From the perspective of new growth models and new international trade theories, both exports and imports play an important role in a dynamic economy. Economically, we cannot merely emphasize on a positive trade balance and we ignore the benefits of imports, and also we cannot advise negative trade balance. But the main concern of policy makers about international trade should be making stability and equilibrium of trade balance in the long run. Especially, deficit agricultural trade balance can be a great danger for food security. Hence, the main objective of this study was to test the sustainability of Iran's agricultural trade balance during the period 1961-2011 (1340-1390). For this purpose, After confirming the existence of cointegration by Gregory-Hansen Test, Hasted and Arize models was estimated with OLS, FMOLS and DOLS methods and it released we can verify sustainability of agricultural trade balance during the period of the study. Also, the estimation of error correction model showed that there is a bidirectional causality relationship between import and export in long-run while in short-run export only cause import in agriculture sector.