1. As disinflation has been going on for a considerable period of time globally, a possible pivot in central banks’ monetary policy has been put under the spotlight of market interest. However, it is still unclear when, under what conditions, and with what certainty inflation will return to their respective target levels, mostly 2% in major economies. To shed light on this matter, in this article, we examine past episodes of high inflation in major economies and identify the main characteristics of inflation which need to be checked in bringing back price stability. Based on these findings, we diagnose the current inflation situation, and draw implications for monetary policy.

2. We trace out three main characteristics of inflation from past transitions of high-inflation to low-inflation regimes which are ① reduced cross-sectional spillovers of any sectoral inflation shock, ② reduced interactions between inflation and inflation sentiment, and ③ a gradual convergence of headline inflation on underlying inflation. All these traits can be inferred from economic agents' rational inattention (Sims 2003) toward inflation. That is, inflation is not an important factor in the daily decision-making of economic agents, and thus sectoral inflation shocks do not significantly carry over to other sectors. In contrast, high inflation episodes are involved with economic agents putting sharp focus on inflation which usually lead to major adjustments in prices and consumption behavior. This process can be observed indirectly in the interaction between headline and underlying inflation. Looking at composite inflation indices, which are the weighted sum of inflation by sector, in a low stable inflation regime we see that headline inflation indices, which contain items directly affected by external shocks, converge on the underlying indices, which mostly contain items subject to spillovers from the shocks, and since the shock disappears on its own, the spillovers are negligible. Conversely, in a high-inflation regime, sectoral inflation shocks spread to neighboring sectors, and the underlying indices follow the headline inflation because of spillovers that appear with a time lag.

3. Judging from past episodes, disinflation and the transition to price stability is well under way in Korea as in other major parts of the world, however, we do observe that the so-called ‘last-mile risks’ still remain on our way to bringing back price stability. Specifically, looking at expected inflation and the distribution of inflation by item, it appears that some momentum for price adjustment is yet to die out. Also, additional cost shocks remain as an ongoing risk along the way. Therefore, we argue that it is necessary to continue efforts to patiently analyze and judge the movements of various inflation indicators in order to avoid overemphasizing momentary false positive signals from certain price indicators.