Slide 1

Joaquim Bento de Souza Ferreira Filho Luclio Rogrio Aparecido AlvesEscola Superior de Agricultura Luiz de Queiroz.University of So Paulo, Brazil.

ECONOMIC IMPACTS OF GM CORN ADOPTION IN BRAZIL

MOTIVATIONA field survey on corn production costs in 2010/2011 showed that the cost of producing GM corn is higher than non GM corn.Even then the adoption of GM seeds is spreading very fast in Brazil.In this paper we analyze some economic aspects of GM corn adoption in Brazil. The research consists of two main parts:A field survey on production costs conducted by CEPEA in 2010/2011 year.A CGE evaluation of the impacts on the economy.2Background: GM corn in BrazilFirst authorization: 2007.Fourteen events presently:7 insect resistant (IR).2 herbicide tolerant (HT).5 stacked (IR+HT).Use of GM corn: widespread in Brazil.Mostly IR (commercial producers).2010/2011: 55% total area (CEPEA); 58% (CELERES).2012/2013: CELERES estimates 64.8% (5.3 Mha first crop); 87.8% (6.9 Mha, second crop).3Corn production, area and productivity in Brazil4

GM introductionSurvey: meetings with stakeholders in crop production, 28 cost structures, GM x non-GM.5

CORN, FIRST CROP. BRAZIL. (% variation in relation to non-GM)6VARIATION [(GM-Non GM)/non GM]ItemLDNCVELGPVACSTXNXCNVRVDUBR% variation% variation% variation% variation% variation% variation% variation% variationFertilizers0,000,000,000,000,000,000,000,00Chemical inputs-32,17-20,90-1,85-31,08-15,87-14,66-33,24-24,96 Herbicides0,000,000,000,000,000,000,000,00 Insecticides-85,78-82,45-50,76-79,38-51,60-37,04-97,18-100,00 Fungicides0,000,000,000,000,000,000,000,00 Seed treatment0,000,000,000,000,00-26,670,000,00Seeds26,6740,0040,0030,4312,1329,6350,0056,25Emulsionable oil0,000,000,000,000,000,000,000,00Mechanical operations-1,89-7,290,00-4,360,00-3,10-0,630,00Transportation0,000,000,000,000,000,000,000,00Labor-1,59-4,730,00-6,390,00-3,91-2,480,00Trade/Storage0,000,000,000,000,000,000,000,00Taxes0,000,000,000,006,2511,110,0014,29Insurance-3,35-3,470,00-2,260,00-1,31-0,540,00Technical assistance0,114,017,09-0,010,724,010,856,38Interest over capital0,093,576,51-0,010,663,770,775,80CO0,093,726,64-0,030,904,280,806,62COT0,182,796,02-0,320,833,660,716,12CT-0,011,784,86-0,440,672,750,564,84CORN, SECOND CROP. (% variation in relation to non-GM)7Variation [(GM-Non GM)/non GM]ItemRVDMNRCVELLDNUNAI% variation% variation% variation% variation% variationFertilizers0,000,000,000,000,00Chemical inputs-18,88-25,37-11,67-36,52-27,37 Herbicides0,000,000,000,000,00 Insecticides-100,00-71,84-43,57-85,78-96,91 Fungicides0,000,000,000,000,00 Seed treatment0,000,000,000,000.00Seeds75,3155,5643,4835,7120,00Emulsionable oil00,000,000,000,00Mechanical operations-0,62-2,33-14,01-3,53-8,28Transportation0,000,000,000,000,00Labor-2,62-0,73-3,06-10,44-2,59Trade/Storage0,000,000,000,000,00Taxes0,000,000,000,000,00Insurance0,10-1,180,00-7,03-5,91Technical assistance7,403,624,771,19-2,33Interest over capital7,403,624,771,19-2,33CO7,153,474,561,12-2,29COT6,572,894,110,41-2,95CT5,662,323,37-0,10-3,32Observation from field surveyNo consistent increase in productivity (2010/2011).GM technology (Insect Resistant) is cost increasing in average: consistent with the fast increase in GM technology in corn in Brazil?Reason for GM technology adoption: question asked in the survey.Predominant answer: risk reduction. Control of worms in the initial stages of corn can be very hard in Brazilian conditions, especially in large areas.Producers accept a reduction in their rate of return in exchange for risk reduction: which size??8Simulating this effect with the aid of a CGE model9Static, inter-regional, bottom-up.42 sectors.42 products (11 agricultural products)10 types of workers (wage classes)27 regions inside Brazil10 household types (income classes)Linearized, solved with GEMPACK.

Closure: environment in which model reaches a new equilibriumCapital stock is endogenous by industry, while the Gross Rate of Return (GRET) in each production sector is exogenous. The exception to this rule is the corn production activity, where:Production is fixed at base year level; the GRET fall necessary to ratify the fixed production level in the presence of cost increase is calculated endogenously: SHADOW VALUE OF RISK REDUCTION.Land stocks are fixed in each region, mobile between activities.Real wages endogenous, aggregate employment fixed. Labor can migrate between regions and activities, driven by real wages. Total (aggregated) investment in the economy is endogenous, and follows aggregated capital stock.10ScenariosModel shocked with the change in cost structure (weighted average): increase in costs (approximately 1.9% in operational costs).Corn production level fixed at base year level.GRET variation: SHADOW VALUE of risk reduction. 11

Results: aggregated (% change)12Land useProductionAgricultural sectorRice0,070,02Corn-1,260,00 (fixed)Other 0,080,01Sugar cane0,070,01Soybean0,100,03Cotton0,060,02Forestry0,080,01Livestock0,090,03Milk0,110,02Results: regional GRET variation. % change.13GRET % variationStateBahia-48,3Minas Gerais-46,6So Paulo-32,9Parana-30,0Santa Catarina-30,5Rio Grande do Sul-30,9Mato Grosso do Sul-38,2Mato Grosso -46,0Goias-46,6Brazil-38,2Results: labor demand by occupation and region (% change)14Fall in unskilled labor demand.State OCC1 OCC2 OCC3 OCC4 OCC5 OCC6 OCC7 OCC8 OCC9 OCC10Bahia-0,07-0,020,00-0,010,000,010,010,010,000,00Minas Gerais-0,27-0,09-0,01-0,01-0,010,010,000,010,010,01So Paulo-0,03-0,02-0,01-0,01-0,010,000,000,000,000,00Parana-0,33-0,15-0,02-0,010,02-0,010,00-0,01-0,010,00Santa Catarina-0,36-0,06-0,03-0,030,020,000,010,010,010,00Rio Grande do Sul-0,14-0,030,00-0,010,000,000,000,010,000,00Mato Grosso Sul-0,14-0,100,02-0,10-0,09-0,010,020,020,010,01Mato Grosso -0,65-0,24-0,060,03-0,010,010,01-0,030,020,02Goias-0,09-0,03-0,030,00-0,04-0,04-0,010,000,000,01Unskilled laborFinal remarksLIMITATION: observations based in just one year. More field surveys needed to assess the impacts on:ProductivityCosts.The reduction in GRET (Shadow value of risk reduction) is calculated on a ex post basis: not really a measure of the willingness to pay for risk, what is probably determined based on a probability distribution of pests infestation.15Thank you.

Email: jbsferre@usp.br16