Futures and realized cash or settle prices for gold, silver, and copper

Abstract

AbstractThis article examines the relation between 15‐, 30‐, 45‐, and 60‐day gold, silver, and copper futures, and their realized cash or delivery settle prices, for deliveries on the first, middle, and last business day (FD, MD, and LD, respectively) of the delivery month. Samples of futures prices for gold, silver, and copper, and the realized cash or delivery settle prices, based on fixed maturities for a cross‐section of contracts are used, because time‐series over fixed maturities are unfeasible owing to the restrictive maturity periods in these contracts. This study contrasts with similar studies for forwards, which are not hampered by this problem (e.g., Baillie and Bollerslev in 1989), studies that use contemporaneous time‐series of futures and cash prices to obtain optimal hedge ratios (e.g., Ghosh and Clayton in 1996), and studies that examine for parity between different futures prices (e.g., Franses and Kofman in 1991). Near‐term gold, closest to delivery silver and all copper futures (the latter only with settle prices at delivery) are good predictors of the future cash price, except for silver and copper deliveries on the last day of delivery months. These results are consistent with previous studies on short‐term cash (Ma in 1985) and futures (Ma and Soenen in 1988) parity between gold and silver, and with Gross's (1988) results for copper. The longer‐term gold (45‐ and 60‐day) and silver (30‐, 45‐, and 60‐day) futures reject an unbiased expectations hypothesis. This result is consistent with Leistikow (1990) because cash prices respond less than futures prices in these instances.