Afforestation programs have become increasingly significant as policymakers attempt to protect the environment and reduce climate change. Although many studies have examined the participation decisions of forest farm households in afforestation programs, these studies fail to consider different types of these policies. This paper fills this knowledge gap by studying the determinants of forest farms participating in two afforestation programs on plain and sloped land in Taiwan. We construct a population-based sample of forest farms drawn from agricultural census surveys in Taiwan and estimate the multinomial logit model. We find that failing to distinguish between afforestation programs may result in misleading findings. Moreover, socio-demographic and farm production characteristics also affect participation decisions. Additional results from a simple simulation exercise show that forest farms are more likely to enroll in afforestation programs on sloped land, possibly due to lower opportunity costs.
Keywords: Taiwan; afforestation program; multinomial logit model; policy simulation.