سیاست های پولی، شوک های قیمت نفت و اقتصاد ژاپن

Hamilton’s [J. Monetary Econ. 38 (1996) 215] measure of oil price shock is statistically significant in explaining real activity in Japan and also has a statistically significant impact on the stance of Japanese monetary policy. It is estimated that the call money rate was higher by 2.0 percentage points due to the first major oil price shock in the mid 1970s and was higher by about 2.5 percentage points due to the second major oil price shock in 1979–1980. It is found that between 30 and 50 percent of the negative impact of oil price shocks on Japanese output is attributable to monetary tightening induced by oil price shocks. This result continues to hold when the effects of domestic fiscal policy, US monetary policy, and exchange rate regime are recognized.

A large body of empirical evidence indicates that oil price shocks are related to slowdowns in output growth. Hamilton (1983) found that oil price shocks and real economic activity have been strongly correlated in the United States since World War II, even in the period 1948–1972 prior to the first oil price shock. More recently, Hamilton (1996) has recognized that oil price changes are an unreliable instrument for capturing the macroeconomic influence of oil shocks subsequent to 1986. He finds that a new variable, “net oil price increase”, has a statistically significant negative relationship with GDP growth from 1948:1 to 1994:2 which strengthens his earlier conviction concerning the macroeconomic importance of oil price shocks in the US economy.1,2 Oil price shocks have also been found to have influence on real activity in other countries. Burbidge and Harrison (1984) find that oil price rises have a statistically significant effect on industrial production in the US, Japan, Germany, the United Kingdom, and Canada, and Hutchison (1993) and Takenaka (1991) find that oil price shocks influence real GDP in Japan.3 Following similar work for the US, Mork et al. (1994) report asymmetric effects of oil price shocks in a number of countries. They find a statistically significant negative correlation between oil price increases and GDP growth (but no statistically significant effect of oil price decreases) for the US, Canada, Japan, West Germany, France, the UK, and Norway over the period 1967 through 1992.4 Thus, the evidence that oil price shocks play a significant role in explaining macroeconomic activity is well established for a number of countries, including Japan. In a related work, Hoover and Perez (1994) note that US recessions of the past 30 years have been preceded by oil price increases and by contractionary monetary policies. This raises the question as to what extent the ensuing economic declines can be attributed to oil price increases and to tight monetary policies.5Bernanke et al. (1997) set out to answer this question. They find that Hamilton’s net oil price increase is the most appropriate indicator for investigating the macroeconomic effects of oil prices in the US, and that a positive innovation in oil prices is followed by a rise in the federal funds rate. Bernanke et al. estimate that most of the impacts of oil price shocks on the real economy are attributable to the Fed’s tightening in response to adverse oil price shocks. They find that between 2/3 and 3/4 of the reduction in US output following an oil price shock is accounted for by the monetary policy tightening in response to adverse oil price shocks. Fig. 1 shows time-series movements of the call money rate, an indicator of monetary policy for Japan, which is comparable to the federal funds rate in the US. Examination of Fig. 1 indicates that episodes of monetary tightening in Japan, measured by increases in the call money rate, correspond to major oil price shock dates identified by Hamilton (1983) and Hoover and Perez (1994). Increases in the call money rate at the times of the first and second major oil price shocks in the early and late 1970s are especially marked.The objective of this paper is to investigate the extent to which oil price shocks affect the Japanese economy when the effects on real activity of monetary policy responses to oil price shocks are eliminated. To achieve the objective, we employ two separate models. The first model is the conventional VAR in which a monetary policy indicator variable is used without transformation. The second model uses an adjusted monetary policy indicator which is obtained by eliminating its impulse responses to oil price shocks. The effects of oil price shocks on real output obtained from the two models are compared to investigate the extent to which monetary policy responses strengthen or weaken the recessionary effects of oil shocks. The finding that the stance of US monetary policy has been influenced by oil price shocks makes it worth investigating whether the monetary policy in other major economies has also been influenced by oil price shocks. It is found that Hamilton (1996) measure of oil price shock is statistically significant in explaining real activity in Japan, and also it has a statistically significant impact on the stance of Japanese monetary policy. It is estimated that the call money rate in Japan was higher by about 2.0 percentage points as a direct result of the first major oil price shock in the mid 1970s and was higher by about 2.5 percentage points as a direct result of the second major oil price shock in 1979–1980. When the influence of oil price shocks on monetary policy is identified, it is found that between 30 and 50 percent of the negative impact of oil price shocks on Japanese output is attributable to monetary tightening induced by oil price shocks. This result is robust in that it continues to hold when the model incorporates the potential dependence of Japanese monetary policy on US monetary policy measured by the federal funds rate. We also find that 30–50 percent of the negative impact of oil price shocks on Japanese output is attributable to monetary tightening induced by oil price shocks following the introduction of a fiscal policy variable and during the period of flexible exchange rates beginning in March 1973. Throughout the analysis the effect of the exchange rate on the economy through prices other than that of oil is explicitly considered by the introduction of a variable that measures commodity prices in terms of Japanese yen. Section 2 of the paper takes up the issue of measurement of oil price shocks for Japan. Section 3 focuses on isolating the extent of contribution of monetary policy to the recessionary effects of oil price shocks. In Section 4, the relationship between US monetary policy represented by the federal funds rate and Japanese monetary policy represented by the call money rate is explored. The conclusion is given in Section 5.

This paper finds that movements in both oil price and the call money rate have statistically significant forecast power for real economic activities in Japan. It also finds that oil price movements have a significant forecast power for the stance of Japanese monetary policy. An increase in oil price induces an increase in the call money rate, and the increase in the call money rate strengthens the contractionary effect of oil shocks. It is estimated that the call money rate in Japan was higher by about 2.0 percentage points as a direct results of the first major oil price shock in the mid 1970s and was higher by about 2.5 percentage points as a direct result of the second major oil price shock in 1979–1980. It is found that between 30 and 50 percent of the negative effect of oil price shocks on real activity in Japan is due to monetary tightening following oil price shocks. Throughout the analysis the effect of the exchange rate on commodity prices in addition to that on the price of oil is explicitly considered. Results are shown to be robust to changes in model specification. The influence of fiscal policy was found to not change the basic finding in the paper, in that monetary tightening contributed about 35 percent of the contractionary effect of oil price shocks when fiscal effects are introduced. Fiscal policy was an important variable to include in the analysis since there is evidence that oil price shocks also induce fiscal action, possibly through government desire to combat rising unemployment. The result that a substantial fraction of the negative effects on output of positive oil price shocks is due to induced monetary tightening continues to hold in a sample period confined to the floating exchange rate period of 1973:3–1996:5. US monetary policy was included as an influence on Japanese monetary policy and the other variables in the model by introducing the federal funds rate as an additional variable in the model. Despite the introduction of the federal funds rate, 30–50 percent of the contractionary effect of oil price shocks in Japan continued to be due to monetary tightening following oil price shocks.