The Impact of COVID-19 Lockdown on Digital Lifestyle in China

Abstract

<p>A basic question in digital marketing is the extent to which different aspects of digital lifestyle and consumption depend on physical interactions and mobility. We disentangle the relationship between digital consumption and physical activities by combining the mobile app data with city-level foot traffic and instore credit card spending data. In particular, we use anonymized individual-level mobile phone data of 80,533 randomly selected active users of a national carrier in Chengdu, Sichuan. The dataset includes, at the individual-user level, daily data usage for all mobile apps, mini-programs, and web browser destinations (https), and spans December 1, 2019 to April 30, 2020, and July 2 to August 14, 2020, during which there were 3 deescalating levels of lockdowns.</p><p>We leverage the mobility restrictions in regression discontinuity in time with multiple cutoffs analysis to quantify the impact of the mobility restrictions on mobile app usage, and also test for age and gender inequalities. Furthermore, we investigate if consumers’ digital consumption patterns (e.g., social media vs. messenger vs. education apps, etc.) are driven (and constrained) by their physical mobility and offline behavior. We used three different mediators; intra-city mobility, outflow mobility (Baidu), and transaction frequency of offline Union Pay payments (at the aggregate-level). Overall, the mediation models demonstrate that physical mobility and activities are fundamental forces driving digital behaviors and lifestyles (for which there are significant demographic heterogeneities), and that restrictions on human mobility were largely responsible for the reduction of digital activities during the pandemic in China.</p>