Abstract: Ownership, as an institutional factor, has been enmeshed in the Chinese culture and has affected the employment of rural migrant workers since the economic reform in 1978. This paper identifies the determinants of the employment choices of rural migrant workers across different ownership enterprises. We first classify the non-state owned enterprises (non-SOEs) based on the adjusted Chinese official statistics and then comparatively analyze the determinants of employment choices using two discrete-choice models. Using CHIP data for 1995 and 2002, the multinomial logit model indicates that the benefits of pension and housing increase the probability of participation in state-owned enterprises (SOEs), while wage has a positive influence on choosing non-SOEs, especially private enterprises. The nested multinomial logit model shows that unobserved factors related to migration decisions are correlated with employment choices. Furthermore, schooling significantly increases the probability of employment in SOEs compared to other enterprises. Purpose: Understand the employment preferences of rural migrant workers across different ownership enterprises and to identify the determinants of employment choices after the economic reform in China. Originality: 1. Classify non-SOEs into three categories based on the adjusted China’s official statistics; 2. Consider the impact of socio-economic variables of the labor market on employment choice, such as benefits and work conditions; 3. Investigate how factors related to migration decisions affect employment choices. Key literature/theoretical perspective: Miyamoto, K. & Liu, H. (2004) Sector Choice and Wage Determinants: Evidence from Urban China. Journal of Economics, Business and Law, 6, 27-43; Roberts, K. D. (2001) The determinants of job choice by rural labor migrants in Shanghai. China Economic Review, 12, 15-39 and Démurger, S., Gurgand, M., Li, S. & Yue, X. (2009) Migrants as second-class workers in urban China? A decomposition analysis. Journal of Comparative Economics, 37, 610-628. Design/methodology/approach: The multinomial logit model and the nested multinomial logit model. Findings: 1. The benefits of pension and housing increase the probability of participation in state-owned enterprises (SOEs), while wage has a positive influence on choosing non-SOEs, especially private enterprises; 2. Unobserved factors related to migration decisions are correlated with employment choices; 3. Schooling significantly increases the probability of employment in SOEs compared to other enterprises. Research limitations/implications: Data, for example, the data reshaping makes the corresponding data miss out and the alternative-specific variables cannot be applied in the estimation of the nested multinomial logit model. Practical and social implications: 1. The classification of non-SOEs based on the adjusted China's official statistics provides a benchmark for studying the issue of employment choice in China. 2. Low wages in SOEs are disadvantage in attracting the employment of rural migrant workers. This implies the necessity for policy makers to increase wages of rural migrant workers in SOEs and attract the employment of these cheap labor in order to rise profits of SOEs.