Seite - (000165) - in Evaluating Climate Change Action for Sustainable Development

travel time remain small. A 10% increase in travel time to themarket increases agricultural landby1.6%. This finding that travel time has modest effects on agricultural decisions in Forest Reserves of Chiang Mai is consistent with other studies of the region: Cropper et al. (1999) find that a 10% increase in road density leads to a 2% decrease in forest cover inNorth Thailand. Cropper et al. (2001) find that a 10% increase in travel timeto themarket leads toa2.4%decrease inforestedarea in the forest areasofNorthThailand.Similarly inNorth-eastThailand,Panayatou (1991) finds that changes in roaddensity have an insignificant impact. Onepolicyconclusionfromthis is that roadbuildingmaynothaveadeleterious effect on forest cover in this area. This is different fromwhat has been found in other parts of the world. To the extent that roads provide increased access to services and markets, improving access within Forest Reserves might help to alleviate poverty without affecting forests. However this result should also be treatedwith caution.34 8.6.3 PropertyRights In this study, I make a distinction betweenNPR villages andAPR villages. It is important tomake thisdistinction:villageswithnosecureproperty rightsare likely to bemore remote andpoorer thanvillages that have ambiguousproperty rights. Animportanteffect in thestudy is thatvillageswithnoproperty rightsare likely to likely to cultivate their land less intensively (being in anNPR village reduces intensityofcultivationby71percentagepoints).Howevermagnitudesof impactof the twomainvariables– travel timeandpopulation–oncroppingdecisionsarenot verydifferent for the twogroupsofvillages.Particularly, travel time tomarkethas anegligibleeffectonupland ricecultivationandagricultural land inNPRvillages. Themixed evidence is explained by the fact that the distinction between the two groupswith respect to their property rights is not sharp.Villageswith noproperty rights (NPR villages) are located in the same region as those with ambiguous property rights and are likely to behave similarly. Feder et al. (1988a, b) in their study of Forest Reserves inNortheast Thailand show that villageswithout secure property rights are less likely to invest in land. This may help to explain the significantly lower intensity of cultivation in NPR villages. They also conclude that secureproperty rightsallowbetter access tocredit. In this study thedistinction 34Therandomeffectsestimators in thestudyreflectprimarilycross-sectionalvariation in thedata. Differences ineffectsof transportationcostscouldthusbepickingupdifferencesbetweenlocation of villages. 8 UsingMixedMethods toAssessTrade-OffsBetweenAgriculturalDecisions. . . 149