The 'welcomed lockdown' hypothesis? Mental wellbeing and mobility restrictions

Abstract

The COVID-19 pandemic and its mobility restrictions have been an external shock, influencing mental wellbeing. However, does risk exposure to COVID-19 affect the mental wellbeing effect of lockdowns? This paper examines the 'welcomed lockdown' hypothesis, namely the extent to which there is a level of risk where mobility restrictions are not a hindrance to mental wellbeing. We exploit the differential timing of exposure the pandemic, and the different stringency of lockdown policies across European countries and we focus on the effects on two mental health conditions, namely anxiety and depression. We study whether differences in the individual symptoms of anxiety and depression are explained by the combination of pandemic mortality and stringency of lockdown. We draw on an event study approach, complemented with a Difference-in-Difference (DiD), and Regression Discontinuity Design (RDD). Our estimates suggest an average increase in depression (3.95%) and anxiety (10%) symptoms relative to the mean level on the day that lockdown took effect. However, such effects are wiped out when a country's exhibits high mortality ('pandemic category 5'). Hence, we conclude that in an environment of high mortality, lockdowns no longer give rise to a reduction in mental wellbeing consistent with the 'welcome lockdown' hypothesis.

Keywords: Anxiety; COVID-19; Depression; Lockdown; Pandemic.

Fig. 1

Stringency Index (blue bricks) and risk exposure (green circles) and deaths per million (red areas). Date: March 20, 2020. Red areas correspond to the mortality rate (deaths per 1,000,000 inhabitants). Higher color intensity denotes higher mortality rate. Data come from https://ourworldindata.org/coronavirus-data-explorer. Green circles correspond to the exposure rate to COVID-19 (confirmed cases per 1,000,000 inhabitants). Larger diameter denotes higher exposure to the virus. Data come from https://ourworldindata.org/coronavirus-data-explorer. Blue Bricks corresponds to for COVID-19 Government Response Stringency Index (Stringency Index). Higher height denotes higher stringency. Data come from https://www.bsg.ox.ac.uk/research/research-projects/oxford-COVID-19-government-response-tracker

Fig. 2

Event study results. Effect of the days before/after lockdown and interaction between days before/after lockdown and pandemic of category 5 over Depression Index and Anxiety Index. The upper panel graphs show the estimated coefficients for ${\sum }_{j=-7}^{j=7}{\gamma }_{0k}{D}_{\mathit{kc}}{L}_{\mathit{ct}}$ of Eq. 2 for Depression Index (left) and Anxiety Index (right). The lower panel graphs show the estimated coefficients for ${\sum }_{j=-7}^{j=7}{\gamma }_{2k}{D}_{\mathit{kc}}{L}_{\mathit{ct}}{P}_{\mathit{ct}}$ of Eq. 2 for Depression Index (left) and Anxiety Index (right). See Table 3 for the detail of coefficients and standard deviations. The red dashed line used to signal the day when lockdown became effective

Fig. 3

Regression Discontinuity plots for the Depression Index and Anxiety Index. Kernel-weighted local polynomial smoothing discontinuity plot with a triangular kernel. The upper panel graphs show discontinuity for PHQ-8 Depression Index and Anxiety Index around the day when lockdown became into force. The lower panel graphs show discontinuity for PHQ-8 Depression and Anxiety Index around the day when COVID-19 reached category 5 in the Pandemic Severity Index