World Bank Group2023-04-052023-04-052023-04-27https://openknowledge.worldbank.org/handle/10986/39633Global commodity prices fell 14 percent in the first quarter of 2023, and by the end of March, they were roughly 30 percent below their June 2022 peak. The unwinding of prices reflects a combination of slowing economic activity, favorable winter weather, and a global reallocation of commodity trade flows. Commodity prices are expected to fall by 21 percent this year and remain mostly stable in 2024, although the outlook is subject to multiple risks in a highly uncertain environment. These risks include intensification of geopolitical tensions, the strength of demand from China following its post-COVID reopening, likely energy supply disruptions, and weather conditions, including the emerging El Niño. A Special Focus section evaluates the performance of several approaches used to forecast prices of seven industrial commodities. It finds that futures prices, which are widely used for price forecasts, often lead to large forecast errors. Time-series models based on multiple independent variables tend to outperform other model-based approaches as well as futures prices. Machine-learning techniques yield better forecasts than some of the traditional approaches. The analysis suggests that augmenting model-based forecasting approaches—by incorporating the dynamics of commodity prices over time and controlling for other economic factors—enhances forecast accuracy.CC BY 3.0 IGOCOMMODITY PRICESENERGY PRICESINDUSTRIAL COMMODITIESOIL AND GASOPECOPEC+AGRICULTURE PRICESMETAL PRICESSerial10.1596/39633https://doi.org/10.1596/39633