Price Rigidity during Japan’s Deflationary Period:

Causes and Consequences

Abstract

A notable characteristic of Japan’s deflation in 1995-2013 is the mild pace of price decline, with the CPI falling at an annual rate of only around 1 percent. Moreover, even though unemployment increased, prices hardly reacted, giving rise to a flattening of the Phillips curve. In this paper, we address why deflation was so mild and why the Phillips curve flattened, focusing on changes in price stickiness. Our first finding is that, for the majority of the 588 items constituting the CPI, making up about 50 percent of the CPI in terms of weight, the year-on-year rate of price change was close to zero, indicating the presence of very high price stickiness. This situation started during the onset of deflation in the second half of the 1990s and continued even after the CPI inflation rate turned positive in spring 2013. Second, using monthly data from 1970 onward, we find that there is an inverse linear relationship between trend inflation and the share of items whose rate of price change is close to zero. This is consistent with Ball and Mankiw’s（1994） argument based on the menu cost model that the opportunity cost of leaving prices unchanged decreases as the trend inflation rate approaches zero. Third and finally, our simulation analysis based on a state- dependent pricing model shows that when deflationary pressure builds over a prolonged period, this gives rise to a situation in which there are more firms than usual for which actual prices are above the “appropriate” level. In other words, there is a large reserve of firms that would lower prices if they could. Under these circumstance, the impact of monetary easing on prices is limited.