Abstract: The paper use the two stage DEA model to calculate the rural collective construction land market efficiency, and then analyze the influence of transaction cost, asset specificity; transaction uncertainty and other factors on the demand side of efficiency, which is based on the 398 survey data of the rural collective construction land market in Nanhai District. The results show that: (1) The overall efficiency of rural collective construction land market is 0.230, the pure technical efficiency is 0.390, and the scale efficiency is 0.658. Most of the market scale is in the increasing stage, which can attract more social capital to participate in the construction of new countryside for promoting urbanization and urban upgrading. (2) The transaction cost has no significant influence on land market efficiency, which is related to the sublease behavior in the market. (3) The aspect of asset specificity: transaction subject, enterprises with high skill level workers are helpful to improve the land market efficiency. Transaction object, in equilibrium, the higher factor price in agglomeration area mean better production factor matching, and the efficiency of the rural collective construction land market is higher. (4) The side of transaction uncertainty, transaction behavior, enterprises will choose to increase production input to optimize the level of market resource allocation when the enterprise life cycle is more matched with the land use cycle. Transaction environment, the way of price determination has a significant impact on market efficiency. The lower the transaction uncertainty is, the more favorable it is for the demander to improve the expectation of rural collective land market and market efficiency. Therefore, the rural collective construction land market can promote the consolidation and rectification of rural collective land space and then reduce the fragmentation degree. At the same time, the transaction system of the collective construction primary land market and the secondary market should be constructed to improve the inputoutput efficiency on the demand side.