Itisusefultodistinguishtwotypesofcontractualrenegotiation（revision）,thoughattimesonerelatestotheother.Thestipulatedtermsinanytenurecontractinessencespecifytwothings:（a）thestateofresourceuse,orallocation,mutuallyagreeduponbythecontractingparties,and（b）thecontracteddistributionofinefortheparties.Torevise（a）throughrenegotiationformoreefficientresourceusemaybenefitallpartiestoacontract,thatis,allpartiesmaygainorloseless.Torevise（b）,however,onepartymustlose.
Considerthecontractualrenegotiationwhichentailsmainlyareallocationofresources;forexample,changesinrelativeproductpriceswhichcallforshiftstodifferentcrops,orinnovationswhichcallfortheadoptionofnewseedsornewmethodsofcultivation.Renegotiationsofthistypearelargelyconfinedtosharecontracts,sinceunderfixedrentsthetenantsarelefttomaketheirowndecisionsonresourceuseexceptimprovementsinlandandmaintenanceofthelandowner'sassets.Inprinciple,sinceallcontractingpartiesexpecttobenefitfromtherevision,renegotiationcantakeplaceatanytimeandleaseterminationbeesunnecessary.Butdifferentindividualknowledgeofthemarketmaygiverisetodifferenceinopinionsastowhethertherevisionisdesirable.Arelativelyshortleasedurationisaconvenientdevicewhichallowsresourcereallocationintheeventofunsuccessfulrenegotiation.[9]This,togetherwiththemoreplicatedcontractualenforcementrequiredforsharecontracts,explainswhydurationsofshareleasesaregenerallyshorterthanthoseoffixedrent.[10]
Considerfurtherthecontractualrenegotiationwhichentailstherevisionofinedistribution,whenonepartygainsattheexpenseoftheother.Itappliestofixedandsharecontractsalike.Resourceallocationmayalsobeaffected.Forexample,changesinrelativeassetpricesofthecontractedresources,acash-rentcontractwithunanticipatedinflation,ordecisionerrorsmadeintheinitialcontract-whichcallforarevisionoftherentalrates-areexamplesinpoint.Sincesomepartymustlosewhentheinitialdistributionaltermsarerevised,thatis,thegainereithercannotorwillnotfullypensatetheloserinmakingtherevision,leasetermination（hence,thechoiceofanappropriatelyshortduration）isessential.[11]Again,thiswouldnotbenecessaryiftransactioncostwerezero（evenifunanticipatedeventsoccurindependently）.Intheabsenceoftransactioncost,acontractwouldbedesignedtoallowdaytodaychangesinrentalpayments;withinanyleaseduration,thedistributionofinewouldnotbeheldfixedthroughout.
[1].ThesepercentagesareputedfromDepartmentofRealEstates,ChinaEconomicYearbook（Shanghai:CommercialPress,1935）,pp.G101-4.Asimilarinvestigationconductedinthesamelocalitiestenyearsearlieryieldedanalmostidenticaldistribution（ibid.）.
[2].AccordingtoasurveyconductedbytheExecutiveYuan（ChangesinLandRights）,coveringfourteenprovincesinChina（1937）,7.5percentoftheleasecontractsweredismissedinthatyear.But,sinceinflationbeganinthesameyear,thecitedpercentagemightbehigherthanthatofprecedingyears.FortheriseinpricesowingtotheSino-Japanesewar,seeChangandWang,QuestionsofFarmTenancy,chap.9.
[3].Seesourcescitedinchapter1,notes10and13.
[4].ArmenA.Alchianhasarguedthatthedesireforsecurityleadsto"long-term"contracts.Buthisanalysisisbasedonapropertyrightsystemwithisnotprivate,wheretheprivatecostofacquiringsecurityisrelativelylow.Seehis"PrivatePropertyandtheRelativeCostofTenure,"inThePublicStakeinUnionPower,ed.PhilipBradley（UniversityofVirginia,1959）,pp.350-71.
[5].Twoindependentsurveys（China,1921-24and1935）revealthat,amongtenantfarms,landownersownedabout60to70percentofthehousingassets;tenantsownedabout75percentofthedraftanimalsand95percentofthefarmingequipment.Thetotalvaluesofnonlandassetsonownerandtenantfarmswereroughlythesame.SeeNationalGovernment,StatisticalAnalysis,pp.99-116.
[6].Chiao,StudyofFarmVillages,p.261.ForsimilarobservationsseeNationalGovernment,StatisticalAnalysis,pp.56-58；andChinaEconomicResearchDepartment,SourceMaterials,pp.84-89.
[7].IapplyherethethinkinginRonaldH.Coase,"TheProblemofSocialCost,"JournalofLawandEconomics（1960）.
[8].Legalrecordswhichcoverfifty-sixprefecturesinsixprovinces（China,1934-35）revealatotalof124tenancydisputes（mostlyinrentalpayments）overaone-yearperiod.Eventhoughthetotalnumberoftenantcontractsisnotavailable,thenumberofdisputesbroughttocourtappearstobesosmallthatonesuspectsmanymoreneverreachedcourt.Overtwo-thirdsoftheserecordedcasesendedintenancydismissals,togetherwithpaymentsettlements.SeeDepartmentofRealEstates,ChinaEconomicYearbook,（1935）,pp.G118-20；and（1936）,pp.G143-44.
[9].Withleasetermination,forexample,asharetenantwhoalonewantschangesintheproductionplancanrequestafixed-rentcontract,purchasethelandoutright,orseektenancywithanotherlandowner.Withoutleasetermination,furthernegotiationmaystilltakeplaceifonepartywhowantstherevisionpaysthereluctantpartyanamounttomaketherevision"convincing."
[10].Localitieswithhigherfrequenciesofshareleases（China,1934）wereassociatedwithhigherfrequenciesofshort-termleases.SeeNationalGovernment,StatisticalAnalysis,p.43,tables20and21；andp.59,table26.
[11].Givenanunexpiredleasewhichfixestherentalrate,changingeconomicconditionsmayleadtoaredistributionofine.Buttheefficiencyofresourceallocationmaynottherebybehindered.
D.ConcludingRemarks