Example: dental hygienist

Brazil's Beef Production and Its Efficiency

Brazil's Beef Production and Its Efficiency: A Comparative Study of Scale Economies Brazil ranks as the second largest world beef producer with the world’s biggest




Link to this page:

Please notify us if you found a problem with this document:

Other abuse

Text of Brazil's Beef Production and Its Efficiency

Brazil's Beef Production and Its Efficiency: A Comparative Study of Scale Economies by Agapi Somwaru and Constanza Valdes*, In recent years, Brazil one of the world s main suppliers of agricultural products- has been raising beef productivity and exports. In Brazil, large farm land availability, ample feedstuffs supplies, a large domestic consumer market, and liberalization of trade barriers have allowed large firms to achieve economies of size that have made the country a major, growing source of meat production. Major differences exist between the modern and the traditional segments of the beef-cattle sub-sector. To assess competitive strength of Brazil's livestock operations, we estimate non-parametrically the efficiency and scale elasticity of 450 cattle operations. Our hypothesis is that large-size specialized farms exhibit cost economies and higher efficiency levels than other beef production operations. GTAP Seventh Annual Conference on Global Economic Analysis Trade, Poverty, and the Environment June 17-19, 2004 The World Bank, Washington, , United States * Agapi Somwaru, Senior Economist and Constanza Valdes, Brazilian Analyst are both with the Economic Research Service, Department of Agriculture. The authors extend thanks to John Dyck, Kenneth Mathews, Mary Burfisher and Demcey Johnson for helpful review comments and to Agnes Prentice for her statistical assistance. All errors and omissions remain with the authors alone. 1Brazil's Beef Production and Its Efficiency: A Comparative Study of Scale Economies Brazil ranks as the second largest world beef producer with the world s biggest commercial herd (170 million head). Brazil has abundant grazing land for calf and grass-fed beef production in the Cerrados region. Embodied technology advancements, lower labor costs, and a large domestic market have encouraged the development of large beef processing operations. However, Brazil's livestock sector has gone through a process of selective modernization. Major differences exist between the modern and the traditional segments of the beef-cattle sub-sector. The purpose of this paper is to assess the competitive strength of Brazil's meat production and gain insights into the country s divergent livestock production activities, cost efficiencies, feed use, and competitiveness. In particular, the paper examines beef production s overall efficiency in Brazil and measures its competitiveness using a non-parametric technique. This paper looks closely at Brazils regional diversity in livestock raising activities and examines the intensity of the various types of operation in order to evaluate their competitiveness and identify the sources of differences. Between 1970 and 1991, Brazil's beef-cattle herd grew at a percent average yearly rate, from million to million head. From 1963 to 2003 beef-cattle slaughter increased from million to million head; and the total carcass weight increased from million to million tons. However, these numbers hide large differences between regions. The beef-cattle industry in areas near the country's more developed core has experienced considerable modernization along with expansion of a dynamic agribusiness sector, which supplies the industry with modern inputs and 2slaughters and processes animals for domestic and world markets. As a result, Brazil's beef exports increased from total value of US$ million (in 1992 dollars) at a total value of US$ million. A still substantial traditional beef-cattle industry can be found in the frontier areas and the more backward parts of Brazil; its productivity remains very low, and it is plagued by serious livestock diseases and management problems. Brazil still has low per capita consumption of beef products, averaging kg/year in 2001. Brazil s Beef Production System Beef production accounts for about 20 percent of Brazilian agribusiness gross income. Brazil s production system is based on grass with less than 3 percent in feedlots. There are primarily three types of beef operations in Brazil: grass-fed small enterprises, grass-fed medium specialized operations, and grain-fed/grass-fed large commercialized beef operations. The small beef enterprises raise less than 500 head per household per year. The so-called medium specialized beef operations on average produce over 1000 head annually. The last category is called commercialized beef production enterprises with over 4000 head per year. Production costs in Brazil are estimated to be 60 percent lower than in Australia and 50 percent lower than in the United States. The average slaughter age is at 4 years and the slaughter rate is just 21 percent, compared to 2 years and 37 percent in the United States (see Dyck and Nelson, 2003). In the last fifteen years, beef and veal production in Brazil experienced a dramatic increase in total quantity produced and the number of animals raised (figure 1). Yield per animal from 1980 s to 2000 s increased by almost 20%. The beef industry in Brazil is 3characterized by dispersion and lack of integration. This is unlike the Brazilian poultry industry, which has been developed around the concept of strategic groups for commodities (chicken) and specialties (processed products), and is characterized by high productivity and high technology use. The land ares in the Cerrados totals 127 million hectares, with 49 million hectares of cultivated pasture and 12 million hectares of annual crops. Available land in the Cerrados is estimated at 60 million hectares, which translates into a potential production of 240 million tons of grains. Brazil s Cattle Production--a Closer Look We use a unique database, the Livestock Costs and Practices Survey conducted annually since 1993 by FNP Consultoria & AgroInformativos, a consulting company specializing in agribusiness. The survey includes detailed information on cost of production, feed use, and other important economic characteristics of beef production. Carried out in the most important beef producing areas in Brazil, the survey provides a picture of beef production efficiency, costs, and feed use by size and types of production during the 2000-2002 period. The survey provides detailed cost of production information for 25 beef producing locations in Brazil, corresponding to the 5 major agricultural regions in Brazil, grouped by State. In the next sections we present the distributional characteristics and the production profile of cattle operations in Brazil by region, type, and size. Cattle production by region 4Although cattle production occurs in all regions, it is generally concentrated in the northern, northeastern, southern, southeastern, and central regions of Brazil. Both figure 2 and table 2, which presents the balance sheet of cattle operations by region, reveal that operations in the northeastern and southeastern regions are more profitable. Detailed balance sheets of revenue and cost by state reveal the heterogeneity in cattle operations in Brazil. Cattle production by type and size Revenue and cost vary greatly by type of operation (table 4). Profitability increases as operations become less intensive in raising cattle. Overall, type I (cow/calf) and type II operations (background/feedlot) are most profitable followed by type III operations (calfing (birth) to slaughter). Table 5 presents the balance sheet revenues and costs by type and size of operation. The table indicates that profitability increases with size. Independent of the type of operation, large cattle operations or operations with 5,000 AU (animal unit) and more are much more profitable than small cattle operations (with 500 AU). Globalization of Brazil s Meat Products With rising incomes and population, changing diets, and liberalization of trade barriers, global meat trade has reached a value of over $40 billion, and has grown at about 6 percent annually since 1990 (Dyck and Nelson, 2003). Meat trade flows among countries are determined largely by differences in countries resource bases, preferences for meat types and cuts, barriers to trade, and the domestic industry structure. Producing countries with low-priced inputs and favorable natural resource bases for beef exports, 5like Brazil, have competitive advantages in meat production. Land for forage and grain production is crucial for livestock success and Brazil s exports of meat products have increased drastically in the last decade (figure 3). Beef exports from Brazil are the third largest in the world, behind the United States and Australia, with one-third in processed meats (mainly corned beef) and two-thirds in frozen/chilled meats value (see Table 1). Despite growth in domestic beef demand -Brazil is the world s second largest consumer market with over 7 million tons- exports have increased at a much faster pace, making this country a net exporter of livestock and livestock products. Given the potential for feedstuffs production (corn, soybeans, etc.), Brazil is also able to keep the costs of feed rations low for grain-fed beef. Despite lack of financing capital, lower labor costs and a large domestic market have encouraged the development of large beef-processing operations in Brazil. Research Methodology To evaluate the competitive strengths of Brazilian beef production, we use the data envelopment analysis (DEA) non-parametric method to measure the economic efficiency and scale economies while accounting for resource costs (fixed costs, operational costs, feed costs, feed efficiency). The analysis identifies cost economies and efficiency levels for each type of operation. The Model and Estimating Method Our study attempts to assess the economic performance, measured by overall efficiency, of the country s cattle production and structural changes. Even though the 6beef industry in Brazil is characterized by dispersion and lack of integration and is dominated by many small operations, cattle production has begun to shift and become more concentrated in large specialized operations. These changes have profound effects on the industry s performance, and measuring the impacts of these changes is therefore of great interest. The method used to assess these changes is presented below. Analysis of structure and performance of cattle operations begins with the underlying production technology. This can be formalized by specifying a transformation function, S(Xn,Y) = 0, which minimizes the production frontier in terms of inputs Xn and output Y. Information on the production technology can be characterized via an input set, F(Y,Xn), that represents the set of all inputs Xn that can produce Y. An input distance function (denoted by superscript i) recognizes the least input use possible for producing the given output vector as defined by F(Y, Xn): (1) {)}X,Y(F)/x(:max)X,Y(Dnni = where Di is the distance of unit i, Y is output, Xn (n=6) are the inputs labor, feed expenses, expenses for purchasing animals (feeder cattle), other variable expenses, fixed cost, and other expenses (indirect cost). We use a programming method to estimate the input distance function and capture the distance from the frontier assuming a radial contraction of inputs to the frontier of cattle operations. The ratio of estimated potential efficient input use compared to the actual observed use provides an estimate of technical efficiency. Further, scale economies can be measured by identifying variations in the input and output ratio at different scale levels when variable returns to scale are allowed. 7For our empirical implementation, the solutions to this problem were developed and computed in GAMS (General Algebraic Modeling Systems, Brooke et al., 1988). Functional relationships of production or distance functions represent a foundation for data envelopment analysis (DEA) procedures, which use programming rather than econometric (parametric) techniques. Formally, an input-oriented programming problem may be written as: (2) JjnmXXYYtsJjjnjJjjniimimjJjji,...,1,6,.. .,1,1,1,0, ,,==== = === There are J observations and the non-negative weights, j, determine the reference points on the frontier for unit i, where unit in this study is represented by state or region or type of operation under evaluation. For notational simplicity, the unit index i is suppressed on the -weights. The input vector in (2) for unit i is adjusted by the efficiency score, i, (Di = i) and then compared with the reference point, , on the frontier. njJjjx =1 The Lagrangian of equation (2) is set up in such a way that the shadow prices of outputs and inputs, umi and vni, respectively, are non-negative: (3) )1()()(11111 = =====JjjininjJjjniiNnnimimjJjjMmmiiuxxyy uL where is the shadow value of the equality constraint on the sum of the s. Since the value of the shadow value is unique for inefficient units, we utilize the radial projection iniu 8approach for calculating scale elasticity values (see Forsund, and L. Hjalmarsson (1979). Estimation of scale elasticities using non-parametric techniques, while straight forward for efficient points ( , points that satisfy F(Y,X) = 0), are ambiguous for inefficient points. Following Forsund et al. we use the radial projection approach and calculate for each farm the scale elasticity as follows: (4) Ii,uEE)XEY(iniiiiii = where Ei is the input-oriented efficiency score and is the shadow price on the equality constraint iniu11= =Jjj . The farm exhibits increasing returns to scale if > 0, constant returns to scale if =0, and decreasing returns to scale if <0. iniuiniuiniu Results Using the survey data we employed the methods described in the previous section and constructed measures of efficiency and scale economies for our sample of cattle operations in Brazil. Our deterministic procedure estimates the best-practice production frontier from these data and compares individual state or region or type of operation to the estimated frontier composed by all states or all regions or all types of operations. For this purpose, we use programming to estimate an input distance function. Our model, as stated previously, is based on one output and six inputs. Estimates of the scale (SEC) and efficiency performance (TE) indicators by region are presented in Table 6. The TE measures suggest that farms have a greater 9potential to reduce costs by moving to a scale-efficient point as the SEC measures reveal scale economies (IRS). These results support the idea that performance across farms arises largely from efficiency and netput (output and input) composition changes accompanying growth. The efficiency or performance measures at the regional level reveal that cattle operations in the northern region are more efficient ( on average) while operations in the central region have the lowest scores ( on average). The detailed state-level results also find increasing returns to scale while TE exhibits low variability. This is the case because the frontier is estimated using all types of operations in each state. On a regional basis, the efficiency indicators (TE) for the southern region (Sao Paulo, Rio Grande do Sul and Parana) reflect highly mechanized commercial farming with infrastructure in place and high input intensity. Yet even in these regions, competitiveness has been constrained by a number of factors, including macroeconomic shocks, weakness in the financial system, and trade protection in other countries. The estimates over the whole sample, by class size, and by region, display some interesting differences (Table 7). The overall efficiency estimate, using all 450 cattle operations, is , indicating low overall efficiency, but the scale elasticity shows increasing returns to scale ( ). The most obvious differences are revealed by the measures of efficiency and scale elasticity by class size. For type III integrated operations (from calfing (birth) to slaughter) have the largest overall efficiency from to while the scale elasticity suggests increasing returns to scale. TE varies when the frontier is estimated by type (type I (Cria), type II (Recria/Engorda), and type III (Cria/Recria/Engorda)) and intensity. 10In sum, the results indicate that integrated operations (type III-- calfing (birth) to slaughter) have higher performance and netput savings, as expected. The scale elasticity by region indicates potential to reduce costs by expanding operation rather than by rearranging input composition. Summary and Conclusions This paper examines the changing structure of beef production in Brazil and measures its efficiency and competitiveness (reduced costs). After analyzing the feed efficiency and structure of production costs directly from survey data and estimating the overall efficiency and scale economies of cattle operations in the sample, we can conclude that the most integrated operations (type III) are most efficient with increasing returns to scale economies. The overall efficiency estimate, using all 450 operations, is , indicating low overall efficiency. However, the scale elasticity ( ) indicates increasing returns. For operations type I to type III , overall efficiency varies from to , respectively. Overall efficiency, measured by the efficiency scores i, is the highest for the integrated operations (type III). The estimated scale elasticity indicates increasing returns to all types of cattle operations. References Dyck, J. and K. Nelson (2003). Structure of Global Markets for Meat, Agriculture Information Bulletin Number 785, Economic Research Service, Department of Agriculture. 11Forsund, and L. Hjalmarsson (1979). Generalized Farrell Measures of Efficiency: An Application to Milk Processing in Swedish Dairy Plant, Economic Journal, 89, 29 4-315). Skoufias, E. Using Shadow Wages to Estimate Labor Supply of Agricultural Households. American Journal of Agricultural Economics, May 2000 (82): 287-297. Zhnag, X. The Current Situation and Benefit/Cost Analysis of Rural Household Hog Production in China. Research Report of China s Rural Economy: 1990-1998. Edited by the Research Center of Rural Economy, Ministry of Agriculture, Beijing, March 1999. 12Figure 1. Brazil's Meat Production, 1961-20030500010000150002000025000300003 5000400001961196319651967196919711973197 5197719791981198319851987198919911993199 51997199920012003Slaughtered/Production Animals010002000300040005000600070008000 Production1,000 Heads 1,000 Mt Source: FAOSTAT Figure 2. Revenue, Cost, and Profit of Cattle Operations in Brazil by Region, 2002 Brazilian Reals (R$) 0 100,000 200,000 300,000 400,000 500,000 600,000 700,000 800,000 900,000 1,000,000 North Northeast SouthSoutheastCentrals (R$ Revenue Total cost Profit Source: FNP Survey, 2002 13 Figure 3. Exports of Brazil's Meat Products, 1961-20030500,0001,000,0001,500,0002,000 ,0002,500,0003,000,0003,500,000196119631 9651967196919711973197519771979198119831 98519871989199119931995199719992001Thous and $US Meat and Meat PreparationMeat Bovine FreshMeat Fresh+Ch+Frozen Source: FAOSTAT 14 Table 1. Leading beef exporting countries, 2000 Rank Volume Million metric tons Value Billion US$ 1 2 Australia Australia 3 EU Canada 4 Canada Brazil 5 New Zealand New Zealand 6 Brazil EU 7 Argentina Argentina 8 India Uruguay 9 Uruguay India 10 Ukraine Ukraine Includes preparations of bovine meat. Data: FAOSTAT, bovine meat. Table 2. Balance Sheet of Production and Cost of Cattle Operations by Region, 2002 (in Reals R$) Region Revenue Labor expenses Variable expenses Feed Cattle purchased Fixed cost Other expenses Total cost Returns (above total cost) North 819,518 69,013 13,846 119,96559,708208,055 97,626 568,213251,305Northeast 908,534 67,521 134,164 54,751197,86298,294 14,148 566,740341,793South 882,906 82,637 102,808 53,239222,831117,024 19,699 598,239284,667Southeast 940,470 76,417 123,176 55,704223,763106,008 17,264 602,332338,139Central 890,553 71,226 111,531 52,905208,619102,531 15,177 561,988328,565Data Source: FNP Survey, 2002 15 Table 3. Balance Sheet of Revenue and Cost by State, 2002 (in Reals R$) Item State State name Revenue Labor expenses Variable expenses Feed Cattle purchased Fixed cost Other Cost Total cost Returns (above total cost) North TO Gurupi 817,866 68,343 133,382 63,335 186,554 95,615 13,565 560,794 257,072 North TO Aragua na 858,190 71,108 113,437 49,781 217,057 97,186 14,113 562,682 295,508 North PA Reden o 811,362 69,648 118,556 60,881 214,966 95,491 13,057 572,599 238,764 North PA Paragominas 783,409 69,334 124,699 80,516 206,508 98,327 14,529 593,912 189,497 North RO Ariquemes 826,761 66,632 109,751 44,029 215,190 101,512 13,964 551,077 275,684 Northeast BA Barreiras 913,166 67,673 133,584 57,335 200,532 97,989 14,179 571,291 341,875 Northeast BA Itapetinga 903,902 67,369 134,744 52,166 195,192 98,600 14,117 562,190 341,712 South RS Alegrete 790,335 81,892 81,826 48,830 211,415 110,534 19,518 554,016 236,318 South PR Paranava 975,477 83,382 123,789 57,648 234,248 123,514 19,879 642,461 333,015 Southeast SP P. Prudente 1,013,756 88,636 125,376 57,234 252,895 120,719 21,491 666,351 347,404 Southeast RJ Campos 940,437 74,390 100,980 58,571 217,265 109,693 16,405 577,304 363,133 Southeast MG Ituiutaba 967,749 76,666 133,870 60,069 236,150 105,124 17,751 629,631 338,118 Southeast MG 7 Lagoas 895,970 72,943 129,144 57,453 203,348 104,558 16,890 584,335 311,635 Southeast MG M. Claros 884,439 69,448 126,511 45,195 209,156 89,945 13,783 554,038 330,401 Central MS Camapu 942,026 74,196 128,455 59,240 221,198 102,665 16,363 602,117 339,909 Central MS C. Grande 923,585 77,394 127,877 57,790 221,983 96,372 17,064 598,481 325,104 Central MS Corumb 737,912 61,510 59,330 32,808 163,751 95,814 12,204 425,418 312,494 Central MS Navira 990,928 74,504 114,398 48,089 252,112 91,754 17,252 598,109 392,818 Central MT B. Gar as 890,730 71,021 131,833 58,617 194,058 93,930 15,663 565,122 325,608 Central MT A. Floresta 878,405 70,466 106,163 54,872 218,889 102,800 13,989 567,178 311,227 Central MT P. Lacerda 917,873 70,022 100,785 53,423 227,840 99,778 13,900 565,747 352,126 Central MT Pocon 702,101 61,897 57,209 34,825 157,614 120,200 12,283 444,028 258,073 Central GO Ca u 951,442 75,898 135,886 61,948 217,096 107,311 15,896 614,034 337,408 Central GO Goi nia 924,405 73,489 134,479 60,848 207,774 111,584 17,016 605,190 319,215 Central GO N. Crix s 936,673 73,087 130,429 59,498 212,492 105,627 15,311 596,445 340,229 Data Source: FNP Survey, 2002 16 Table 4. Balance Sheet of Revenue and Cost by Type of Cattle Operation, 2002 (in Reals R$) Revenue Labor expenses Variable expenses Feed Cattle purchased Fixed cost Other expenses Total cost Returns (above total cost) Type I (Cow/calf) Intensive 1,718,698 199,978 310,662 257,792 5,708 251,120 418,824 1,068,761 649,937 Semi-Intensive 1,449,871 161,254 232,998 65,147 11,275 208,452 65,887 714,197 735,674 Extensive 1,194,243 133,660 189,731 92,361 20,172 182,467 99,179 646,711 547,532 Type II (background/feedlot) Intensive 2,704,381 137,781 181,492 139,501 1,497,490 209,650 1,686,806 2,195,594 508,788 Semi-Intensive 2,703,375 126,859 216,480 47,988 1,274,690 191,172 1,065,616 1,884,495 818,879 Extensive 2,110,030 101,894 188,477 52,973 980,726 166,081 913,333 1,511,682 598,349 Type III (calfing (birth) to slaughter) Intensive 1,505,291 175,719 316,177 191,046 3,448 239,379 422,657 963,974 541,317 Semi-Intensive 1,421,801 147,637 255,774 59,226 6,681 213,272 69,315 714,691 707,110 Extensive 1,161,116 119,099 208,084 83,967 12,411 193,591 123,625 642,370 518,746 Data Source: FNP Survey, 2002 Table 5. Balance Sheet of Revenue and Cost by Type and Size of Cattle Operation, 2002 (in Reals R$) Revenue Labor expenses Variable expenses Feed Cattle purchased Fixed cost Other expenses Total cost Returns (above total cost) Type I (Cow/calf) Small (500 UA) 396,518 86,850 67,744 34,982 4,634 111,530 250,634 323,576 72,942 Large (5,000 UA) 3,966,293 408,042 665,646 380,317 32,521 530,509 333,256 2,106,093 1,860,201 Type II (background/feedlot) Small (500 UA) 676,193 73,430 54,404 21,057 348,178 99,549 546,098 611,146 65,047 Large (5,000 UA) 6,841,593 293,104 532,045 219,405 3,404,728 467,354 3,119,657 4,980,625 1,860,968 Type III (calfing (birth) to slaughter) Small (500 UA) 368,583 82,977 71,345 24,797 1,999 112,253 252,265 310,424 58,159 Large (5,000 UA) 3,719,625 359,478 708,691 309,443 20,542 533,988 363,333 2,010,612 1,709,013 Data Source: FNP Survey, 2002 17 Table 6. Efficiency and Scale Estimation by Region Regions Efficiency (TE) Scale Elasticity (SEC) North Northeast South Southeast Central 1 RS Alegrete 2 PR Paranava 3 MS Camapu 4 MS C. Grande 5 MS Corumb 6 MS Navira 7 MT B. Gar as 8 MT A. Floresta 9 MT P. Lacerda 10 MT Pocon 11 SP P. Prudente 12 RJ Campos 13 MG Ituiutaba 14 MG 7 Lagoas 15 MG M. Claros 16 GO Ca u 17 GO Goi nia 18 GO N. Crix s 19 TO Gurupi 20 TO Aragua na 21 PA Reden o 22 PA Paragominas 23 RO Ariquemes 24 BA Barreiras 25 BA Itapetinga Data Source: Model results 18 Table 7. Efficiency and Scale Estimation by Type and Size of Cattle Operations Farms Efficiency (TE) Scale Elasticity (SEC) All Farms 450 All Intensive 150 All Semi-Intensive 150 All Extensive 150 All Type I (Cow/calf) 150 Intensive 50 Semi-Intensive 50 Extensive 50 All Type II (background/feedlot) 150 Intensive 50 Semi-Intensive 50 Extensive 50 All Type III (calfing (birth) to slaughter) 150 Intensive 50 Semi-Intensive 50 Extensive 50 Data Source: Model results. 19

Related search results