The model of Brazilian land reform is based on settlements of families without land, in unoccupied public lands, or by expropriation of unproductive estates. The market assisted land reform programs are considered complementary instruments of land reform. They are used to provide credit for land purchase by groups of landless farmers directly from the land market. The aim of this paper was to measure the profit efficiency, considering two different mechanisms of land access. In this paper, analysis of profit efficiency was performed using the stochastic frontier analysis. The frontier of production was estimated based on Cobb-Douglas production function. The data set represents the states: Bahia, Ceará, Maranhão, Pernambuco and the north region of Minas Gerais. The overall efficiency estimates were 0.3163. The efficiency estimates were 0.3678 for the farmers assigned by expropriation, and 0.3537 for the farmers with market land access. The production showed decreasing gains with the major partial elasticity for land. The positive efficiency effect sources are: off-farm labor, collective labor, location in superior soil class county areas, location in Maranhão and Ceará states, age of head of household, crops with irrigation or flood, and use of machinery. The negative efficiency effect sources are: outer incomes and family consumption of its own production.