本研究剖析民國89年10月份國安基金首次以臺股期貨進行護盤的合理性，並根據國安基金實際交易臺股期貨的日內資料，利用向量自我迴歸模型(VAR)，探討其對市場價格及定價誤差的影響。實證顯示國安基金在進行當時顯然達成拉抬期貨，造成超額正價差的目的，但是同時間的現貸指數並未因國安基金進場所造成的正向套利空間的引導而上升。所以本研究認為國安基金干預市場雖然在短期之內有穩定市場的功能，但並未達到如香港行政當局拉抬期貨市場而造成現貨市場上揚的效果，其可能的原因為當時我國期貨市場規模較不對稱，期貨價格還不具備市場領先指標的功能，同時機構投資者的成效金額在股市所佔比例不高，造成國安基金無法用拉抬期貨價格產生正向套利空間進而達到穩定現貨指數的目的。

The National Financial Stabilization Fund (NFSF) was established in February 2000 to stabilize the financial markets in Taiwan. In order to prop up the Taiwan Stock Exchange Capitalization Weighted Index (TAIEX), the NFSF first intervened in the TAIEX Futures market in October 2000. There is no research evaluating the reasons why the NFSF intervened in the stock index futures market or whether its intervention strategy was effective to keep the TAIEX from panic selling pressure. Using intraday NFSF trading data obtained from the Taiwan Futures Exchange (TAIFEX), this paper investigates the intervention effectiveness by the NFSF on Taiwan stock index futures markets. We also examine whether the NFSF achieved its objectiveness of stabilizing the stock index through intervene in the stock index futures market. Finally, we explore the reasons why intervention by the NFSF did not have a long-run stabilization effect on the stock market. Government interventions in the foreign exchange and interest rates markets are often seen, and there is much empirical work on the effectiveness of these interventions. Except for the intervention by Hong Kong Monetary Authority (HKMA) during the Asian financial crisis period, prior studies focus on government intervention in the stock or futures market is spare. The HKMA launched a series of intervention operations between August 14-28, 1998 to pop up the Hang Seng Index (HSI). These intervention programs involved buying 33 constituent stocks of the Hang Seng Index, large numbers of HSI futures contracts, and Hong Kong dollars in the foreign exchange market. Draper and Fung (2003) showed that there were frequent overpricing of futures during periods of HKMA intervention, but they did not provide explanations on why overpricing of HSI futures contracts is useful to stabilize the Hong Kong's stock index. This study tried to explore the impact of NFSF intervention using Taiwan data. We also provide suggestions for policy makers on how to design a more effective way to pop up the stock market through futures market interventions. We find that there are three major reasons why overpricing in the futures market is helpful to stabilize the spot stock index. First, the government can spend less money to intervention in the futures market than in the stock market because the leverage in the futures market is much higher. Margins requirement in the futures market is less than 15% of the contract size, so intervention in the futures market should be more effective. Second, one of the functions of the futures market is price discovery, that is, the futures market may respond to new information much faster than the spot market. Chan (1992) finds evidence that futures prices lead spot index prices. Hsieh (2002) indicates that the TAIEX futures price appears to be more informative than the spot prices. Therefore, if NFSF successfully boost the TAIEX futures index, it could potentially alter traders' expectation and stabilize the spot index level. Third, by propping up the futures price, NFSF can create excess overpricing in the futures market and invite cash-and-carry arbitrageur to enter the market. By taking long positions of the underlying stocks and short positions in the futures market, arbitrageurs can help to alleviate the TAIEX panic selling pressure. Consistent with the findings by Draper and Fung (2003), we find prevalent excess overpricing of TAIEX futures contacts during periods of government intervention. The excess overpricing persists at least one trading day. It shows that interventions by the NFSF effectively boosted prices in the futures market and resulted in a positive spread between the stock and futures markets in the short run. Nevertheless, the stock index level was lower after the interventions. We conclude that the reasons for these unsustainable intervention effects are due to the facts that trading volume in Taiwan' futures market was thin and transactions of institutional investors, who have ability to arbitrage between spot and futures markets, were small during the sample periods. Hsieh (2002) shows that TAIEX futures contracts lead the stock market when the futures trading volumes are larger. During the sample period, the volumes of TAIEX futures were about 7% to 15% of the TAIEX stock market. As such, TAIEX futures price was not considered a leading indicator for investors in the stock market during the sample period. The proportion of institutional investors' trading volume in the spot market was also lower than the amount of individual investors. Therefore, our results show that even when the NFSF effectively boosted the TAIEX futures price and widened the cash-and-carry arbitrage opportunities, the NFSF could not prevent the TAIEX from further declining because of small arbitrage-related trading volumes results from institutional investors' trading in the spot market.