Present paper provides a detailed critique of the traditional methodology of taking wage differential as equal to the differential of marginal productivity of labour and proposes to replace it by the differential of average productivity of labour types. The purpose in this paper is not to provide a concrete estimate of the labour quality index, but to discuss possibility of following an alternative method and expand the scope of measurement from the traditional age–sex–education classification to a wider range including some directly non-quantifiable labour qualities. The traditional method and the previous efforts made in India ignored classifications of workers based on the urban–rural residence, organized–unorganized sectors, public versus private sector employment, self-employed versus hired workers versus casual employees, and health and nutritional status. If all these qualitative aspects are properly considered, we can get a reasonably accurate marginal contribution of formal education, urbanization, privatization, self-employment and health and nutrition status of workers to the economic growth.