Spread of E-commerce, prices and inflation dynamics: Evidence from online price big data in Korea

Abstract

The so-called Amazon effect is generally defined as such that the increasing competition in the online market and between online and traditional retailers is reducing retail markups and putting downward pressure on prices. This paper investigates the existence of the Amazon effect using online price big data in Korea where e-commerce has been rapidly spreading. For this task, the direct comparison between the online and offline prices were conducted in terms of the levels, trends, inflations, and dynamic correlations for two prices. Online prices for products were collected and classified under the identical classification in the CPI for 14 items in two divisions: food and non-alcoholic beverages and clothing and footwear. Laspeyres formula with identical weights as for the CPIs was applied to compilation of the item- and division-level OPIs (online price indexes) from July 2018 to June 2019. The empirical analyses overall indicated that persistent decreasing trends in the online prices were found when compared to those in the offline prices represented by the CPIs, indicating the existence of the Amazon effect in Korea. More specifically, the OPI covering two divisions decreased by 1.8%, while the CPI increased by 3.6% for the period. In addition, the close dynamic correlations between on-month inflations of two indexes were also found in panel regression and VAR estimations, indicating that a 1% increasing shock to OPI inflation led to an around 0.3% increasing response in the CPI inflation. The dynamic correlations, however, were quite different across the divisions. Those were more pronounced for the division of food and non-alcoholic beverages while they were found little in the division of clothing and footwear.