Two approaches to saving the economy: Micro-level effects of Covid-19 lockdowns in Italy

In response to the two waves of Covid-19 in 2020, the Italian government implemented a general lockdown in March, but geographically targeted policies during fall. We exploit this natural experiment to compare the effects of the two policies in a difference-in-differences design, leveraging a unique database combining traditional, municipality-level and big data at weekly frequency. We find that the general lockdown of the first wave strongly reduced mobility at a high price in terms of employment, while the targeted policies during the second wave induced a lower decrease in mobility and little additional economic cost. We also study the role of pre-existing municipality characteristics and labour market policies in shaping these responses. Our results suggest that working from home and short-term working schemes buffered the adverse consequences of the drop in economic activity on the labour market. Both mechanisms, however, acted more strongly in high-income areas and among white collar workers, further exacerbating existing inequalities.