Abstract: Embedded in the deeply rooted rural clan networks, do village leaders favor their clan relatives in policy implementation and public resources allocation? This paper aims to answer this question. Based on the monitoring data of a World Bank financed poverty alleviation and agriculture development project in China, we find that village leaders' clan relatives are indeed more likely to participate in the project. The result remains robust even when we control for household characteristics and clan size. We find that the higher the level of democratic and governmental supervision, the stronger the effect of village leaders on the project participation of their clan relatives, which excludes the existence of capture or favoritism, and supports the mobilization theory. In other words, village leaders did not favor their clan relatives in the project, but were more likely to mobilize them to participate in the project through the clan network. Moreover, we show that the effect is mainly driven by village chairmen rather than party secretaries, which also supports the mobilization theory. The main result remains robust even after eliminating abnormal data and replacing some variables.