Price transmission between Global and Australian wheat markets: A GARCH-based NARDL analysis

Price transmission between distant markets is a measure of market performance, weighing the fairness of the distribution of market surplus among producers, traders, and consumers. We estimate price transmission from global wheat futures (CBOT and Euronext) to Australian regional wheat spot markets (Kwinana and Newcastle) using the Non-Linear Autoregressive Distributed Lag (NARDL) model. The NARDL captures non-linearity and asymmetric adjustments when price series are stationary or of mixed order at the level. The model is further extended into the Generalised Autoregressive Conditional Heteroskedasticity (GARCH) framework to account for time-varying volatility. The results reveal that price transmission from CBOT and Euronext futures to Kwinana is negatively asymmetric in both the short and long run, while it is symmetric to the Newcastle spot market. The export-oriented Kwinana market is highly concentrated, with a single major buyer, whereas the Newcastle market has multiple buyers and significant domestic demand. This study adds to the existing spatial price transmission literature, especially in the underexplored context of Australian grain markets. It also advances the econometric practices by integrating the NARDL framework with a GARCH specification to account for time-varying volatility in high-frequency time series data.