This paper examines the effect that gender and racial diversity of a board have on the performance of REIT in South Africa. The legacy of the Apartheid system and the creation of affirmative action programmes make South Africa a justifiable and relevant case study for such research. The dataset used is obtained from the McGregor Database and covers thirty-three (33) JSE-listed REITS. The dependent variable is the squared of return on asset (ROA). The independent variables are the proportion of females and non-whites on the boards. The control variables used are the board size, board independence and market capitalisation. We hypothesised that the proportion of females and non-whites on the boards lead to improved performance. Using the multiple regression model, we find gender (the proportion of females on the board) and racial diversity (non-whites on the board) to affect the performance of the REITs. While the gender affect ROA Positively, the racial diversity has a negative effect on ROA. The results show that when the proportion of females on the board increases by 1%, all other things being equal, ROA also increases by approximately 4.1%. However, when the proportion of blacks on the board increases by 1%, all other things being equal, the performance of REITs reduces by some 2.95%. Two of the control variables, board independence and the size of the board also have a positive and negative effect on performance respectively. These results have are useful for various decision makings for REITs companies.