwise, how- ever, it is not a strong performer. It ranks seventh among food-processing industries in terms of out- put multipliers, eighth in employment multipli- ers, and ninth in income multipliers. Preserved Foods The preserved food industry has the second highest income multiplier and total employment effect of any food-processing industry. Thus, when a food-processing industry is being tar- geted for an initial increase in income or an in- crease in employment through an increase in 24 output, this industry is worthy of investigation. The ‘demand factors considered in the second section of this bulletin are not clearly in favor of or against this sector. Preserved foods are in the middle of food-processing industries in terms of output multipliers, ranking fourth, and fifth in employment multipliers. “Other Food” Products Products in the “other food” category are not , prominent in their impacts on the Texas economy. Among food-processing industries, it ranks eighth in income multipliers and ninth in output and employment multipliers. This situ- ation is rather disadvantageous because the demand for many of these products is trending upward. The important consideration is that the market is more likely to sustain additional food- processing plants of this type. Beverages When an initial increase in employment is sought through a food-processing industry, the beverage industry should be considered. It ranks second in the agricultural processing sectors for p employment multipliers. It also ranks fourth l‘ among food-processing industries in terms of A i income multipliers and fifth in terms of output multipliers. The demand factors are mixed in terms of its future demand. Applications of this Research The usefulness of this research can be dem- onstrated through an example. The figures in Table 1 1 indicate that a $1 million change in final demand for poultry products generates the highest impact on output and employment of any food-processing industry and the fourth highest impact on income. The poultry industry is the highest ranked industry overall by these indicators and thus will be examined more closely in the following case study. The Census of Manufactures (U.S. Depart- ment of Commerce 1987) indicates that the meat and poultry industries combined lost 28 plants (11.48%), and their value-added contributions decreased 20.02% between 1982 and 1987. These losses were some of the highest losses ex- perienced by the Texas food-processing indus- tries over this period. However, the value of their shipments increased 17.04% and their employ- ; ment increased 11.85%. Trends since 1987 in '\ the Texas poultry-processing sector are not readily available from secondary data. From the input-output model, the main mar- kets for poultry products include the households (60.26%), exports (27.15%), service (3.28%), ani- mal feeds (2.82%), and meat-processing (2.23%) sectors. Although increasing demand for poul- try products by any sector helps the Texas economy, the economy benefits more from fur- ther processing of poultry products when value is added at each processing, handling, or dis- tribution stage (including handling by the ser- vice sector). These sectors are not identified in- dividually for the export markets by the input- output model. Determining whether poultry is further processed out-of-state would facilitate the determination of whether poultry could be further processed in Texas instead. According to the input-output modeling re- sults, major cost components of poultry-process- ing plants were devoted to inputs from the poul- try and egg (49.47%), household (l4.97%), and import (16.05%) sectors. The poultry and egg production sector provides the poultry slaugh- tered by poultry processors, while the house- hold sector provides the laborers to process the poultry. The input-output model does not allow identification of the import markets, so further investigation is required to determine What types of inputs are imported and whether Texas in- put suppliers could provide these inputs. To put these figures into perspective, Texas poultry processors shipped $60 1.4 million worth of processed poultry products in 1986. In addi- tion, median sales of Texas poultry-processing plants in 1990 were $10 to $50 million. Fur- thermore, in 1986, final demand constituted 91% of total sales for processed poultry prod- ucts. For simplicity, assume that the median poultry plant size and final demand as a per- centage of sales were approximately the same between 1986 and 1990. A 10% change in final demand would be $54.7 million, which could be met by two new poultry-processing plants having sales of approximately $25 million each. The location of these plants could be determined by mathematical programming or spatial equi- librium analysis. Further information from con- sumer focus group interviews, econometric mod- els, and / or poultry industry experts can be com- bined to arrive at a clearer assessment of the ‘demand trends for poultry. The other issue is whether additional poultry demanded could be 25 processed by a Texas supplier. Either the poul- try and egg, household labor, imported, and other inputs would need to be provided in a greater quantity or an expanding poultry-pro- cessing sector would have to seek supplies else- where. A further issue to be investigated is whether poultry processors could be attracted to Texas. Currently, most poultry is processed by one large firm in Arkansas. Institutional bar- riers might deter plants from locating in Texas. Suggestions for Future Research Many suggestions for future research were previously mentioned in this chapter. Note that this research has investigated food products at a general food classification level. An initial starting point would be to select one of these general food-processing industries for further study and evaluate the food products at a more detailed level of SIC classification. For example, if the meat industry were chosen, the beef, pork, lamb, veal, sausage, and other meat industries would be evaluated in greater detail to deter- mine any positive impacts on the Texas economy if it were expanded. As an application, economic and employment impacts associated with the Texas rice industry can be identified and quantified. The rice indus- try contributes much to the economy of the up- per Texas Gulf Coast. Using the previously cal- culated multipliers, Taylor et al. (1993) calcu- lated the economic impact of the Texas rice in- dustry for the 1992 marketing year to be nearly $850 million. They also reported that 6,266 per- son-years of employment are created and main- tained in association with the industry. Further evaluation of these specific indus- tries could take the form of econometric analy- sis of trends in value of shipments, value added, number of plants, and number of employees, as well as demand for the different products. Con- sultation with industry experts would enhance model estimates because experts can provide in- formation about trends that would not be rep- resented in time-series data, such as new tech- niques creating new products that may not fit neatly into current product classifications or such as changing production processes and as- sociated costs. Experts could also inform the researcher about institutional or other factors that would preclude locations from consider- ation by plant managers. These factors could be used to interpret results from mathematical programming or spatial equilibrium models employed to determine the potential locations of plants. Input-output models could be further modified to yield multiplier and elasticity val- ues for more detailed food-processing classifi- cations. These input-output model results could also be modified to produce other information that might better suit policy makers’ objectives, such as water use or tax multipliers. Many suggestions mentioned in this section are related to further investigation of the gen- eral reasons that food processors indicated were factors in their decisions to purchase inputs in Texas or outside Texas. These reasons merit further investigation because they point out ar- eas in which Texas can improve or promote its inputs to potential buyers (new food-processing plants). Further investigation into the destina- tion of exports is warranted to determine whether exported products are further processed and, if so, whether any of this further process- ing could occur in Texas. These improvements would make the multipliers presented here ob- solete, probably increasing their values. Other research issues that follow up on this research include the creation of regional mod- els for Texas. Because of the size of the state, different regions have different strengths and weaknesses in terms of input availability. In addition, in-depth demand and locational analy- sis needs to be conducted for the food-process- ing industries because the basic premise of multiplier analysis is that industries are expand- ing because of a change in final demand. An- other issue for further study is the comparative advantage of Texas relative to other states in food and commodity production. Furthermore, the issue of whether Texas can increase its out- put, income, and employment by using more Texas (versus out-of-state) commodities in res- taurants, transportation, and retail trade mer- its further investigation. The question of loyalty- based purchase of Texas inputs also needs to be addressed. Conclusions Many state governments are interested in in- vestigating opportunities to increase state output, income, and employment. This research has pre- sented a framework for investigating opportuni- ties in the Texas food-processing industry. While additional research will definitely be useful, this fiamework will aid researchers in organizing a plan 26 for further research that investigates specific op l‘ portunities for food-processing plants. LITERATURE CITED Archibald, S.O., and D.C. Dahl. Regulating U.S. Food Processing Industries. Economics of the U.S. Food Processing Industry, ed. C.O. McCorkle, Jr., pp. 167- 214. San Diego, CA: Academic Press, 1988. Babb, E.M., and J.V. Chambers. Economics of Dairy Processing in the United States. Economics of the U.S. Food Processing Industry, ed. C.O. McCorkle, Jr., pp. 279-313. San Diego, CA: Academic Press, 1988. Brown, D.J., and L.F.A Schrader. Cholesterol Informa- tion and Shell Egg Consumption. American Journal of Agricultural Economics 72( l990):548-555. Bureau of Business Research. 1990 Directory of Texas Manufacturers. University of Texas at Austin, mag- netic copy. Buse, R.C., and A. Fleischner. Factors Influencing Food Choices and Expenditures. University of Wiscon- sin, Department of Agricultural Economics. Eco- nomic Issues 68(1982). Capps, O. Jr., S.W. Fuller, and J .P. Nichols. Assessing Opportunities in Food and Fiber Processing and Distribution. American Journal of Agricultural Eco- \ nomics 70(1988):462-468. Capps, O. Jr., J .R. Tedford, and J . Havlicek, Jr. House-w hold Demand for Convenience and N onconvenience Foods. American Journal of Agricultural Economics 67(l985):862-869. Capps, O.J., and J .D. Schmitz. A Recognition of Health and Nutrition Factors in Food Demand Analysis. Western Journal of Agricultural Economics 16(1)(l991):2l-35. ‘ Carman, and B.C. French. Economics of Fruit and Vegetable Processing in the United States. Econom- ics of the U.S. Food Processing Industry, ed. C.O. McCorkle, Jr., pp. 215-239. San Diego, CA: Aca- demic Press, 1988. Christy, R.D., and J .M. Connor. Economic Forces Influ- encing Value-Added Food Industries: Implications for Southern Agriculture. Southern Journal of Agri- cultural Economics 2 1 (1 989): 13-22. Connor, J .M. Food Processing: An Industrial Powerhouse in Transition. Lexington, MA: Lexington Books, 1988. Connor, J ., D. Heien, J . Kinsey, and R. Wills. Economic Forces Shaping the Food Processing Industry. Arneri- can Journal of Agricultural Economics 67(1985):1 136- 1 142. Connor, J .M., and R.L. Wills. Marketing and Market Structure of the U.S. Food Processing Industries. Economics of the U. S. Food Processing Industry, ed. C.O. McCorkle, Jr., pp. 117-166. San Diego, CA Academic Press, 1988. Dommeyer, C. J . Does Response to an Offer of Mail Sur- vey Results Interact with Questionnaire Interest? Journal of Market Research Society (1985):27-38. Farris, P.L, R.T. Crowder, R.P. Dahl, and S. Thompson. Economics of Grain and Soybean Processing in the United States. Economics of the U.S. Food Process- ing Indusuy, ed. C.O. McCorkle, Jr., pp. 315-348. San Diego, CA: Academic Press, 1988. Ferris, J . State Value Added Activities as Related to Eco- nomic Efficiency. Value-Added Research Invest- ments: Boon or Boondoggle? Proceedings of an Or- ganized Symposium. Ed. G.W. Williams. TAMRDC Special Series Report No. SS-1-89, August 1989. Gallo, A.E., L.E. Salathe, and W.T. Boehm. Senior Citi- zens: Food Expenditure Patterns and Assistance. Washington, DC: U.S. Department of Agriculture. Econ. Statistics and Cooperative Service Agricultural Economics Report 426, 1979. Giertz, J .F., and J .B. Crihfield. Impact of Business Cli- mate Factors on Value-Added Activities in Illinois Agriculture: A First Analysis. University of Illinois at Urbana-Champaign, College of Agriculture Bulle- tin 800, 1990. Greene, M.J. Agricultural Diversyication Initiatives: State Government Roles in Rural Revitalization. Lexing- ton, Kentucky: Council of State Governments. Ru- ral Economic Alternative Series Technical Assistance Bulletin No. 2, 1988. Holt, D.A. A Research Administrator's Perspective on Value-Added Research. Value-Added Research In- vestments: Boon or Boondoggle? Proceedings of an Organized Symposium, ed. G.W. Williams. TAMRDC Special Series Report No. SS-1-89, August 1989. lsard, W. Methods of Regional Analyses: An Introduction to Regional Science, pp. 251-254. Cambridge, MA: MIT Press, 1960. Jones, L.L., and G. Mustafa. Structure of the Texas Economy: Emphasis on Agriculture. College Station, TX: Texas A&M University, Texas Agricultural Ex- periment Station, B-1121, 1972. Kinsey, J ., and D. Heien. Trends in the Consumption and Production of Processed Foods. Economics of the U.S. Food Processing Industry, ed. C.O. McCorkle, Jr., pp. 47-81. San Diego, CA: Academic Press, 1988. Kraybill, D., and T. Johnson. Development of Value- Added Industries as a Rural Development Strategl. Southern Journal of Agricultural Economics 21(1989):13-22. Logan, S.H., L.J. Steinmann, and D.E. Farris. Econom- ics of Meat Processing in the United States. Eco- nomics of the U.S. Food Processing Industry, ed. C.O. McCorkle, Jr., pp. 241-277. San Diego, CA: Aca- demic Press, 1988. Mautz, R.F., and F.L. Neumann. The Effective Corpo- rate Audit Committee. Harvard Business Review (1970):58. 27 Mazzocco, M.A. Personal communication, Department of Agricultural Economics, University of Illinois, 1991. McCorkle, C.O., Jr. U.S. Food Manufacturing Industries: An Overview. Economics of Food Processing in the United States. ed. C.O. McCorkle, Jr., pp. l-46. San Diego, CA: Academic Press, Inc., 1988. McCorkle, C.O., S.O. Archibald, and A.F. McCalla. Food Processing Industry Dynamics and Economic Policy. Economics of the U.S. Food Processing Industry, ed. C.O. McCorkle, Jr., pp. 409-434. San Diego, CA: Academic Press, Inc., 1988. McMenamin, D., and J. Haring. An Appraisal of Nonsurvey Techniques for Estimating Regional In- put-Output Models. Journal of Regional Science 14(1974):191-204. Meyers, L.H. Economic Forces Influencing Value-Added Food Industries: Implications for Southern Agricul- ture: Discussion. Southern Journal of Agricultural Economics 2 1(1989):23-25. Moore, K.L. Food and Fiber Center Adds Value to Mississippi's Ag Economy. Rural Enterprise (Fall 19871120-22. Mustafa, G. An Input-Output Model for the Texas Economy with Emphasis on Agriculture. Ph.D. dis- sertation, College Station,Texas A&M University, 1971. Nichols, J .P. Food Industry Trends in the Southwest: Size, Structure, and Value-Added. Presentation to the Texas Food Processors Association Armual Meet- ing, April 28, 1988, San Antonio, TX. O'Rourke, A.D. Value-Added Research as a Strategl to Boost Foreign Demand for Domestically Produced Agricultural Commodities. Value-Added Research Investments: Boon or Boondoggle? Proceedings of an Organized Symposium, ed. Gary W. Williams. TAMRDC Special Series Report No. SS-1-89, Au- gust 1989. Otto, D.M., and G.W. Williams. Value-Added Research as a State Economic Development Program. Value- Added Research Investments: Boon or Boondoggle? Proceedings of an Organized Symposium, ed. G.W. Williams. TAMRDC Special Series Report No. SS-1- 89, August 1989. Ozuna, T. Fortran programs with which to update an input-output model. Department of Agricultural Economics, College Station, Texas A&M University, 1986. Paarlberg, P. L. Export Promotion of High-Value Agricul- tural Products: Who Gains? Value-Added Research Investments: Boon or Boondoggle? Proceedings of an Organized Symposium, ed. G.W. Williams. TAMRDC Special Series Report No. SS-1-89, August 1989. Reeh, E. The Potential for Sustainable Competitive Ad- ' vantage in Texas Wool and Mohair Processing. M.S. thesis, College Station, Texas A&M University, 1990. Redman, B.J. The Impact of Women's Time Allocation on Expenditures for Meals Away From Home. Ameri- can Journal of Agricultural Economics 62(1980):234- 237. Salathe, L. Demog1'aphics and Food Consumption. Eco’- nomic Statistics and Cooperative Service, U.S. De- partment of Agriculture. National Food Review 8(1979]:31-34. Washington D.C. Sanderson, G.W., and B.S. Schweigert. Changing Tech- nical Processes in U.S. Food Industries. Economics of the U.S. Food Processing Industry, ed. C.O. McCorkle, Jr., pp. 83- 1 15. San Diego, CA: Academic Press, Inc., 1988. Sexauer, B.H., and J.S. Mann. Food Expenditures of ‘Single-Person Households. Washington DC: U.S. Department of Agriculture Agricultural Economic Report 428, 1979. Sporleder, T.L., and L.J. Hushak. OHFOOD: An Ohio Food Industries Input-Output Model. Columbus, Ohio: Farm Income Enhancement Program, The Ohio State University, unpublished manuscript, February 1991. 28 Taylor, E.L., M.E. Rister, H.L. Goodwin, and M.L. Waller. " The Economic Impact of the Texas Rice Industry. FP93-6, Department of Agricultural Economics, College Station, Texas A&M University, June 1993. Texas Department of Agriculture. Economic Growth through Agricultural Development? A Blueprint for Action. Austin, TX, 1986. ' Tilley, D. Value-Added Activities as a Rural Development Strategy: Discussion. Southern Journal of Agricul- tural Economics 2 1 (1989) :37-40. Tull, D., and D. Hawkins. Marketing Research: Measure- ment and Method New York: Macmillan Publishing Company, 1987. U. S. Department of Commerce, Bureau of the Census. Census of Manufactures. Washington, DC: MC87- A-44, 1987. Wildenthal, M. Opportunities for Enhancing the Texas Food Processing Industry. Ph.D. dissertation, Col- lege Station, Texas A&M University, 1992. ‘a [Blank Page in Original Bulletin] ‘ Texas Agricultural. Experiment Station The Texas A&M University System Mention of a trademark or a proprietary product does not constitute a guarantee or a 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. The Texas Agricultural Experiment Station is an equal opportunity/affirmative action employer. Copies printed: 1,000 W