Labor Market Impacts of COVID-19 on Hourly Workers in Small- and Medium-Sized Businesses: Four Facts from Homebase Data

0

An up-to-date picture of COVID-19’s labor market impact is now available, thanks to researchers at IRLE, the California Policy Lab, Rustandy Center for Social Sector Innovation at Chicago Booth, and the University of Chicago Poverty Lab. They are taking advantage of granular data on exact hours worked among employees of firms that use the Homebase scheduling software. Check back for weekly updates:

Week Four: Update with Homebase Data Through May 9
Week Three: Update with Homebase Data Through April 25
Week Two: Update with Homebase Data Through April 11
Week One: Initial Analysis of Data Through March 28 and Methodology


Week Four: Update with Homebase Data Through May 9

We update our analysis of the impact of COVID-19 on small- and medium-sized businesses and their employees using Homebase time-card records updated through May 9th. We highlight three new facts from our analysis of the last two weeks’ of data below.

Fact #1: Much of the hours reductions persist, but some firms are reopening.

Though the lion’s share of hours reductions and firm shutdowns persist, the slight bounceback in hours identified in our last update continued through May 9th. The share of firms operating as of early February that remained shutdown fell from around 45 percent in late April to below 40 percent last week (Figure 1). As of the week ending on April 18, total hours across all firms fell by about 60 percent from the baseline period. As of last week, that fraction had shrunk to about 35 percent. Much of the improvement in hours is attributable to some firms reopening (Figure 2).

Figure 1: Ratio of total hours worked at each firm relative to hours worked between January 19-February 1

Fact #2: Beauty and personal care businesses, as well as firms in states where shutdown orders have been rolled back, were more likely to reopen.

Propensity to reopen and restore hours varies slightly by region and industry. For example, beauty and personal care firms were most likely to have reopened and restored some share of hours. Hours also bounced back more in states that moved to reopen than in states that never shutdown or remained shutdown (Figure 3). In so-called “reopened” states, total hours in the past week (May 3-9) were 35 percent higher than their respective low. In other states, hours have only come back about 26 percent from their respective lows. Hours remain far below their baseline averages in all regions and industries.

Fact #3: Reopened firms tended to re-hire original employees, but they’ve only restored a portion of their original hours.

Of the roughly 42,000 unique firms in our baseline sample, approximately half (around 21,000) shutdown for at least one week in March or April. Of these firms that ever have shut-down, almost 6,500 have reopened and remained open through last week. In Figure 4, we show how much these firms’ baseline hours and workforce were restored through each week. By last week, these re-opened firms had collectively regained about 35 percent of their baseline hours and 40 percent of their baseline employment levels. Almost 90 percent of this reemployment came through re-hiring employees who worked at the firms before they shutdown, as opposed to new hires. This suggests that, so far, most worker-firm matches at these firms have been maintained through the crisis, firm-closures, and re-openings.

Although some firms are re-opening, most remain closed. Figure 5 shows how hours have been restored (or are still missing) among firms that shut down before April 4. Across these firms, two-thirds of their collective baseline hours remain missing due simply to ongoing firm closures. Another 19% of hours are lost because reopened firms are operating at reduced scale. Counting all of the firms that have ever shut down, only about 14 percent of pre-shutdown hours have been restored.

Describing Homebase Firm Characteristics

Our analysis is based on firms that use Homebase, which are not representative along many dimensions. They tend to be smaller than average, and concentrated in particular industries (such as food and drink and retail).
Below, we provide information on three characteristics of the firms in the Homebase data: the regions they are located in, the industries they operate in, and the number of employees they have. This information is based on firm characteristics between January 19 and February 1 and includes all firms operating during this time period, regardless of whether they have continued to operate through the COVID-19 crisis or not.

Figure 6: Homebase Firms Broadly Match the Distribution Across US Regions

This figure shows the share of Homebase firms located in each Census region (in light blue) and compares these shares to each region’s share of total employment (in dark blue), as measured by the Bureau of Labor Statistics (BLS) data. Homebase firms are somewhat concentrated in the West region and less concentrated in the Northeast and Midwest than overall employment.

Figure 7: A Majority of Homebase Firms Are in the Food and Drink or Retail Industries

This figure shows the share of Homebase firms in each of the nine industry categories that Homebase uses. As might be expected given that Homebase offers scheduling and time-card software, Homebase firms are predominantly in industries like food & drink and retail that employ many hourly workers.

Figure 8: Almost All Homebase Firms Have 50 Full-Time Equivalent Employees or Less

This figure shows the share of Homebase firms in each of five size categories. We define categories based on the number of hours worked at the firm between January 19 and February 1st, which we convert to full-time equivalent (FTE) workers by dividing by 40 hours per week. Roughly 60 percent of Homebase firms have one to five FTE employees (i.e., less than 400 hours worked in our two base period) and the vast majority have fifty or fewer FTE employees.


Week Three: Update with Homebase Data Through April 25

Below, we update our four figures analyzing the labor market impacts of COVID-19 on small businesses using Homebase time-card information with data through April 25.

Broadly, the patterns remain similar as before. We’ve added additional figures, 5 through 7, that provide information on three characteristics of the firms in the Homebase data: the regions they are located in, the industries they operate in, and the number of employees they have. These figures are included to help readers better understand the Homebase sample and interpret our results.

We will update these facts frequently to track these patterns over time and add new information as the COVID-19 situation develops. We expect the next update to take place in a few days, when we have data extending through May 5.

An up-to-date version of this summary will be maintained here. These analyses build upon the work done by Homebase itself on their blog, where they have provided frequent analyses on what their data is telling us about labor market developments.

Fact 1: Firms have dramatically reduced employee hours

Figure 1 plots the evolution of total hours per week worked among firms in our sample from January 19 through April 25 (the last full week in the data). Each sub-plot shows the distribution of hours across firms, measuring each firm relative to its average hours per week in a base period of January 19 – February 1. Through early-March, the distribution of hours is centered around one, corresponding to stable hours worked, with a few firms shutting down (at least for the week) and reporting zero hours but little net change on average. The week of March 8, the distribution of hours starts to shift slightly left and then the week of March 15 the distribution shifts dramatically left. Most firms have significantly fewer hours than in the base period and over 15 percent of firms shut down entirely. By the week of March 29, over 40 percent of firms shut down entirely, with zero recorded hours, and many of the remaining firms having quite large reductions in hours.

Fact 2: Hours reductions vary by ability of industry to operate under stay-at-home orders


Figure 2 investigates how hours reductions vary by firm industry. We see that reductions in hours are largest in Beauty and Personal Care and Leisure and Entertainment, where hours have declined over 90 percent. Hours declines are smallest in industries like Home and Repair and Transportation. However, even in these industries hours have declined by around 50 percent. Broadly, the magnitudes of industry-level job declines appear to closely map the extent to which an industries’ workers are “essential” according to government social distancing orders, and whether consumption requires in-person interaction (i.e. whether remote work is possible).

Fact 3: Hours start falling earlier in states with stay-at-home orders, but start falling sharply by March 16 in almost all states


Figure 3 explores how hours reductions vary across states by the timing of the announcement of a stay-at-home or shelter-in-place orders (if any) by industry. We group states into four categories, those that announced shelter in place orders on or before March 22, those that announced shelter in place orders between March 23 and March 30, those that announced shelter in place orders on March 31 or later, and those states that have yet to announce shelter-in-place or stay-at-home orders.

Fact 4: Hours reductions are primarily explained by firm shutdowns and hours reductions, not layoffs


Figure 4 separates the total hours reductions documented in Figures 1 – 3 into three channels: shutdowns, layoffs, and cuts in hours. We define firms as having fully shut down in a given week if the Homebase data records zero employees clocking in at that firm during that week. We identify a worker as having been laid off in a given week if that employee works zero hours at a firm which is still operating. We define hours cuts as the reduction in hours, relative to that initial baseline, among workers still employed at still operating firms. The figure distinguishes which fraction of the percent change in hours each week since early February is attributable to these three forms of hours reductions. The total number of hours worked in the first week of April are less than half what they were in late January. Most of that reduction is due to firms fully shutting down or asking retained employees to work fewer hours. A smaller percentage is due to firms laying off a portion of their workforce. This suggests that the principal driver of unemployment claims is total firm shut downs. It also suggests that even still employed workers are suffering a cutback in their hours.
One important caveat to this decomposition is what we refer to as a firm shut-down is a shut-down of Homebase measured employment. If firms employ workers that do not schedule their time using Homebase and some of these workers remain employed, some of the hours losses that we attribute to shut-downs may instead be properly attributed to layoffs.

Describing Homebase Firm Characteristics

Our analysis is based on firms that use Homebase, which are not representative along many dimensions. They tend to be smaller than average, and concentrated in particular industries (such as food and drink and retail).

Below, we provide information on three characteristics of the firms in the Homebase data: the regions they are located in, the industries they operate in, and the number of employees they have. This information is based on firm characteristics between January 19 and February 1 and includes all firms operating during this time period, regardless of whether they have continued to operate through the COVID-19 crisis or not.

Fact 5:Homebase firms broadly match the distribution of employment across US regions


This figure shows the share of Homebase firms located in each Census region (in light blue) and compares these shares to each region’s share of total employment (in dark blue), as measured by the Bureau of Labor Statistics (BLS) data. Homebase firms are somewhat concentrated in the West region and less concentrated in the Northeast and Midwest than overall employment.

Fact 6: A majority of Homebase firms are in the food and drink or retail industries


This figure shows the share of Homebase firms in each of the nine industry categories that Homebase uses. As might be expected given that Homebase offers scheduling and time-card software, Homebase firms are predominantly in industries like food & drink and retail that employ many hourly workers.

Fact 7: Almost all Homebase firms have 50 full-time equivalent employees or less


This figure shows the share of Homebase firms in each of five size categories. We define categories based on the number of hours worked at the firm between January 19 and February 1st, which we convert to full-time equivalent (FTE) workers by dividing by 40 hours per week. Roughly 60 percent of Homebase firms have one to five FTE employees (i.e., less than 400 hours worked in our two base period) and the vast majority have fifty or fewer FTE employees.

<h2


Week Two: Update with Homebase Data Through April 11

Below, we update our four figures analyzing the labor market impacts of COVID-19 on small businesses using Homebase time-card information with data through April 11. (Our previous analysis used data through March 28.)

Broadly, the patterns remain similar as before, but three features of these updated data are worth highlighting:

First, hours losses appear to have stabilized somewhat post March 28. States announcing shelter-in-place orders later than other places have not fully converged to those states announcing shelter in places earlier.

Second, there appears to be within-week variation in the effects of COVID-19 on the labor market, with the reduction in total hours worked being larger on the weekend than on weekdays. This suggests firms are disproportionately reducing weekend hours.

Third, we are more confident than we were last week that firms are primarily implementing reductions either via total shutdowns, with no workers recording positive hours in a week, or by reducing hours of all workers without substantial layoffs. This pattern showed up in Figure 4 in our post last week, but we added a caveat that it could have been a statistical artifact due to the fact that workers who had hours in the first half of the week and were then laid off would initially appear in the “hours cut” category and only appear in the correct “layoff” category in the following week. That seems not to have happened: The layoffs category grew only a small amount in the week ending April 11, and the “hours cut” category shrunk only a bit. It seems that a large share of the still-open firms have distributed hours reductions among their workers without substantial layoffs.

We will update these facts frequently to track these patterns over time and add new information as the COVID-19 situation develops.

An up-to-date version of this summary will be maintained here. These analyses build upon the work done by Homebase itself on their blog, where they have provided frequent analyses on what their data is telling us about labor market developments.

Fact 1: Firms have dramatically reduced employee hours

Figure 1 plots the evolution of total hours per week worked among firms in our sample from January 19 through April 11 (the last full week in the data). Each sub-plot shows the distribution of hours across firms, measuring each firm relative to its average hours per week in a base period of January 19 – February 1. Through early-March, the distribution of hours is centered around one, corresponding to stable hours worked, with a few firms shutting down (at least for the week) and reporting zero hours but little net change on average. The week of March 8, the distribution of hours starts to shift slightly left and then the week of March 15 the distribution shifts dramatically left. Most firms have significantly fewer hours than in the base period and over 15 percent of firms shut down entirely. By the week of March 29, over 40 percent of firms shut down entirely, with zero recorded hours, and many of the remaining firms having quite large reductions in hours.

Fact 2: Hours reductions vary by ability of industry to operate under stay-at-home orders

Figure 2 investigates how hours reductions vary by firm industry. We see that reductions in hours are largest in Beauty and Personal Care and Leisure and Entertainment, where hours have declined over 90 percent. Hours declines are smallest in industries like Home and Repair and Transportation. However, even in these industries hours have declined by around 50 percent. Broadly, the magnitudes of industry-level job declines appear to closely map the extent to which an industries’ workers are “essential” according to government social distancing orders, and whether consumption requires in-person interaction (i.e. whether remote work is possible).

Fact 3: Hours start falling earlier in states with stay-at-home orders, but start falling sharply by March 16 in almost all states

Figure 3 explores how hours reductions vary across states by the timing of the announcement of a stay-at-home or shelter-in-place order (if any). We group states into four categories, those that announced shelter in place orders on or before March 22, those that announced shelter in place orders between March 23 and March 30, those that announced shelter in place orders on March 31 or later, and those states that have yet to announce shelter-in-place or stay-at-home orders.

Fact 4: Hours reductions are primarily explained by firm shutdowns and hours reductions, not layoffs

Figure 4 separates the total hours reductions documented in Figures 1 – 3 into three channels: shutdowns, layoffs, and cuts in hours. We define firms as having fully shut down in a given week if the Homebase data records zero employees clocking in at that firm during that week. We identify a worker as having been laid off in a given week if that employee works zero hours at a firm which is still operating. We define hours cuts as the reduction in hours, relative to that initial baseline, among workers still employed at still operating firms. The figure distinguishes which fraction of the percent change in hours each week since early February is attributable to these three forms of hours reductions. The total number of hours worked in the first week of April are less than half what they were in late January. Most of that reduction is due to firms fully shutting down or asking retained employees to work fewer hours. A smaller percentage is due to firms laying off a portion of their workforce. This suggests that the principal driver of unemployment claims is total firm shut downs. It also suggests that even still employed workers are suffering a cutback in their hours.
One important caveat to this decomposition is what we refer to as a firm shut-down is a shut-down of Homebase measured employment. If firms employ workers that do not schedule their time using Homebase and some of these workers remain employed, some of the hours losses that we attribute to shut-downs may instead be properly attributed to layoffs.


Week One: Initial Analysis and Methodology

COVID-19, and the policies enacted in response to the disease, have resulted in dramatic changes in many aspects of American society. These changes have been particularly large in the labor market. It has been challenging to understand the magnitude of these changes because standard data sources become available only with a lag of several weeks – we will not receive data on employment and unemployment after shelter-in-place orders took effect until the first week of May. Below, we take advantage of granular data on exact hours worked among employees of firms that use the Homebase scheduling software to provide an up-to-date picture of the labor market impact of COVID-19. We measure how the impact varies across geography and industry, how it evolves in response to state and local social distancing guidelines and orders, and how concentrated it is among particular sets of workers.

Homebase provides scheduling and time clock software to tens of thousands of small businesses employing hundreds of thousands of workers across the US and Canada. This scheduling software generates granular data on exact hours worked every day for all hourly employees at customer firms, providing a much higher-frequency and more detailed picture of employment and hours than traditional labor market datasets. This greater detail and higher frequency come at some costs; Homebase’s customer base is disproportionately composed of small firms in food service, retail, and other sectors that employ many hourly workers.  The data exclude most salaried employees, firms who do not require this type of scheduling software for their operations, and larger firms who would use their own software for this purpose.  Consequently, insights derived from the Homebase data should be viewed as relevant to hourly workers in small and medium sized businesses, rather than to the labor market at large.

Despite these limitations, we think these data provide valuable information for two main reasons. First, hourly workers at small businesses may be most vulnerable to the economic disruption caused by COVID-19, so understanding how they are being impacted is of significant policy importance. Second, hourly employees are one of the easiest employment margins for firms to adjust, so measuring changes in outcomes for these workers may provide a more real-time picture of the labor market than other types of workers whose hours firms find more difficult to adjust.

Below, we describe four facts about the effects of COVID-19 on the labor market for hourly workers at small businesses. We will update these facts frequently to track these patterns over time and add new information as the COVID-19 situation develops. An up-to-date version of this summary will be maintained here. These analyses build upon the work done by Homebase itself on their blog, where they have provided frequent analyses on what their data is telling us about labor market developments.

Fact 1: Firms have dramatically reduced employee hours

Figure 1 plots the evolution of total hours per week worked among firms in our sample from January 19 through March 28.  Each sub-plot shows the distribution of hours across firms, measuring each firm relative to its average hours per week in a base period of January 19 – February 1. Through early-March, the distribution of hours is centered around one, corresponding to stable hours worked, with a few firms shutting down (at least for the week) and reporting zero hours but little net change on average. The week of March 8, the distribution of hours starts to shift slightly left and then the week of March 15 the distribution shifts dramatically left. Most firms have significantly fewer hours than in the base period and over 15 percent of firms shut down entirely. In the week of March 22, over 40 percent of firms shut down entirely, with zero recorded hours, and many of the remaining firms having quite large reductions in hours. Fully 91 percent of firms had fewer hours that week than in the base period.  

Fact 2: Hours reductions vary by ability of industry to operate under stay-at-home orders

Figure 2 investigates how hours reductions vary by firm industry.  We see that reductions in hours are largest in Beauty and Personal Care and Leisure and Entertainment, where hours have declined over 90 percent. Hours declines are smallest in industries like Home and Repair and Transportation.  However, even in these industries hours have declined by around 50 percent. Broadly, the magnitudes of industry-level job declines appear to closely map the extent to which an industries’ workers are “essential” according to government social distancing orders, and whether consumption requires in-person interaction (i.e. whether remote work is possible).

Fact 3: Hours start falling earlier in states with stay-at-home orders, but start falling sharply by March 16 in almost all states

Figure 3 explores how hours reductions vary across states by the timing of the announcement of a stay-at-home or shelter-in-place order (if any).  We group states into four categories, those that announced shelter in place orders on or before March 22, those that announced shelter in place orders between March 23 and March 30, those that announced shelter in place orders on March 31 or later, and those states that have yet to announce shelter-in-place or stay-at-home orders.  Although reductions in hours have been large throughout the country, with states in all four categories experiencing total hours declines of at least 50 percent, there are differences in the magnitude of hours declines between states that announced shelter in place or stay-at-homes earlier than others.  By March 28, total hours have declined by over 70 percent in states with the earlier stay at home orders, over 15 percentage points more than states that have not issued stay-at-home orders.  The timing of the onset of hours declines is fairly similar across categories of states – although hours started falling three or four days earlier and fell more rapidly in the states implementing shelter-in-place and stay-at-home policies the earliest. [1]

Fact 4: Hours reductions are primarily explained by firm shutdowns and hours reductions, not layoffs

Figure 4 separates the total hours reductions documented in Figures 1 – 3 into three channels: shutdowns, layoffs, and cuts in hours. We define firms as having fully shut down in a given week if the Homebase data records zero employees clocking in at that firm during that week. We identify a worker as having been laid off in a given week if that employee works zero hours at a firm which is still operating. We define hours cuts as the reduction in hours, relative to that initial baseline, among workers still employed at still operating firms. The figure distinguishes which fraction of the percent change in hours each week since early February is attributable to these three forms of hours reductions. The total number of hours worked in the last week of March are less than half what they were in late January. Most of that reduction is due to firms fully shutting down or asking retained employees to work fewer hours. A smaller percentage is due to firms laying off a portion of their workforce. This suggests that the principal driver of unemployment claims is total firm shut downs. It also suggests that even still employed workers are suffering a cutback in their hours.

One important caveat to this decomposition is what we refer to as a firm shut-down is a shut-down of Homebase measured employment. If firms employ workers that do not schedule their time using Homebase and some of these workers remain employed, some of the hours losses that we attribute to shut-downs may instead be properly attributed to layoffs. Another caveat is that the “hours cut” category includes all workers with positive hours during a given week. Firms that shut down in the middle of a week, as well as workers who are laid off mid-week, will be counted in this category in the first week, and will not appear in the correct category until the first week in which they have zero hours.

Methodology

Our analyses are based on data on hours worked at the establishment-worker-day level generously made available by Homebase. These data extend from January 1, 2020 through March 28, 2020. We aggregate the Homebase data to the firm-MSA-industry-day level.  We restrict the sample to firms whose employees worked at least 80 hours between January 19 and February 1 and to states for which we observe at least 50 such firms.  We refer to this two-week window as the “base period.” All analyses weight firms by their total hours during the base period.

In our analyses of weekly outcomes (e.g., Figures 1 and 4), we normalize each firm’s hours by dividing by the average hours worked per week over the base period at the firm. In our analyses of daily outcomes (e.g. Figures 2 and 3), we normalize by dividing by the average value of the outcome at the given firm on the same day of the week during our base period.  For example, if total hours for a firm on Friday, March 13 was 100 and total hours for the same firm on Friday, January 24 and Friday, January 31 was 300, (150 on each day), the outcome variable total hours’ value would be .66. (This is 100 divided by (300/2), the average Friday hours in the base period.)

We use the data compiled by The New York Times on the timing of stay-at-home and shelter-in-place orders in different states.

[1] It’s important to emphasize that reducing activity and interactions between people is the goal of stay-at-home and shelter-in-place orders, so the earlier reduction in employment, particularly among businesses involving significant closer-contact such as Beauty and Personal Care and Leisure and entertainment, can be interpreted as the laws working as intended.  Other work, for example from Villas-Boas, et al (2020) suggests that, by reducing activity and mobility, stay-at-home and shelter-in-place orders may reduce hospitalizations and death due to COVID-19.

Authors

Alexander W. Bartik, Assistant Professor Economics, University of Illinois at Urbana-Champaign, and Research Affiliate, UChicago’s Poverty Lab
Marianne Bertrand, Chris P. Dialynas Distinguished Service Professor of Economics, University of Chicago Booth School of Business, and Faculty Director, Chicago Booth’s Rustandy Center for Social Sector Innovation and UChicago’s Poverty Lab
Feng Lin, Research Professional, Chicago Booth
Jesse Rothstein, Professor of Public Policy and Economics, University of California, Berkeley, and Director, Institute for Research on Labor and Employment (IRLE) and California Policy Lab
Matt Unrath, PhD Candidate, Goldman School of Public Policy, UC Berkeley, and Research Fellow, California Policy Lab

Acknowledgments

We thank Homebase and Ray Sandza in particular for generously allowing access to their data and sharing their time to answer questions and help us understand the data. We also thank Jingwei Maggie Li, Salma Nassar, and Greg Saldutte at Booth’s Rustandy Center for Social Sector Innovation and Manal Saleh at the Poverty Lab for excellent assistance on this project and Michael Stepner for comments.

Share: