The causes and unintended consequences of a paradigm shift in corn production practices

Full text

(1)

The

causes

and

unintended

consequences

of

a

paradigm

shift

in

corn

production

practices

Scott

W.

Fausti

1,

*

DepartmentofEconomics,SouthDakotaStateUniversity,Brookings,SD57007-0895,UnitedStates

1.

Trends

in

U.S.

row

crop

production

Thecorn/soybeanmonoculturecroppingsystemhasbecome a dominantfixture inmodern cropproduction inthe U.S. Roughly30%oftotalfieldcropacreagewasdevotedtocornin thepast5years(NASS,2014),whichannuallyequatesto4.6% oftheterrestriallandsurfaceofthecontinentalUnitedStates (U.S.)andrepresented35%oftotalcropprofitsbetween2010 and 2012(NASS,2014).Soybeans arethe next mostwidely plantedcrop(24%oftotalfield cropacreage),butsoybeans producedonly19%oftotalU.S.cropvalue.

In the Midwest Corn Belt, crop production has shifted significantlytoward corn and away from other crops over thelast15years(Table1;Fig.1).Thistrendcoincideswitha significantdeclineinthediversityofagriculturalproduction

systems in the Midwest and the Northern Great Plains (Wallanderetal.,2011;Larsonetal.,2010).Thisshifttoward greater homogeneity in agricultural production systems (landscapesimplification)hasalsotakenplaceinotherregions of the U.S. and Canada (Wiens et al., 2011). Landscape simplificationdoesnotonlyimplyareductioninagricultural productionsystemdiversity,butalsoadecreaseinecosystem biodiversity(e.g.,Purtaufetal.,2005;Meehanetal.,2011).The increasesincornproductionareduetoacombinationofyield productivityincreases(Wallingtonetal.,2012),andregional landusechanges(Johnston,2014;WrightandWimberly,2013; Wallanderetal.,2011;Larsonetal.,2010).However,thefuture of corn yield productivity increases is uncertain. Recent increasesinyieldproductivityattributedtoGMcornvarieties arepartiallyduetoimprovementsinnon-GMgermplasm(Shi et al.,2013). Shietal.arguesthatduetopatentlaws,seed

a

r

t

i

c

l

e

i

n

f

o

Articlehistory:

Availableonline26May2015 Keywords:

Causalrelationships Btcorn

Croprotationpractices Croplandsimplification Ethanolmandate FreedomtoFarmAct RenewableFuelAct GrangerCausality

a

b

s

t

r

a

c

t

IndependentbutsimultaneouslyoccurringchangesinU.S.agriculturalandenergypolicies in conjunction with advances in biotechnology convergedto create an economic and regulatory environmentthatincentivizedcornacreageexpansion. Advancements inBt seedandethanolproductiontechnologiescontributedtoscaleefficiencygainsincornand biofuelproduction.Theseadvancementswereaccompaniedbychangesinmarketforces thatalteredthebalancebetweencornandotheragriculturalcropproduction.Thecausal linkagesamongBtadoption,ethanolproduction,andcornproductionareexploredalong withadiscussionofhowthisshifttowardcornproductiongeneratedunexpectedeconomic andenvironmentalconsequences.Alternative policysolutionstomitigatethenegative consequencesandenhancetheresiliencyofU.S.agriculturearediscussed.

#2015TheAuthor.PublishedbyElsevierLtd.ThisisanopenaccessarticleundertheCC

BY-NC-NDlicense(http://creativecommons.org/licenses/by-nc-nd/4.0/).

*Tel.:+16056884868.

E-mailaddress:Scott.Fausti@sdstate.edu.

1 IwouldliketoacknowledgeDr.JonathanG.Lundgren’scontributiontothismanuscript.Dr.Lundgrenisanentomologistemployedby theUSDAAgriculturalResearchService(ARS).However,theARShasrequiredDr.Lundgrentoremovehisnameasjointfirstauthorfrom thisarticle.Ibelievethisactionraisesaseriousquestionconcerningpolicyneutralitytowardscientificinquiry.

Available

online

at

www.sciencedirect.com

ScienceDirect

journalhomepage:www.elsevier.com/locate/envsci

http://dx.doi.org/10.1016/j.envsci.2015.04.017

1462-9011/# 2015 The Author. Publishedby Elsevier Ltd.This is an open access article underthe CC BY-NC-ND license (http://

(2)

companyresearchdollarshavegoneintoGMtechnologyand notgermplasmproductivityimprovement.Thishas implica-tionsforthefutureofcornyieldproductivityincreases.

Thesetrendsraisethefollowingquestions:(a)whatarethe factorsthatencouragedU.S.agriculturalproductiontomove towardmonoculturecroppingpracticesfocusedoncorn,and (b)what have been the unintended consequences and the unforeseenfutureconsequencesassociatedwiththecurrent U.S.cornproductionsystem?Toaddressthefirstquestion,the causal linkagesbetween state-level Bt corn seed adoption rates,ethanolproductioncapacity,andtheproportionofcrop acres planted to corn are empirically tested. The second questionis addressedthrough asynthesisofthe pertinent literature.Finally,potentiallong-runeconomicand environ-mentalimplicationsofthecurrentsystemarediscussed.

2.

The

convergence

hypothesis

Overthelast20years,theU.S.hasexperiencedashiftinrow cropproductionpractices.Rowcropproducershave moved awayfromarotational-based-multi-cropproductionsystem andtowardamonoculturebased(corn/soybean)production system. It is proposed here that the recent shifts in U.S. agricultureandenergypoliciesprovidedtheopportunityand themotivationfortherapidchangetowardacorn-dominated agriculturalsystem.

2.1. Opportunity

Thepolicythatprovidedthe opportunityforcornexpansion wastheFederalAgricultureImprovementandReformActof 1996(P.L.104-127)(alsoknownasTheFreedomtoFarmAct; FFA). The FFA made two fundamental changes to U.S. agriculturalpolicy:(a)itremovedthelinkagebetweenprices ofagriculturalproductsandincomesupportpayments,and(b) it removed acreage restrictions from cropping decisions (Claassen etal., 2011). Claassenet al.(2010)argue thatthe croppingpatternsthatoccurredafterimplementationofFFA would nothave been possibleunderthe old policy regime becauseFFA‘‘...allowedproducerstorespondmorefreelyto marketsignals,policyincentives,andtechnologychange.’’ 2.2. Motivation

Key biofuel policy initiatives occurred shortly after the introduction ofGMseedtechnologyforcorn and soybeans inthelate1990s.Thesebiofuelpoliciesprovidedthemotivation for the expansionofcorn-based ethanol. Thebiotech crop revolutionthenopenedthedoorforproducerstoadoptacorn/ soybean monocultureproductionsystemwhenbiofuel pro-ductionexpansionresultedinasurgeinderiveddemandfor corn (Fig. 1). These independent but simultaneous events providedthemechanismfortheexpansionofcornproduction that in turn supported a further expansion in ethanol

Table1–ChangesinareaplantedtoprincipalcropsintheCorn-Belt(NASS,2014).a

Crops(plantedacres) Pre-ethanolincentive policy(1996–2000)c

Post-ethanolmandates (2009–2013)c

Changeinareaplanted bycrop(%) Corn 64380.0(35.7) 73640.0(41.5) 14.4 Soybeans 55596.0(30.9) 58028.0(32.7) 4.4 Barley 669.4(0.4) 182.8(0.1) 73.0 Oats 2161.8(1.2) 1166.4(0.7) 46.0 Wheat 23724.0(13.2) 18663.4(10.5) 21.3 Hayb 24370.0(13.5) 20462.0(11.5) 16.0 Othercrops 9253.8(5.1) 5423.0(3.1) 41.4

Totalplantedarea 180155.0(100) 177565.6(100) 1.4

a Dataarefromthefollowingstates:IA,IL,IN,KS,MI,MN,MO,NE,OH,SD,andWI. b Harvestedacres.

c Thousandsofacresplanted(%ofarea).

0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400 0.450 0.500 19 96 19 97 19 98 19 99 20 00 20 01 20 02 20 03 20 04 20 05 20 06 20 07 20 08 20 09 20 10 20 11 20 12 20 13 %hay %corn %Soy %barley %Oat %wheat %hay usage

Fig.1–ChangesintheMidwestCorn-Beltcroppingsystem(proportionoftotalacresplanted)andEthanolProductionUsage ofAnnualU.S.CornCrop.

(3)

production capacity. Thus, a feedback mechanism was established.

TheMidwestistheepicenterofU.S.cornproductionand ethanol production capacity (Lambert et al., 2008; Stewart and Lambert, 2011). The causal linkages outlined in Fig. 2 hypothesizes that federal agricultural and energy policy initiativesindependentlyinfluenced producercropplanting decisionsandindustrialcorn-based-ethanolproduction deci-sions.Theestablishment ofethanolproduction capacityin cornproduction regionsaltered producers’cropproduction decisionstomeettheanticipatedincreaseddemandforcorn. Inturn,producersadoptedGMseedtechnologytofacilitate the expansion of corn acres planted. Expansion of corn production resulted in the intensification of corn acreage planted relative to other crops. Increased corn production motivatedtheethanolindustrytoincreaseplantcapacityin cornproducingareastocaptureadditionaleconomic incen-tives associated with federal biofuel mandates. Thus, it is hypothesized that energy and agricultural policy actions implementedindependentlycreatedafeedbackmechanism

that resulted in rapidexpansion of U.S. corn and ethanol productionfrom2000to2013.

Fig. 2 also suggests a linkage between the unintended consequencesassociatedwiththisfeedbackmechanismfor ecologicalsystemsincornproductionregionsandworldgrain markets. These untended consequences have been widely documentedintheacademicliteratureandarediscussedin Section5.

3.

Ethanol

production,

gm

seed

adoption,

and

market

incentives

3.1. Ethanolproduction

California’s 2003 decisionto replaceMTBE (methyl tertiary-butylether)withethanolpromptedrefinersnationwidetomake arapidconversionfromMTBEtoethanol(EPA,2014).Thisshift inproductionwasacceleratedbythepassageofthe2005Energy PolicyAct(EPAct)andtheEnergyIndependenceandSecurity

Fig.2–U.S.corncroppingsystemsandtheroleofAgandbio-fuelpolicy.TheempiricalresultsreportedinTable3support thecausalrelationshipsdepictedbytheeconomicprocessessection.Theeconomicprocesssection(center)hypothesizes thepresenceofanethanolandcornproductionlinkageandafeedbackmechanism.GrangerCausalitytestprovides statisticalsupportfortheexistenceoftheselinkages.Theestablishmentofethanolplantsincentivizedcornproducersto expandproductionbyadoptingBtseedvarieties.Bttechnologyallowedproducerstoexpandcornacreage.Increasedcorn productionincentivizedethanolcompaniestoexpandcapacity.Risingethanolfuelmandatesallowedtheeconomic feedbackmechanismtointensity.Croppingsystemsimplificationanditsassociateduntendedconsequencesaccelerated asthefeedbackmechanismintensified.

(4)

Act(EISA)of2007.Thesepolicyinitiativesestablishedagoalof blending36billiongallonsofrenewablefuel(ofwhich15billion gallons wouldcome from corn)into gasolineby2022. As a resultofthesepolicyinitiatives,ethanolproductionexpanded rapidlyfrom2.1billiongallonsin2002to13.44billionin2013 (U.S. EnergyInformation Administration(EIA), 2015a,b).The numberofcorn-basedethanolrefineries morethandoubled since2005(95and210refineriesin2005and2014,respectively), and 90% of theserefineries are located inCorn Belt states (RenewableFuelsAssociation,2014;CaiandStiegert,2014). 3.2. GMseedadoption

Theincreasedflexibilityprovidedbythebiotechnology revolu-tion enabled row crop producers to reduce labor input requirementsforcropproductionduringtheplantingseason asaresultofGMseed(Fernandez-CornejoandMcBride,2002). Tomeettheethanol-drivenincreaseddemandforcorn,many farmersabandoned traditionalcrop rotation practices. This shiftwasonlyagronomicallyandeconomicallyfeasiblewith theadoptionofBtcornhybrids.Croprotationsaretraditionally usedtomitigateyieldreductions(rotationeffect)frominsect pests(e.g.,thecornrootworm,Diabroticaspp.;Grayetal.,2009). Btcornreducestheneedforcroprotationforpestmanagement intheshortrun,allowingcorn-on-cornproductionpracticesto maximizeshort-termprofitabilityoffarms.By2014,adoption ofBtcornincreasedtoanaverageof80%ofacresplantedin theCornBelt,matchingtherapidexpansionofcorn-ethanol production(Fig.3).TheproliferationofmodernBttechnology andthelinkagebetweengrainmarketsandenergymarketsvia ethanol(CaiandStiegert,2014;Wallanderetal.,2011)facilitated theexpansionthatoccurredintheU.S.cornproductionsystem. 3.3. Marketincentives

The surge in U.S. corn-based ethanol production changed relativecropprices(Wallanderetal.,2011).Producercropping

decision flexibility in the western Corn Belt increased and became more responsive to relative crop prices after FFA (Claassenetal.,2010).Conversionofgrazinglandtorowcrops inthesestatescoincidedwithhigherthanexpectedreturnto cropsrelativetograzing.Thesestatesexperienceda signifi-cantconversionofgrassland,pasture,andwetlandstorow cropproductionoverthelastdecade(Johnston,2014).

Claassenetal.(2011)estimatethatFederalCropPrograms (commoditysupport,cropinsurance,anddisasterpayments) added8.5%toannualcroprevenuesfrom1998to2007and reducedthefinancialriskassociatedwithconverting grass-landstocropsinNorthandSouthDakota.Thedecouplingof USDAcommodityprogramsfromcropproductiondecisions after the passage of FFA providedproducers an additional economicincentivetoexpandrowcropproduction.

U.S. agriculture and energy policy changes facilitated privatesectoractivities,includingtheexpansionofethanol production capacity;and thetechnologicaladvancementin Ht, Bt, andstacked GMcorn seedvarieties. Advancements in theseareas ofbiotechnology evolved independently but then converged to alter crop production practices in the MidwestandtheNorthernGreatPlains.

4.

Empirical

evidence

in

support

of

the

convergence

hypothesis:

a

test

for

statistical

causality

Theconceptofcausalitywithinatimeseriesframeworkwas introducedbyGranger(1969).Essentially,a‘‘GrangerCausal Relationship’’existsifpastvaluesofXtcanbeusedtobetter

predict current values of Yt. If this is true, then this

relationship is expressedas Xt ‘‘GrangerCauses’’ Yt. Thus,

GrangerCausalityisastatisticalconceptofcausalitythatis basedonprediction.

There are several caveats that influence the degree of statisticalrobustness whenusingGranger’sempirical tech-nique. First, for bilateral causality, both random variables must be stationary or cointegrated. Next, the selection of thelaglengthforthesamplingperiodneedstobecarefully considered.Finally,relevantvariableswhichinfluenceboth XtandYtmaybethesourceofthecausalrelationshipbetween

XtandYt.

Formally, Fig. 2 hypothesizes a causal linkage between ethanol production capacity, the proportion of corn acres planted, and Bt corn seed adoption rates. To test the robustness ofthis proposition,astatistically basedtestfor evidenceofstatisticalcausalityisconducted.Therearefour possibleGrangerCausalityoutcomes betweenXtandYt: (a)

bidirectional(coinciding)causality,(b)Grangernon-causality, and(c)unidirectionalcausality(Xt!YtorXt Yt).

Statisticalanalyseswereconductedonthedatareportedin Table1,usingSAS(2009).Giventhatthedataspan11states over14years,apanelGrangerCausalitytestwasconducted usingamodifiedversionofaSASpooledpaneldataprogram for Granger Causality developed by J. Morrison (2015). Significant modifications to Morrison’s SAS program were madetoovercometheeconometricissuesassociatedwiththe statisticalanalysisassociatedwiththisproject.Modifications include unit root tests, adding an individual equation lag

Fig.3–AnnualdataforacresplantedtoBtcornandcorn ethanolproductionsince2000.Thenonlinearrelationship betweenBtandcornusagerevealsthatannualBtcorn acreagewashighlypredictiveofgraindevotedtocorn ethanol(y=76.04T (1S e(S0.050x))(F1,12=685,P<0.001;

adjustedR2=0.98).

(5)

selectionprocess,an endogeneitytest,and atestforserial correlation.

Forclarityofmathematicalpresentation,the dependent variable is defined as Yt, and the independent variable is

definedasXt.ThepotentialrelationshipbetweenYtandXtis

definedinEqs.(1)and(2).ThedirectionofGrangerCausalityis notassumed.Towardthatend,aVectorAutoregressive(VAR (n))modelisutilizedthatallowsforvaryinglaglengths(Ytj,

Xtk)forYtandXt: Yt¼ Xn j¼1 CjYtjþ Xn k¼1 BkXtkþe1t (1) Xt¼ Xn k¼1 BkXtkþ Xn j¼1 CjYtjþe2t (2)

ThenullhypothesisofXtdoesnotGrangercauseYtcanbe

specifiedas

H1

0:B1¼B2¼...¼Bn¼0 (3)

andthenullhypothesisofYtdoesnotGrangercauseXtcanbe

specifiedas

H2

0:C1¼C2¼...¼Cn¼0 (4)

Summary statistics for the data used in the Granger Causality analysis are provided in Table 2. The sensitivity caveat of the Granger test for lag length is addressed by adopting an optimal VAR lag length selection criteria rule basedontheAIC‘‘goodnessoffit’’statistic.Theoptimallag lengthforthedependentvariablewasselectedfirstandthen the independent variable lag length was determined. The adoptionofanoptimalVARlaglengthruleisconsistentwith thebasiceconomicprincipleofmaximizationofanobjective function.ResidualwhitenoisewasconfirmedusingtheWald– Wolfowitzrun testforserialcorrelation(http://support.sas. com/kb/33/092.html).Anendogeneitytestwasconductedby examiningthecorrelationbetweentheindependent covari-atesandtheresidual.ThenullhypothesisofCorr(X,e)=0was confirmedforallVARmodelspresented.

The common variable caveat was addressed by: (a) incorporatingthesoybean/cornpriceratioasavariable(PBCR) to capture market forces affecting the possible causal relationships,(b)includingstatedummyvariablestoaccount foruniquecharacteristicsofstatesaffectingcausal relation-ships (controlling for fixed effects), (c) including a biofuel policydummyvariablethatequalsonefortheyear2007or later and zero otherwise, and (d) including the selected

common variables inthe lag length selectionprocess. The policyvariable(ethdum)definitionisbasedonAkinfenwaand Qasmi (2014) who demonstrate that a structural break occurredinU.S.ethanolproductiontimeseriesin2007.

The last issue is stationarity and it is addressed by conducting unit root tests (Phillips–Perron).Unit root tests indicatedthepresenceofunitroots inallthreevariablesof interest and PBCR. Therefore, Granger Causality tests were conductedusingfirstdifferencesofthesevariables.Thethree variablesofinterestare;thechangeintheratioofcornacres tototalacresplanted(DCorn),thechangeintheratioofBtcorn acresplantedtototalcornacresplanted(DBt),thechangein ethanolplantcapacity(DEth),andthechangeinthesoybean/ cornpriceratio(DPBCR).Unitrootanalysiswasconductedusing SASAuto-RegprocedureinSAS/ETSVersion9.2(SAS,2009).

The empirical results provide statistical evidence to support the hypotheses graphically depicted inFig. 2of a causalrelationshipconnectingtheincreaseintheproportion ofcornacreageplantedrelativetototalacres,theshareofcorn acresplantedwithBtseed,andethanolproductioncapacityin theCornBeltregion(Table3).

Grangertests(Table3)providestatisticalevidence indicat-ing that changes in ethanol production capacity were influencedbychangesintheproportionofcornacresplanted (P-value=0.009). This suggests that changes in ethanol productioncapacityoccurredinareaswherecornproduction was increasing. The bi-directional statistical relationship betweenDCornandDBt(P-value<0.01)indicatesafeedback mechanismresultingfromproducerssimultaneouslymaking decisionsonhowmanyacresofcorntoplantandwhattype of seed to plant. Finally, there is statistical evidence that the change in ethanol production capacity influenced the producers’ Bt adoption rate decision (P-value<0.001). The Granger results indicate a production system feedback mechanism operating in Midwest corn production areas: DCorn!DEth!DBt$DCorn.

TheVAR parameterestimatesforpotentialconfounding variables(statedummies,ethdum,andDPBCR)areprovidedin Table4.Thestatedummyvariables(MIbase)wereincludedto capture fixed effects due to heterogeneity in agricultural production among states. Given that dependent variables were expressedas first differences, the empirical evidence indicatesthatstateheterogeneitydidnotinfluenceDCorn.The same istruefortheDEth equations(exceptforIAand NE). However,withrespecttotheBtequations,thereare8of11 states with significant coefficients (positive or negative), suggestingthatstatesvariedinhowquicklytheyadoptedBt

Table2–Statelevelsummarystatisticsfor2000to2013.

Variable No.obs. Mean Standarddev. Statelevelmin Statelevelmax

%Cornacresplanteda 154 38.41742 12.03083 12.4801 58.04749

%BTadoptionrateb 154 45.25974 21.39477 6 84

Ethanolproductioncapacity(1000U.S.BBL)c 154 14088.39 17166.82 0 87811.0

PBCRa 154 2.490437 0.292475 2.076412 3.060729

a DatacollectedfromtheNationalAgriculturalStatisticsService(2014). b ERS(2014).

c RFA(2014)andEIA(2015b,c).Productioncapacityfor2013calculatedusingNebraskaEnergyOffice(December2013report)data.NoteEIAdate collectedusingEIAindividualstatedata.

(6)

seedovertime.Thereportedstatisticalevidenceof heteroge-neityisconsistentwiththeERSreportonGMadoptionrates (ERS,2014).

Thepolicy-inducedstructuralshiftintherateofincreasein ethanolproductionandcapacityafter2006wasfoundtohave astatisticallysignificantandpositiveeffectonDCornandDBt. Themagnitudesofthesechangesaresurprising.Relativeto thepre-2007period,thepolicyinducedestimatedrangeforthe changeinDCornisbetween0.69and0.88%(P-value<0.04and <0.02,respectively) basedon the two DCorn VAR equation estimates (Table 4). However, for DBt, the policy induced estimatedchangeincornacresplantedwithBtisbetween6.8 and 9.01% (P-value <0.02 and <0.01, respectively). This suggests the acceleration in ethanol production capacity inducedbyU.S.energypolicyhadamuchgreatereffecton Bt adoption rates than DCorn. This finding supports the suppositionposedearlierthatGMseedtechnologyandbiofuel technology may have begun as independent technological

forcesinU.S.agriculture,butU.S.policyinitiativesprovided aneconomicenvironmentthatcreatedafeedbackmechanism that linked these two technologies. Thus, the empirical evidence suggests that U.S. biofuel energy policy is a key contributing factor in the rapid adoption of Bt corn seed technologyintheU.S.cornproductionsystem.

Thelastissuetobeaddressediswhethermicroeconomic forcesexertedthroughthemarketplaceinfluenceda produ-cer’sdecisiontoplantcorn.Thesoybean/cornpriceratiowas includedtoaddressthecaveatsassociatedwithconducting GrangerCausalitytests.Marketpricessendeconomicsignals to producers on market demand and supply conditions. Soybeans and corn are complementary members in the mono-cropping systemcurrently gaining popularityamong U.S.rowcropproducers.Historically,thelongrunaverageof the soybean/corn price ratio has been 2.52 as reported by Zulauf (2013). Ratiosexceedingthehistoricallevel signalto producerstoplantmoresoybeansandlesscorn.

Table4–VARanalysisresultsforGrangerCausalitymodels.

Models(Y/X)a DCorn/DBt DBt/DCorn DCorn/DEth DEth/DCorn DEth/DBt DBt/DEth

AICCstat 614 515 621 1674 1676 520 Rsq. 0.44 0.61 0.42 0.58 0.51 0.60 No.ofOBS 143 88 143 88 88 88 Variablesb DPBCR 1.01* NS 1.23** NS NS NS Ethdum 0.69** 6.79** 0.88** NS NS 9.01*** IA NS S NS S S S IL NS NS NS NS NS NS IN NS S NS NS NS S KS NS S NS NS NS NS MI-Base NA NA NA NA NA NA MN NS S NS NS NS NS MO NS S NS NS NS S NE NS S NS NS S NS OH NS NS NS NS NS NS SD NS S NS NS NS S WI NS S NS NS NS NS ARCH0 NA NA NA *** *** NA ARCH1–3 NA NA NA NS NS NA RunstestP-Vc 0.21 0.93 0.21 0.74 0.44 0.93

Note: S, NS, and NA denote;statistically significantat the 10%level, not statistically significantat the10% level, and not applicable, respectively.

a Delta(D)denotesfirstdifferenceofvariable(XtXt1).

b Statisticalsignificancelevelsof0.1,0.05,and0.01aredenotedbythefollowingasterisks,*,**,and***,respectively.Valuesreportedare estimatedVARcoefficients.

c Wald–Wolfowitzruntestforserialcorrelation.Ho:whitenoise.P-VdenotesP-values.

Table3–VAR(optimal)model:directionofGrangerCausality.

Ya Xa No.ofobs.b XGrangercausesY Ylaglength Xlaglength P-valueofWaldx2testc

DCorn DBt 143 Yes 1 1 0.001

DBt DCorn 88 Yes 6 4 0.001

DCorn DEth 143 no 1 1 0.194

DEthd DCorn 88 Yes 6 5 0.009

DEthd DBt 88 No 6 1 0.26

DBT DEth 88 Yes 6 4 0.001

a Delta(D)denotesfirstdifferenceofvariable(Xt Xt1).

b Datasetcontained11statesand14timeperiodsforatotalof154observations. c GrangerasymptoticequivalencyFtest.

d EthanolproductionVARequationhadtobecorrectedforheteroscedasticityusinganARCH(1)correction.Normalitytestacceptedatthe5% level.

(7)

TheDPBCR coefficientsigns estimated inthe DCorn/DBt andDCorn/DEthVARequationsarenegativeandstatistically significant(P-value<0.09and<0.04,respectively),astheory predicts.Furthermore,thepriceratiowasstatistically insig-nificant for the Bt and ethanol production capacity VAR equations(Table 4), whichisnotsurprising fortheethanol VAR equations. However, for the Bt VAR equations, this suggestthattherelativemarketvaluationofsoybeanstocorn didnotplayaroleinBtseedadoption,andfurtherindicates thatBtseedcostdidnotplayasignificantroleinthedecision toplantcornrelativesoybeans.GiventhatbothGMsoybeans andGMcornhavetechnologyfees,itappearsthatseedcost isnottheprimaryfactoraffectingthedecisiontoplantcorn orsoybeans,ratherrelativepriceplaystheprimaryrole.

5.

Consequences

of

the

corn-ethanol

feedback

mechanism

Therapidincreaseinthederiveddemandforcorn,asaresult of expanding ethanol production, has shifted corn usage (Fig.4)awayfromitstraditionalroleasafoodsourcefor(1) animalproduction, and(2) human consumption(Anderson etal., 2008;Wienset al.,2011).Ethanolexpansionhashad widespreadeffectsontheworldeconomyandpricevolatility inthegrainmarkets,andcontinuedrelianceoncorn-based ethanolwilllikelyresultin(1)higherandmorevolatilefood prices,and(2)afurtherintensificationofcornproductionin theU.S.

Theeconomicandsocialconsequencesofthecorn-ethanol feedbackmechanismhavebeendiscussedintheeconomics literature. Wright(2014) estimates that corn-based ethanol expansionhascausedan$800billionincreaseinagricultural landpricesandatransferofwealthfromconsumerstoland owners.Bellemare(2015)empiricallylinkstherecentspikesin commodity grain prices to increased social unrest in the developingworld.

Thedisproportionateincreaseincornproductionrelative to other crops has created unintended ecological conse-quencesasaresultoflandscapesimplificationandincreased environmentalpollution (Hill etal., 2006). Highcropprices

(driven by the corn-ethanol feedback mechanism) have incentivizedthereplacementofnaturalareas(wetlandsand grasslands)inhighlycroppedregions(Johnston,2014;Wright andWimberly,2013).Reductionsinbiodiversityassociatedwith cornintensificationreducetheecosystemservicesprovidedby healthy biological communities (Landis et al., 2008), and challengewildlifeconservation(Meehanetal.,2010).

Sciencehasalsoestablishedthatpollution(fertilizersand pesticideuse)associatedwithcornproductionhasimportant environmentalconsequences.Severalstudieshavelinkedthe increasednitrogenlevelsinthelowerMississippitochanges incornandsoyproductionpracticesintheMidwest(Donner and Kucharik, 2008). Increased nitrogen levels have been linked to hypoxia and a dead zone in the Gulfof Mexico (Turneretal.,2007).Larsonetal.(2010)providesevidencefrom asimulationanalysispredictinganincreaseinfertilizerand chemicaluse,andadecreaseinsoilcarbonstocksasaresult of increased corn production. Fausti et al. (2012) provide evidence of a positive association between increased per acre insecticide usage and increased corn acreageplanted atthecountylevelinSouthDakota.

When land-clearingisconsidered,greenhouse gas emis-sionsassociatedwithcorn-basedethanolproductionareoften greaterthanthosecreatedbyburningfossilfuels,creatinga netcarbondebtthatwilltakegenerationstorepay(Fargione et al., 2008). Finally, long-term projections indicate that cellulosicethanolproductionbasedoncornstalksor plant-basedfeedstockalternativeswillnotcurethisproblem(Liska etal.,2014;Searchingeretal.,2008;PimentelandPatzek,2005). Furthermore,studiesonalternative energyusingbiofuel feed-stockshavedemonstratedthatcropssuchassugarcane and palm oil have more favorable conversion ratios with respecttonetenergy;e.g.Wiensetal.(2011).However,U.S. tradepolicywithrespecttotheimportquotaonsugar,U.S. ethanolsubsidies,andthehighproductioncostofcommercial cellulosicethanolprecludethecurrentcommercialviabilityof thesefeedstockalternativestocorn-basedethanolproduction intheU.S.Specifically,enzymesrequiredforcellulosicethanol havearelativeproductioncostbasis20–40timesthatofcorn based ethanol(Sainz, 2011).While cellulosicethanol hasa smaller carbonfootprint than cornbased ethanol,it isnot economically competitive without additional technological advances.

6.

Cracks

in

the

ice:

the

brittleness

of

corn

dominated

agriculture?

Thecurrentcorn-basedethanolproductionsystemis depen-dentonthethreeeconomicforces:(a)U.S.agriculturalpolicy, (b)U.S. energypolicy, and (c)technologyinnovationinthe areasofbiofuelproductionandGMseeddevelopment.Ifany oneofthesethreesupportmechanismsbeginstofalter,the systemwillbegintobreakdown.

Consequences to food production will bedependent on whichsupportmechanismbecomesunstable.Recent devel-opments in public sentiment, federal biofuel policy, and vulnerabilities to current technologies may forewarn of impending challenges for corn production. For instance, concern continuesto be raisedabout the efficacy and the

Fig.4–U.S.CornCropUsages. Source:ERS(2015).

(8)

longrunsustainabilityoftheGMseedtechnologytomitigate pestdamageandsustainadvancementsinyieldproductivity. IntheU.S.,widespreadadoptionofBtcorn technologyand abandonment of crop rotations (Claassen et al., 2011; Wallander et al., 2011)and traditional non-Btcorn refuges (Onstadetal.,2011)havehelpedtoselectforBt-resistancein the western corn rootworm (Gassmann et al., 2014). Pest resistance,ifnotovercomewithnewtechnological advance-mentsinGMseedtechnology,willforceproducerstoreturnto a traditional rotational cropping system. In turn, this will increaseeconomicstressongrainmarketsduetoareduction inlandsuitableforgrowingcorn inanygivenyear.Ifyield productivitygainsfailtomaterialize,thenincreasingtotalU.S. productioncapabilitywillrequireextensivegrowthinplanting areas.Inturn,extensiveexpansionofcornacreswillleadto furthersimplificationofrurallandscapes.

The2012droughtintheMidwestrevealedaseriousriskto marketsdependentuponU.S.cornproduction.Thecurrent U.S.renewablefuelstandardmandatesthat15billiongallons ofethanol(roughly5.4billionbushels ofcornannually) be blendedintofuelin2015andthenremainconstantthrough 2022.RecentU.S.annualcornproductionrangedfrom10.8to 13.9 billion bushels per year. Given the current ethanol mandateandrecentU.S.cornproductionhistory;thisimplies thatapproximately38–50%oftheU.S.cropwillbedevotedto ethanolonanannualbasisintothefuture.In2012,drought reducedU.S.cornproductionto10.8billionbushels,andthe resultinglimitedcornsupply(exacerbatedbythemandateto blend minimum levels of corn-based ethanol in gasoline) increasedcornpricestorecordhighs($7.63perbu.)inAugust of2012.SuddenspikesinU.S.grainpricesincreasevolatilityin world grain markets, resulting in economic and social instabilityaroundtheglobe(Wright,2014).

The recent dramatic decline in crude oil prices, the eliminationoftheethanolexportsubsidyandlowcornprices willlikelyresultisdownwardpressureonbothethanoland cornproductionintheshortrun.Thiswillresultinproducers shiftingproductiontowardsoybeans,wheat,andothercrops throughout the region. However, it is unlikely that world petroleumpriceswillremaindepressed.Thus,theimplication oftheethanolmandateiscontinuedincreasedpricevolatility inworldgrainmarkets.Combiningmarketissuesassociated with the current corn/ethanol production system withthe recentchallenges associated withextremeweather events, public sentiment, and pest resistance arguably reveal the brittlenessofU.S.rowcropproductionpractices.

7.

Potential

solutions

to

the

problem

Therapidityandoutcomeoftheshiftawayfromcurrent corn-dominatedcrop production patterns could have important perturbations for agricultural markets if the latter are not preparedtoadapt.Asshown inFig.2,thecurrentstatusof corn-dominatedagriculturewascreatedlargelybyU.S.policy, andthusthesolutiontotheproblemwilllikelyhavetobeat thislevel.

Solution1.Restrategizetheethanolmandate.Afirstpotential solutiontoreducingagriculture’srelianceoncornistolinkthe ethanolmandatetocropproductionlevelsbymandatingthat

a maximum percentage of the corn crop (rather than a mandated fixed ethanol production level) be devoted to ethanolproduction.Inadditiontoreducingtheperturbations that ethanol consumption has on corn prices in years of low production, this would provide incentives to ethanol producers toincrease theirefficiency inextracting ethanol fromcorngrain.Inyearsofbelowaveragecornproduction, aceilingoncornusageforethanolwillreducepricevolatility by reducing uncertainty surrounding the level of residual corn supply that will be available for human and animal consumption.

Solution 2.Growmorecornonlessland. Grain-price-driven highlandpricesrestrictthesizeoffarms,sohigheryieldsare neededtocontinueincreasingethanolfeedstock.Germplasm development (not biotechnology) has historically been the source of increasing corn yields. But investment in corn germplasmresearchhasbeensupplantedbybiotechnology, which hasled toadiminishingrate ofyield increasesand couldleadtoatroughingrainyieldadvancements(Shietal., 2013).Also,cropproductionisinherentlytiedtosoilhealth, andprolongedoveruseoftillageandalackofbiodiversityin farmingoperationshavereducedsoilnutrientstatusandits capacityforsupportingoptimalyields(Lehmanetal.,2015). Investmentingermplasmdevelopmentandsoilhealthand conservationshouldbeprioritized.

Solution3.Incentivizeinnovationincropandethanolproduction. Even if the 15 billion gallon ceilingis lifted, corn willnot producesufficientethanoltomeetfutureEISAmandateof36 billiongallons.Thus,majorresearchinitiativesinto develop-ingothercellulosicfeedstock,andincreasingtheefficiencyof ethanol production from these feedstock alternatives is needed to reduce our reliance on corn-based ethanol and increase the resiliency of our biofuel production system. Finally,as cropproductionisintensified toproducefuelin additiontofoodandfiber,theimplicationsforthesesocietal andeconomicchangesonspeciesconservationonandaround farmsbecomesimperative.Therearenumerouswaystoalter ourcurrentcropproductionsystemsinwaysthatconserve biodiversity (e.g., reduce soil disturbance,diversify in-field plant communities through rotations and ground covers, strategizeuncroppedareasofthefieldtopromotebiodiversity, etc.). Thus, promoting research to make these agronomic strategiesscalable,transferable,andpredictablewillhelpalter farmerbehavior.

8.

Summary

The convergencehypothesis suggeststhat U.S.agricultural and energy policy induced a causal relationship running betweenethanolproductioncapacitydecisionsinhighcorn production states, tothe Bt adoptiondecision byrowcrop producers,totheproducer’sdecisiononhowmanyacresof corn acrestoplantinthe MidwestCornBelt andNorthern GreatPlainsregion(2000–2013).Empiricalevidencesupporting thehypothesisisprovidedintheformofaGrangerCausality test.VARanalysisalsoprovidesempiricalevidencethatU.S. biofuel energy policyhas been acontributing factor ofthe rapidadoption ofBtcornseedtechnologybythe U.S.corn productionsystem.

(9)

Acknowledgements

TheSouthDakotaAgriculturalExperimentStationprovided partialsupportforthisresearchactivity.Theauthorswishto thankDr.EvertVanderSluisforhelpcommentsandedits.I wouldliketothankMr.AlanCarter,seniorsystems program-meratSDSUforhishelpwithdevelopingtheSAS program-mingcode.Theauthorsalsowishtoacknowledgethehelpful suggestionsmadebytheanonymous reviewer.Inaddition, the views expressed in this manuscript do not represent official opinions held by the USDA, and mention of any proprietaryproductsdoesnotconstituteendorsementbythe USDA. Any additional errors are the responsibility of the author.

r

e

f

e

r

e

n

c

e

s

Akinfenwa,S.O.,Qasmi,B.A.,2014.Ethanol,theagricultural economy,andruralincomesintheUnitedStates:abivariate econometricapproach.Agric.Resour.Econ.Rev.43(2). Anderson,D.,Anderson,J.D.,Sawyer,J.,2008.Impactofthe

ethanolboomonlivestockanddairyindustries:whatare theygoingtoeat? J.Agric.Appl.Econ.40(2) 573–579. Bellemare,M.F.,2015.Risingfoodprices,foodpricevolatility,

andsocialunrest.Am.J.Agric.Econ.97(1) 1–21.

Cai,X.,Stiegert,K.W.,2014.Marketanalysisofethanolcapacity. Int.FoodAgribus.Manage.Rev.17,83.

Claassen,R.,Carriazo,F.,Ueda,K.,2010.GrasslandConversion forCropProductionintheUnitedStates:DefiningIndicators forPolicyAnalysis.OECDAgri-EnvironmentalIndicators: LessonsLearnedandFutureDirections.http://www.oecd. org/tad/sustainable-agriculture/44807867.pdf.

Claassen,R.,Carriazo,F.,Cooper,J.C.,Hellerstein,D.,Ueda,K., 2011.GrasslandtoCroplandConversionintheNorthern Plains:TheRoleofCropInsurance,Commodity,andDisaster Programs.U.S.DepartmentofAgriculture,E.R.S.,

Washington,DC.

Donner,S.D.,Kucharik,C.J.,2008.Corn-basedethanol

productioncompromisesgoalofreducingnitrogenexportby theMississippiRiver.Proc.Natl.Acad.Sci.U.S.A.105(11) 4513–4518.

EconomicResearchService,2014.GeneticallyEngineeredVarieties ofCorn,UplandCotton,andSoybeans,byStateandforthe UnitedStates,2000–13.http://www.ers.usda.gov/datafiles/ Adoption_of_Genetically_Engineered_Crops_in_the_US/ alltables.xls(datasetaccessedJanuary).

EconomicResearchService,2015.CornBackground.http:// www.ers.usda.gov/topics/crops/corn/background.aspx (accessed15.01.15).

EnvironmentalProtectionAgency,2014.WebPage:http:// www.epa.gov/mtbe/gas.htm.(accessed15.01.14).

Fargione,J.,Hill,J.,Tilman,D.,Polasky,S.,Hawthorne,P.,2008. Landclearingandthebiofuelcarbondebt.Science319(5867) 1235–1238.

Fausti,S.W.,McDonald,T.M.,Lundgren,J.G.,Li,J.,Keating,A.R., Catangui,M.A.,2012.Insecticideuseandcropselectionin regionswithhighGMadoptionrates.Renew.Agric.Food Syst.27,295–304.

Fernandez-Cornejo,J.,McBride,W.D.,2002.Adoptionof bioengineeredcrops.ERSAgriculturalEconomicReportNo. AER810,Washington,DC.

Gassmann,A.J.,Petzold-Maxwell,J.L.,Clifton,E.H.,Dunbar, M.W.,Hoffmann,A.M.,Ingber,D.A.,Keweshan,R.S.,2014. Field-evolvedresistancebywesterncornrootwormto

multipleBacillusthuringiensistoxinsintransgenicmaize.In: Proc.Natl.Acad.Sci.U.S.A.http://dx.doi.org/10.1073/pnas. 1317179111.

Granger,C.W.J.,1969.Investigatingcausalrelationsby econometricmodelsandcross-spectralmethods. Econometrica37,424–438.

Gray,M.E.,Sappington,T.W.,Miller,N.J.,Moeser,J.,Bohn,M.O., 2009.Adaptationandinvasivenessofwesterncorn rootworm:intensifyingresearchonaworseningpest.Annu. Rev.Entomol.54,303–321.

Hill,J.,Nelson,E.,Tilman,D.,Polasky,S.,Tiffany,D.,2006. Environmental,economic,andenergeticcostsandbenefits ofbiodieselandethanolbiofuels.Proc.Natl.Acad.Sci.U.S. A.103(30) 11206–11210.

Johnston,C.A.,2014.Agriculturalexpansion:landuseshell gameintheU.S.NorthernPlains.Landsc.Ecol.29,81–95. Lambert,D.M.,Wilcox,M.,English,A.,Stewart,L.,2008.Ethanol

plantlocationdeterminantsandcountycomparative advantage.J.Agric.Appl.Econ.40(1) 117.

Landis,D.A.,Gardiner,M.M.,vanderWerf,W.,Swinton,S.M., 2008.Increasingcornforbiofuelproductionreduces biocontrolservicesinagriculturallandscapes.Proc.Natl. Acad.Sci.U.S.A.105,20552–20557.

Larson,J.A.,English,B.C.,Ugarte,D.D.L.T.,Menard,R.J., Hellwinckel,C.M.,West,T.O.,2010.Economicand

environmentalimpactsofthecorngrainethanolindustryon theUnitedStatesagriculturalsector.J.SoilWaterConserv. 65(5) 267–279.

Lehman,R.M.,Cambardella,C.A.,Stott,D.E.,Acosta-Martinez, V.,Manter,D.K.,Buyer,J.S.,Karlen,D.L.,2015.

Understandingandenhancingsoilbiologicalhealth:the solutionforreversingsoildegradation.Sustainability7(1) 988–1027.

Liska,A.J.,Yang,H.,Milner,M.,Goddard,S.,Blanco-Canqui,H., Pelton,M.P.,Fang,X.X.,Zhu,H.,Suyker,A.E.,2014.Biofuels fromcropresiduecanreducesoilcarbonandincreaseCO2 emissions.Nat.Clim.Change4,398–401.

Meehan,T.D.,Hurlbert,A.H.,Gratton,C.,2010.Bird

communitiesinfuturebioenergylandscapesoftheUpper Midwest.Proc.Natl.Acad.Sci.U.S.A.107,18533–18538. Meehan,T.D.,Werling,B.P.,Landis,D.A.,Gratton,C.,2011.

Agriculturallandscapesimplificationandinsecticideusein theMidwesternUnitedStates.Proc.Natl.Acad.Sci.U.S.A. 108(28) 11500–11505.

Morrison,J.,2015.BivariateGrangerCausality.Welcometo ForecastingSolutions.http://www.forecastingsolutions. com/index.html(accessedJanuary15).

NationalAgriculturalStatisticsService,2014.USDAEconomics, StatisticsandMarketInformationSystem,AlbertR.Mann Library.CornellUniversity.WebSite:http://usda.mannlib. cornell.edu/MannUsda/viewDocumentInfo.do?documentID= 1000(accessed15.01.14).

NebraskaEnergyOffice,2014.EthanolProductionbyState: Archive.http://www.neo.ne.gov/statshtml/121_archive.htm (accessed15.01.15).

Onstad,D.W.,Mitchell,P.D.,Hurley,T.M.,Lundgren,J.G.,Porter, R.P.,Krupke,C.H.,Spencer,J.L.,DiFonzo,C.D.,Baute,T.S., Hellmich,R.L.,Buschmann,L.L.,Hutchison,W.D.,Tooker, J.F.,2011.Seedsofchange:cornseedmixturesforresistance managementandIPM.J.Econ.Entomol.104,343–352. Pimentel,D.,Patzek,T.W.,2005.Ethanolproductionusingcorn,

switchgrass,andwood;biodieselproductionusingsoybean andsunflower.Nat.Resour.Res.14(1) 65–76.

Purtauf,T.,Dauber,J.,Wolters,V.,2005.Theresponseof carabidstolandscapesimplificationdiffersbetweentrophic groups.Oecologia142,458–464.

RenewableFuelsAssociation,2014.AnnualIndustryOutlook: VariousYears.WebPage:http://www.ethanolrfa.org/ (accessed15.01.14).

(10)

Sainz,M.B.,2011.Commercialcellulosicethanol:theroleof plant-expressedenzymes.Biofuels237–264.

SASInstitute,2009.SAS/ETS1

,Version9.2.SASInstitute,Inc., Cary,NC.

Searchinger,T.,Heimlich,R.,Houghton,R.A.,Dong,F.,Elobeid,A., Fabiosa,J.,Tokgoz,S.,Hayes,D.,Yau,T.-H.,2008.UseofU.S. croplandsforbiofuelsincreasesgreenhousegasesthrough emissionsfromland-usechange.Science319,1238–1240. Shi,G.,Chavas,J.-P.,Lauer,J.,Nolan,E.,2013.Ananalysisof

selectivityintheproductivityevaluationofbiotechnology: anapplicationtocorn.Am.J.Agric.Econ.95,739–754. Stewart,L.,Lambert,D.M.,2011.Spatialheterogeneityoffactors

determiningethanolproductionsiteselectionintheUS, 2000–2007.BiomassBioenergy35(3) 1273–1285. Turner,R.,Eugene,Rabalais,N.N.,Scavia,D.,McIsaac,G.F.,

Nassauer,J.I.,Santelmann,M.V.,2007.CornBeltlandscapes andhypoxiaoftheGulfofMexico.FromtheCornBelttothe Gulf.10–27.

UnitedStatesDepartmentofAgriculture.NationalAgricultural StatisticsServices,April2014.CropProductionHistorical TrackRecords.http://usda.mannlib.cornell.edu/MannUsda/ viewDocumentInfo.do?documentID=1593(accessed 24.01.15),inpress.

UnitedStatesEnergyInformationAdministration,2015a. AlternativeFuelsDataCenterhttp://www.afdc.energy.gov/ fuels/ethanol_production.html(accessed24.01.15),inpress. UnitedStatesEnergyInformationAdministration,2015b.

Renewable&AlternativeFuels.http://www.eia.gov/

renewable/andhttp://www.eia.gov/state/seds/ seds-data-complete.cfm?sid=SD&CFID=19640954& CFTOKEN=a80b1b5164f2d0a-0C32E0B6-25B3-1C83-54B8C12366F319C3&jsessionid=8430c73fd04f3 a3383642b637a3b65542818/(accessed24.01.15) UnitedStatesEnergyInformationAdministration,2015c.

Renewable&AlternativeFuels:‘‘StateEnergyDataSystem ProductionEstimatesTechnicalNotesFor1960–2012 Estimates.’’ http://www.eia.gov/state/seds/seds-technical-notes-complete.cfm#Production(accessed21.04.15),inpress. Wallander,S.,Claassen,R.,Nickerson,C.,2011.Theethanol

decade:anexpansionofU.S.cornproduction,2000–2009. USDA-ERS,16.

Wallington,T.J.,Anderson,J.E.,Mueller,S.A.,KolinskiMorris,E., Winkler,S.L.,Ginder,J.M.,Nielsen,O.J.,2012.Cornethanol production,foodexports,andindirectlandusechange. Environ.Sci.Technol.46(11) 6379–6384.

Wiens,J.,Fargione,J.,Hill,J.,2011.Biofuelsandbiodiversity. Ecol.Appl.21(4) 1085–1095.

Wright,B.,2014.Globalbiofuels:keytothepuzzleofgrain marketbehavior.J.Econ.Perspect.28,73–98.

Wright,C.K.,Wimberly,M.C.,2013.Recentlandusechangein theWesternCornBeltthreatensgrasslandsandwetlands. Proc.Natl.Acad.Sci.U.S.A.110,4134–4319.

Zulauf,Carl,2013.Soybean–CornPriceRatioSince1975. Farmdocdaily,UniversityofIllinois.http://farmdocdaily. illinois.edu/2013/09/soybean-corn-price-ratios-since-1975. html(accessedJanuary2015).

Figure

Fig. 1 – Changes in the Midwest Corn-Belt cropping system (proportion of total acres planted) and Ethanol Production Usage of Annual U.S
Fig. 1 – Changes in the Midwest Corn-Belt cropping system (proportion of total acres planted) and Ethanol Production Usage of Annual U.S p.2
Fig. 2 also suggests a linkage between the unintended consequences associated with this feedback mechanism for ecological systems in corn production regions and world grain markets
Fig. 2 also suggests a linkage between the unintended consequences associated with this feedback mechanism for ecological systems in corn production regions and world grain markets p.3
Table 2 – State level summary statistics for 2000 to 2013.

Table 2

– State level summary statistics for 2000 to 2013. p.5
Table 3 – VAR (optimal) model: direction of Granger Causality.

Table 3

– VAR (optimal) model: direction of Granger Causality. p.6
Table 4 – VAR analysis results for Granger Causality models.

Table 4

– VAR analysis results for Granger Causality models. p.6
Related subjects :