It is increasingly being discussed whether the Reserve Bank of India should react to rising food prices, generally considered to be due to supply shocks, when overall inflation seems to be under control. This paper first provides evidence, based on the OPEC 1973 price hike, against the supply shock view, and then points to the low trend growth in agriculture as the main factor keeping India's food inflation high. It then presents a simple model of a two person (rich and poor), two commodity (food and non food) economy with rising food and falling nonfood prices. Simulation results show that GDP can go up while aggregate utility goes down. Simultaneously the GDP deflator (considered to be the general price level) falls relative to the consumer price index, a weighted average of the price index for the poor and rich. Accordingly it argues for a population weighted Consumer Price Index to be constructed and its inflation rate made the primary target of monetary policy, eschewing other inflation measures.