第29章 《佃農理論》英語原著 (23)

3.Also,Buckfoundthattherentalpercentagewashigherwhenthelandownerprovidedpartofthefarminginputs（i.e.,thetenantcostwaslower）：

Thepercentageoftotalreceiptsforthelandlordvariesfrom24.6percent,……wheresmallrentsaredemandedto66.6percent,……wherethecroppersystemprevailsandwherethelandlordfurnisheseverythingbutlaborandroutinemanagement.[7]

Similarly,accordingtoanothersurveyconductedbyChing-MohCheninfourprovinces（China,1934）,anaveragerentalpercent-ageof55.98wasfoundintenantfarmswherelandownerspro-videdseeds,fertilizers,andbullocks,asparedtoanaverageof46.37percentwhenthetenantsprovidedthesenonlandin-puts.[8]

Furthermore,giventheproductionfunction,thelandspacerentedtoeachtenantdependsuponthefertilityoflandandthetenant'salternativeearning.Specifically,thefarmsizepertenantfamilywillbesmallerif（a）thelandismorefertileor（b）theten-ant'salternativeearningisloWer.[9]Again,supportingevidenceisstrong:

1.InKorea,theaveragefarmsizeoveraten-yearperiod（1929through1938）was0.58choforpaddy-fieldfarmsand0.97chofordrylandfarms.Inthesameperiodandlocation,thepricesofpaddyfieldsweremorethantwoandone-halftimesashighasthoseofdryfields,confirmingthatpaddyfieldsweregenerallymorefertile.[10]

2.AccordingtoanotherinvestigationconductedbyBuck,whichcovers16,786samplefarmsin168localitiesinChina（1929-33）,wefind（a）theaveragefarmsizeinthemorefertilericeregionis3.09acres,asparedtoanaveragesizeinthelessfertilewheatregionof5.63acres;and（b）oftheseventypesofcropfieldslisted,thesmallestaveragesizeoccursindouble-croppingricefields（2.37acres）,whicharegenerallymostfertile.[11]

Turningtothetax-equivalentapproachtoanalysisofsharetenancy,andreferringagaintofigure5,wemaywellask:IftheassertedequilibriumatAwerevalid,whatwouldweobserve?Thefollowingisimplied:

1.Wewouldobservelowerratiosoflaborandotherinputstolandintenantfarmsthaninfarmsunderownercultivationorinfarmscultivatedbyhiredfarmhands.Italsoimpliesthathectareyieldsintenantfarmswouldbelowerthaninownerfarms.ButasBuckobserved（China,1921—25）：

Contrarytotheprevailingopinionthattenantsdonotfarmaswellasowners,aclassificationaccordingtoyieldsbydifferenttypesoftenureshowsnosignificantvariationinyieldsformostlocalities,andforthefewinwhichadifferencedoesoccur,itisinfavorofthetenantorpartownerasoftenasfortheowner……Insomeplaces,even,itisevidentthatthetenantsfarmbetterthantheowners.[12]

Buck'sdatashowthefollowingcropindexesperacre:Ownerfarms,100and101；part-ownerfarms,99and101；andtenantfarms,103and104.[13]Thirtyyearslater,makingnoreferencetoBuck,JamesO.Brayechoedthesameobservation:

UnderdevelopedcountriesintentontechnicalprogressinagricultureshouldrecognizethefactthatmuchofthemostimpressivegaininagriculturalproductivityoccurredinareasoftheUnitedStateswherethedominantformoflandtenureisasharerentalarrangementbetweenlandlordandtenant.[14]

Withrespecttotheintensitiesoffarming,wefindJapanesedata（1932-38）showingthattheaveragelandholdingperownerfarmeris2.22tan,asparedto2.10tanperpart-ownerfarmerand1.93tanpertenantfarmer,[15]alsodenyingtheimplicationofthetaxapproach.Thehigherlabor-landratiointenantfarmscanbeexplainedbythesomewhathigherproportionofpaddyfieldsundertenancy.

2.Wewouldobserve,sincetherentalearningislowerwithequilibriumatA（underthetaxapproach）,thatthemarketvalueoflandundertenantcultivationislessthanthevalueoflandunderownercultivation.Again,evidenceclearlydeniesthisimplication.Wefindthatthevaluesoflanddifferaccordingtofertilitygradesandlocations,andthat"thevalueoflandforthethreetypesoftenure（owner,part-owner,andtenant）inmostcasesvariesonlyafewdollars."[16]Also,withequilibriumatA,wewouldobservethattheactualrentreceivedbythelandownerislowerundersharetenancythanunderafixedrent（rentperacre）.AsisshowninAppendixB,however,sharerentisgenerallyslightlyhigherthancroprent.[17]Thisslightdifference,asIshallindicatelater,mightbeexplainedasapaymenttothelandownerforhis"risk"bearingunderasharecontract.

3.Wewouldobserve,asimpliedbythetaxapproach,thatahigherrentalpercentage-withthe

（1-r）curveshiftingdownwardinfigure5—wouldbeassociatedwithlessfarminginputsperunitoflandintenantfarms.Wefindevidencetothecontrary:Tenantfarmswithhigherrentalpercentagesusuallyalsodisplayhigherlabor-landratios.[18]Thisisdueeithertotheland'sbeingmorefertileortothetenants'alternativeearnings'beinglower,aconditionimpliedbyourtheoryofsharetenancy.

4.Finally,ifthetax-equivalentapproachwerecorrect,wewouldobservethatmost,ifnotall,sharetenantswouldrentlandfromseverallandlords.Butthisisrarelythecase.TakeTaiwan,forexample.In1949thereexistedonly1.24leasespertenantfamily.[19]Thisextra.24leasecaneasilybeexplainedbytheuseofmarginalplotsasdiscussedinchapter2.Ontheotherhand,landownersissuingoneortwohundredleaseswereregardedasmon.[20]