Measuring lockdown travel rebound: Experimental insights from the Hong Kong COVID-19 omicron outbreak

While localized lockdowns emerged as pivotal Non-pharmaceutical Interventions (NPIs) during the mid-to-late phases of the COVID-19 pandemic, the precise impacts on residential mobility remain understudied, including pre-and post-lockdown travel comparisons, ‘bounce-back’, and long and short-term adjustments. Policymakers were particularly concerned about unanticipated viral transmission from undiagnosed asymptomatic carriers. This study investigates Hong Kong’s Restriction-testing Declaration (RTD) policy, a short-term localized lockdown strategy mainly targeting public housing estates (PHEs) during the omicron outbreak. Employing a Propensity Score Matching and Difference-in-Differences (PSM-DID) framework, we analyzed anonymized mobile positioning data from 117,370 resident’s mobile devices during omicron outbreak, to quantify the lockdown-induced mobility response, defined as the distance to maximum reach position (DMRP).Our findings reveal that RTD events generated substantial increases in the daily average increase of 1.20 km of DMRP over the two-week study duration, and exhibits an inverted-U-shaped rebound pattern in differed epidemic phases. In the rising phase, RTD immediately increased PHE residents’ DMRP by 0.89 km on the implementation day, and this effect persisted for 11 days, with the maximum surge occurring on post-RTD Day 5. In the peak phase, no immediate impact occurred on the implementation day. However, an increase of 1.04 km occurred one day later, and this effect continued for 5 days, reaching its maximum surge on post-RTD Day 2. No significant DMRP increase was observed for the declining phase. This is the first empirical study to investigate how localized lockdowns impact residents’ travel rebound and provides key insights for policymakers to understand unintended mobility responses.