Do Agricultural Price Shocks Persist? A Fractional Integration Analysis for the MENA Agricultural Markets

This study investigates the persistence and mean-reversion properties of major agricultural commodity prices in the Middle East and North Africa (MENA) region using a fractional integration framework. The analysis focuses on wheat, rice, maize, sugar, and soybeans under alternative specifications for the error process, including white-noise, Bloomfield autocorrelated, and seasonal autoregressive errors. The results are found to depend critically on the specification of the disturbances. Under white-noise and seasonal autoregressive errors, the estimated fractional differencing parameters exceed one for all commodities, indicating strong persistence and providing little evidence of mean reversion. However, allowing for autocorrelation as in the Bloomfield specification lowers the estimates below unity for all commodities. Significant mean reversion is found for wheat, rice, and soybeans in the case of the original series, while only soybeans exhibit significant mean reversion in the case of the logged series. Thus, agricultural price shocks do not all have permanent effects: their persistence varies across commodities and depends on the underlying error structure. These findings have implications for strategic reserves, agricultural trade, food-price stabilization, and food-security policies in the MENA region.