In this paper, we estimate government purchase multipliers for Japan, following the approach used previously for a panel of OECD countries (Auerbach and Gorodnichenko, 2013). This approach allows multipliers to vary smoothly according to the state of the economy and uses real-time forecast data to purge policy innovations of their predictable components. For a sample period extending from 1960 to 2012, estimates for Japan are quite consistent with those previously estimated for the OECD as well as those estimated using a slightly different methodology for the United States (Auerbach and Gorodnichenko, 2012). However, estimates based only on more recent observations are less stable and provide weaker support for the effectiveness of government purchases at stimulating economic activity, particularly in recession, although cyclical patterns in Japan make the dating of recessions a challenge.