The earning function clearly supports the existing evidence of the significant positive coefficients for education and the marginal wage effects are increasing with the level of education for both the genders. There are clear evidence of caste bias for males, location and regional bias for both the genders in earning. Our decomposition results show that endowment component which shows the existence of pre-market discrimination is smaller than the discrimination component. Discrimination explains 66.1 per cent of the lower wages of female individuals when compared to males. Discrimination component is the highest for the production workers (81.3 per cent) followed by professionals (77.6 per cent), agriculture/allied workers (77.4 per cent), clerical workers (65.9 per cent) and is least for sales/services workers (61.4 per cent). Gender wage discrimination is very high for the urban areas (86.3 per cent) than the rural settings (71.3 per cent). Large discrimination differences are a matter of concern for thepolicy makers. JEL Classifications: I21, J30, J31.