Cycles in natural-resource commodity prices: An analysis of the frequency domain

A model is derived that relates price cycles to investment behavior in the mineral industries and predicts that these cycles will be roughly twice as long as the investment period (the period between the investment decision and the project completion). Spectral techniques are used to estimate the principal cycles in metal and fuel prices. For the metals, the estimated cycle lengths reinforce the model predictions. However, no significant cycles are found in the prices of the fuels. The lead-lag relationships between prices of different commodities are also investigated and are found to be significant for those that are jointly produced.