Comment on “Chronic Deflation in Japan”

Abstract

Nishizaki et al. (2014) provide a useful summary of stylized facts and existing studies about Japan’s deflation. Put simply, inflation may be written as π = F(Z, MP), where π is the inflation rate, MP represents monetary policy measures, and Z is all the other variables affecting inflation. Deflation may ensue if (i) MP is not effectively used, (ii) the power of MP to affect inflation is low due, say, to the zero lower bound on interest rates (ZLB), or (iii) significant negative shocks, Z, hit the economy. Most of Nishizaki et al.’s discussion is devoted to Z, but in addition, interactions between Z (a decline in the natural rate of interest/inflation expectations) and the ZLB, and those between MP (central bank communication) and Z (inflation expectations), are also discussed. In the following, I would like to point out some factors that might have affected inflation developments in Japan, but that are not quite addressed by Nishizaki et al. Let me first mention the possibility that Nishizaki et al. only briefly discuss, that is, deflation may have been caused by an inappropriate use of MP, or for that matter, by the the Bank of Japan’s (BoJ) implicit policy of targeting a near-zero inflation rate. The case in point are the two exits – one in 2000 and the other in 2006 – from a forward guidance strategy despite a negative core consumer price index (CPI) inflation rate. As Ueda (2013) points out, the BoJ internally discussed what the appropriate inflation target was at the end of year 2000, but was unable to choose between small positive inflation and zero inflation. Such developments may have affected the public’s or the market’s perception of the BoJ’s resolve to fight deflation, despite efforts aimed at improving communication, as discussed in Nishizaki et al. Although the two exits and/or the doubts about the BoJ’s intention to target a positive inflation rate were probably not a direct cause of deflation, they may have undermined the power of subsequent easing measures to stimulate the economy. Nishizaki et al. point out, as do other recent discussions of Japan’s Phillips curve, that the curve has become flatter since the late 1990s. Thus, inflation did not rise much in the mid-2000s despite a fairly strong recovery. Going forward, inflation may not go up to the current 2% inflation target easily. On the other hand, the flattening of the curve has meant that deflation did not accelerate during the last 15 years. In a sense, inflation has been rigid downward. This is puzzling because of the large and persistent gross domestic product gap that is discussed by Nishizaki et al. It is even more puzzling because wages have not been as rigid downward as has the CPI. One wonders if there has been a