This paper examines the output effects of fiscal policy based on twoapproaches. The first approach employs SVARs. However, it differs fromprevious studies in data sources, classification of government expenditures,and treatment of data to allow for permanent effects of fiscal policy. Thesecond one is the first attempt in Korea to estimate the effects of a tax cutusing historical data on tax rates since 1970. Results show that the effects ofan increase in government expenditures are much bigger than those found inprevious studies. The effects of a tax cut estimated with tax rates are morepronounced than in SVARs.