Oil prices react to various types of shocks, and their impacts could reach our lives quickly. For example, on one day, we might hear on the radio that OPEC has decided to cut their oil production for the next year, and, only a few days later, find out that our local gasoline station has just raised the price. This paper examines daily data on gasoline prices, produced by a price comparison site in Japan, to estimate how they respond to a shock that hit the world oil market. In doing so, we take seriously the possibility that an increase in oil prices might cause different reactions depending on the source of the change. This paper focuses on one particular type of shock, namely changes in expectations about future supplies of crude oil. Identification is achieved via estimating a version of the Structural VAR with External Instruments (SVAR-IV or proxy-VAR) coupled with High Frequency Identification (HFI). The result confirms that pass-through is indeed very fast: about 70% of the entire adjustment process is completed within just 18 days.

石油价格对各种形式的冲击做出反应，它们的影响可能会迅速影响我们的生活。例如，有一天，我们可能在广播中听到欧佩克已决定削减其明年的石油产量，仅几天后，我们发现我们当地的加油站刚刚提高了价格。本文研究了日本价格比较网站提供的每日汽油价格数据，以估算汽油如何应对冲击世界石油市场的冲击。在此过程中，我们认真考虑油价上涨可能会根据变化的来源而引起不同反应的可能性。本文着眼于一种特殊类型的冲击，即对未来原油供应的预期变化。通过使用外部仪器（SVAR-IV或proxy-VAR）以及高频识别（HFI）估算结构VAR的版本来实现识别。结果证实了传递速度确实非常快：整个调整过程的大约70％在短短18天内就完成了。