The effects of prawn (Macrobrachium rosenbergii) stocking density and the use of commercial diet on rice–prawn culture were evaluated. The experimental design was completely randomized in a 2 × 2 factorial, with a control (rice monoculture, RM) and four replicates for each treatment. The treatments were: 2 prawns m^− 2 without feed (2W), 2 prawns m^− 2 with feed (2F), 5 prawns m^− 2 without feed (5W) and 5 prawns m^− 2 with feed (5F). Mean survival varied from 17.2 ± 9.3% (5F) to 39.2 ± 14.9% (2F), and mean weight from 20.3 ± 2.2 g (2W) to 25.7 ± 5.85 g (2F). Prawn productivity ranged between 118 ± 58 kg ha^− 1 (2W) and 224 ± 46 kg ha^− 1 (5W). No significant interactions were observed between feeding management and densities. Survival rate was significantly higher (P < 0.05) for the 2 prawn m^− 2 density while the other variables did not vary significantly among treatments. Rice yields were not statistically different among treatments and varied from 4608 ± 1225 kg ha^− 1 (2F) and 5620 ± 1202 kg ha^− 1 (RM). The results indicate that the rice–prawn system is technically feasible and does not need the addition of commercial diet. Economic analysis showed that stoking 5 prawns m^− 2 was not profitable. Total production costs and gross revenue were US$ 1337.84 and US$ 1182.07 for rice monoculture and US$ 2655.60 and US$ 2738.70 for the rice–prawn system (2 prawns m^− 2; without feed), respectively. Therefore, the integrated rice–prawn system is a suitable alternative to make rice production economically feasible on small farms without government subsidies.

Statement of relevance:

New finds to the international literature, such as the synergic effect of stocking density and commercial feed supply in simultaneous rice–prawn system assessed by a technical and economic perspective; discussion about carrying capacity of paddy fields, in simultaneous system; natural feed is not a major limiting factor for stocking densities up to 5 prawns m^− 2 as presumed in other studies.