This article examines the hedge performance and market efficiency through the price discovery of mini-gold futures and trading volume against gold and mini-gold spot market during the sample period covering from September 10, 2010 to March 5, 2015. To test the hedge performance, we introduce both the constant hedge model such as the naive, Ederington(1979)’s minimum variance hedge model and VECM(vector error correction modle) and dynamic hedge model like Bollerslev(1986)’s GARCH. The major empirical results are as follows; First, there are co-integration relations between spot and futures prices of gold and mini-gold futures markets. Second, we find that there is a feedback lead-lag relationship between mini-gold spot and futures markets but the price discovery of futures is dominant. However, there is a unilateral influence from gold spot to gold futures but not vice versa. Third, we also find that the change in trading volume of mini-gold futures has an influence on the returns mini-gold futures but those of gold spot and futures. Overall, the impact from returns to trading volume are more strong than that from trading volume to returns. Fourth, we find that as an hedge instrument mini-gold futures is well functioned and according to hedge performance the static hedge models’ hedge effectiveness is relatively greater than that of dynamic hedge model during the out-of sample period. From these empirical results we infer that the KRX gold and mini-gold spot and futures markets are inefficient and these results would be a little bit helpful for a hedger and policy maker to set up risk management strategy and market activity acts in gold spot market in Korea.