In this paper, first, we examine the history records of macroeconomic policies and identify the disinflation episodes in China since 1978 and then measure the costs of disinflation by employing the narrative method. Second, after introducing a bench-mark search model, and estimating the parameters based on China’s data, we compute the sacrifice ratios of disinflation in China in a ... [Show full abstract] search-theoretical environment. The results of two approaches are in line with each other. Our paper contributes several aspects: From narrative methods, we find that disinflation is costly in China: averagely, to reduce 1 percentage of inflation rate will lose 0.5 percentage in GDP growth rate. The sacrifice ratios vary across episodes from 0.20 for the fifth disinflation (2007-2008) to 0.84 for the third disinflation (1988-1990). It seems that a slow disinflation process yields lower cost in China. The search-theoretical approach provides similar results: for buyers-take-all case, the welfare of the residential measured by consumption decreases about 0.18 percentage and 0.49 percentage by reducing 1 percentage of inflation for and respectively. For the partially bargaining case, the welfare declines about 0.25 percentage for a one percentage reduction in inflation.