Smallholder participation in large forestry programmes.

In recent years, many forestry projects have been implemented in developing countries. In China, a variety of large-scale afforestation and reforestation programmes have been carried out with multiple objectives, such as livelihood improvement and carbon sequestration. As in many developing countries, these projects have been implemented in a smallholder context. This paper investigates the determinants of smallholder participation in large forestry projects. Using the case of camellia, it explores the determinants of smallholder participation using a probit regression model. To distinguish between participation in international and government-run projects, a bivariate probit regression model is used. The findings show that only 37% of households in the sample had participated in the Camellia project; a major reason for the low participation rate is perceived tenure insecurity. The results of the bivariate probit model show that the education level of the household head and household size have a positive impact on the likelihood of household participation. The more 'off-farm' activities are taken up in a household, the less likely a household is to participate in an international project. For a government project, household size also has a positive impact on the likelihood of participation. Chinese forestry is diversifying since the devolution offorestland use rights, with a majority of households hesitating to invest, while some risk investment and others depend on government subsidies. The main policy implication is that, if the Chinese government wishes to achieve its goal of 1.68 million hectares of camellia, then improving tenure security is crucial. [ABSTRACT FROM AUTHOR]