TH E IMPACT OF COVID-19 ON SMALL BUSINESS DYNAMICS AND

2 the pandemic and the subsequent rebound are larger than the estimates for total employment in the four sectors as estimated by the CES.3 Our estimates therefore imply that employment by small businesses contracted more severely in the beginning of the pandemic than employment of larger businesses but then also rebounded more strongly. [...] 8We define these rates relative to the count of active establishments in the reference week as opposed to the count of active establishments around week t because the count of active establishments varies dramatically in the beginning of the pandemic. [...] As described in the previous section, employment in the CES is measured by the number of workers receiving pay for any part of the pay period that includes the 12th of the month independent of whether they actually worked, while employment in HB is measured by the number of workers who logged positive tracked hours plus all workers who used the HB software otherwise in a given week. [...] In sum, the take-away from Figures 4 and 5 is that temporary closings and reopenings are the primary driver of the large contraction and rebound in the beginning of the pandemic. [...] Hence, many of the smallest businesses – the ones that are the focus of our study – were subject to delayed access to PPP loans during the beginning of the pandemic and the extent of this delay depended in large part on geography.