The paper gives a simplified version of a typical dynamic stochastic open economy general equilibrium models used to analyze optimal monetary policy. Then it outlines the chief modifications when dualism in labour and in consumption is introduced to adapt the model to a small open emerging market such as India. The implications of specific labour markets, and the structure of Indian inflation and its measurement are examined. Simulations give the welfare effects of different types of inflation targeting. Flexible CPI inflation targeting (CIT) without lags works best, especially if the economy is more open. But volatile terms of trade make the supply curve even steeper than in a small open economy despite specific labour markets and higher labour supply elasticity. Exchange rate intervention limits the volatility of the terms of trade and improves outcomes, making the supply curve flatter. As long as such intervention is required, domestic inflation targeting (DIT) continues to be more robust and effective. The welfare losses from the lags in CPI, which prevent the implementation of CIT, are low as long as the dualistic structure dominates. As the economy becomes more open, however, the loss from not being able to use CIT rises. The lags in CPI therefore need to be reduced, making its future use possible.