Working Remotely and the Supply-Side Impact of COVID-19 (Papanikolaou and Schmidt, NBER Working Paper 27330)
Summary
NBER Working Paper No. 27330, Working Remotely and the Supply-side Impact of Covid-19, by Dimitris Papanikolaou and Lawrence D.W. Schmidt, dated June 2020 and published by the National Bureau of Economic Research. The paper builds a Covid-19 work exposure measure for each industry from American Time Use Survey data on whether workers can work from home, and classifies some industries as essential. It reports that a one-standard-deviation increase in exposure is associated with an approximately 8 percent larger decline in employment, an 8 percent decline in analysts' revenue forecasts for Q2 and a 0.22 percentage point increase in the probability of default. It also constructs a long-short Covid-19 factor portfolio, which the authors state had lost roughly 50 percent of its value as of May 15, 2020. The paper closes with Table A.2, listing the factor's portfolio weights by NAICS industry.
Summary drafted by a model from the document's text below and checked by script against that text before publication. It is a navigation aid, not a reading of what the document proves. Where AI is used
Cite this document
“Working Remotely and the Supply-Side Impact of COVID-19 (Papanikolaou and Schmidt, NBER Working Paper 27330).”
Original at doi.org. Archived at Pandemic Darlings, https://pandemicdarlings.com/source-documents/working-remotely-and-the-supply-side-impact-of-covid-19-8209d2cde0/
, SHA-256 73afa10ec85d33de080d40605364cf8c7041789d05e2d700782e5287f694973a.Full text
NBER WORKING PAPER SERIES
WORKING REMOTELY AND THE SUPPLY-SIDE IMPACT OF COVID-19
Dimitris Papanikolaou
Lawrence D.W. Schmidt
Working Paper 27330
http://www.nber.org/papers/w27330
NATIONAL BUREAU OF ECONOMIC RESEARCH
1050 Massachusetts Avenue
Cambridge, MA 02138
June 2020
We thank Carola Frydman for helpful comments and discussions. We are grateful to Nathaniel
Barlow and Jorge Colmenares-Miralles for excellent research assistance and Evan Soltas,
SafeGraph, and the Risk Management Institute (RMI) of the National University of Singapore for
sharing their data. Please see https://sites.google.com/site/lawrencedwschmidt/covid19 for
additional information and data related to the analysis conducted in this paper. The views
expressed herein are those of the authors and do not necessarily reflect the views of the National
Bureau of Economic Research.
NBER working papers are circulated for discussion and comment purposes. They have not been
peer-reviewed or been subject to the review by the NBER Board of Directors that accompanies
official NBER publications.
© 2020 by Dimitris Papanikolaou and Lawrence D.W. Schmidt. All rights reserved. Short
sections of text, not to exceed two paragraphs, may be quoted without explicit permission
provided that full credit, including © notice, is given to the source.
Working Remotely and the Supply-side Impact of Covid-19
Dimitris Papanikolaou and Lawrence D.W. Schmidt
NBER Working Paper No. 27330
June 2020
JEL No. E23,G10,J00
ABSTRACT
We analyze the supply-side disruptions associated with Covid-19 across firms and workers. To
do so, we exploit differences in the ability of workers across industries to work remotely using
data from the American Time Use Survey (ATUS). We find that sectors in which a higher
fraction of the workforce is not able to work remotely experienced significantly greater declines
in employment, significantly more reductions in expected revenue growth, worse stock market
performance, and higher expected likelihood of default. In terms of individual employment
outcomes, lower-paid workers, especially female workers with young children, were significantly
more affected by these disruptions. Last, we combine these ex-ante heterogeneous industry
exposures with daily financial market data to create a stock return portfolio that most closely
replicate the supply-side disruptions resulting from the pandemic.
Dimitris Papanikolaou
Kellogg School of Management
Northwestern University
2211 Campus Drive, Office 4319
Evanston, IL 60208
and NBER
d-papanikolaou@kellogg.northwestern.edu
Lawrence D.W. Schmidt
Sloan School of Management
Massachusetts Institute of Technology
100 Main Street
Cambridge, MA
ldws@mit.edu
The Covid-19 pandemic of 2020 has led to severe disruptions to the supply-side of the
world economy as entire sectors shut down. In the first quarter of 2020, US gross domestic
output fell by 4.8 percent in annualized terms, a decline not seen since the Great Recession.
This drop underestimates the full economic impact of the pandemic since the severity of the
crisis became fully apparent to the public and private sector only in the last few weeks of
March. Moreover, the effect has been highly asymmetric: restaurants, entertainment, and
travel services suffered significantly more than food or technology services. Naturally, these
declines reflect not only the supply side disruptions due to the effect of the lockdown but
also demand-side factors, including the collapse of global consumer demand and expectations
of future government policy. This confluence of adverse forces obscures the direct effects of
the supply-side disruptions of the pandemic.
Our goal is to isolate the supply side effects from other forces. Our starting point is
that industries in which a higher fraction of the labor force can work remotely are likely to
experience less severe disruptions in their business operations. We build on existing work
and construct a metric of industry exposure to the lockdowns using information on the
share of the workforce than can work from home (Dingel and Neiman, 2020; Alon, Doepke,
Olmstead-Rumsey, and Tertilt, 2020). Specifically, we follow Alon et al. (2020) and exploit
data from the American Time Use Survey (ATUS) in 2017 and 2018, in which workers disclose
the extent to which they are able to and have historically had experience working remotely.
As noted by Alon et al. (2020), occupations vary immensely by whether people report they
are able to telecommute—ranging from 3% for transportation and material moving to 78%
for computer programmers. We then aggregate these survey responses to build measures of
exposure across industries and groups of workers.
In brief, our assumption is that the extent to which firms are able to effectively continue
their business operations during the pandemic depends on whether workers can perform
their tasks remotely. As such, our measure of supply side disruption for a given industry,
termed “Covid-19 work exposure”, is equal to one minus the fraction of workers that have
telecommuted – more specifically, the fraction that have ever worked full days from home
– in each industry. Importantly, there is considerable dispersion in our exposure measure
across industries. For example, software publishers (NAICS 5112) have an exposure of just
0.38, since much of the production work can be done remotely; by contrast, meat production
(NAICS 3116) or general merchandise stores (NAICS 4523) have an exposure close to one,
since most of the employees cannot perform their work remotely.
In addition, government restrictions also play a role in firms’ ability to continue operations.
Specifically, local governments typically deem certain industries as ‘critical’, namely those
1
that provide ‘essential infrastructure’. Thus, we also manually classify some industries as
‘essential’. Since the definition of essential industries varies greatly across states, we aim
to be conservative, classifying as essential industries related to the production and sale of
food and beverages; utilities; pharmacies; transportation; waste collection and disposal; and
some healthcare and financial services. Data on foot-traffic from SafeGraph validate our
construction: establishments in industries deemed critical experience significantly smaller
declines in traffic than establishments in non-critical industries, which stands in stark contrast
to nearly identical trends in foot traffic in January-February 2020.
Armed with a measure of industry exposure to these supply-side disruptions, we can
isolate these effects from other economic forces that would otherwise affect these sets of firms
symmetrically. Our summary finding is that these supply-side disruptions are economically
large.
Our first set of results concerns employment outcomes. We find that sectors with a larger
fraction of workers who cannot work remotely—higher Covid-19 work exposure—experienced
significantly larger declines in employment than sectors where more of the workforce can
perform tasks remotely. The differences are economically sizable: a one-standard-deviation
increase in our Covid-19 work exposure measure is associated with an approximately 8 percent
larger decline in employment; the differences among sectors is starker when we restrict the
sample to non-critical industries (13 percent). We also find some evidence for heterogeneous
effects across workers. Specifically, a given increase in our Covid-19 work exposure measure is
associated with a somewhat greater increase in the probability of non-employment for women
and lower-earning workers. Among all affected groups, we find that the employment status
of female workers with young children and without a college degree is most sensitive: in the
cross-section of workers, a one-standard-deviation increase in our Covid-19 work exposure
measure is associated with a 15 percent probability of non-employment for these workers,
which is more than three times the magnitude of the baseline coefficient.
Our second set of results focuses on firm outcomes. Given the delays in the availability of
real data on firm outcomes, we focus on a set of forward-looking variables that capture future
expectations. That is, we focus on the revisions in analyst forecasts of expected revenue
growth; the stock market performance of publicly traded firms; and the expected probability
of default. We find that firms in sectors that are more likely to experience work disruptions
also fare significantly worse during the 2020 pandemic: a one-standard-deviation increase
in our Covid-19 exposure metric is associated with a 8 percent decline in analysts’ revenue
forecasts for Q2; a 0.22 percentage point increase in the probability of default over the next 6
months; and a 7 percent decline in stock market performance to date. These magnitudes
2
account for a significant share of cross-industry differences in outcomes during this period.
Last, while financial analysts expect the worst effects to be short-lived, our work exposure
variable is still a significant predictor of differences in expected revenue growth over the
next two years—though magnitudes are significantly muted. Similar patterns emerge in the
differences in projected declines in revenue across non-critical versus critical industries, where
we find that analysts project firms in non-critical industries’ annual revenues to decline by
13, 10, and 8 percent for 2020, 2021, and 2022, respectively. These same differences are 3.2,
2.5, and 2.3 percent for firms in critical industries.
In the last part of the paper, we use some of these forward-looking variables to construct
a real-time indicator of supply-side disruptions. In particular, we use daily data on stock
returns to construct a portfolio that is maximally exposed to the Covid-19 work disruption
using the methodology of Fama and MacBeth (1973). The resulting ‘Covid-19’ factor has
a long-short portfolio interpretation. It overweights industries whose workers cannot work
remotely and underweights industries whose workers can perform their tasks from home. As
of May 15, 2020, this portfolio had lost roughly 50 percent of its value since the beginning
of the year—compared to 10 percent for the broad market index. Naturally, reversing this
investment strategy would deliver a portfolio that could significantly hedge future Covid-19
related uncertainty.
In sum, our work contributes to the voluminous economics literature that has emerged in
response to the pandemic. The key differentiator of our work is its focus on isolating the
supply side disruptions associated with Covid-19. By contrast, existing work has focused
on the overall response of the economy during this period.1 How firms respond during the
pandemic is a function of both the underlying supply-side disruptions (our focus) but also
firms’ exposures to a decline in consumer demand as a result of expected income losses and
increase in uncertainty. Our goal is to isolate the former rather the latter channel. In this
respect, our work is closer to Bonadio, Huo, Levchenko, and Pandalai-Nayar (2020) who
focus on the role of global supply chains; or Hassan, Hollander, van Lent, and Tahoun (2020)
who identify differences in firms’ exposure based on the transcript of analyst calls.
Last, our paper is related to recent work by Dingel and Neiman (2020), Alon et al. (2020)
and Mongey, Pilossoph, and Weinberg (2020). Dingel and Neiman (2020) construct measures
1
Recent work examines the response of employment (Coibion, Gorodnichenko, and Weber, 2020b; Cajner,
Crane, Decker, Grigsby, Hamins-Puertolas, Hurst, Kurz, and Yildirmaz, 2020; Campello, Kankanhalli, and
Muthukrishnan, 2020; Borjas and Cassidy, 2020; Fairlie, Couch, and Xu, 2020); firm revenue (Barrero, Bloom,
and Davis, 2020; Landier and Thesmar, 2020); firm closures (Bartik, Bertrand, Cullen, Glaeser, Luca, and
Stanton, 2020); stock market performance (Ding, Levine, Lin, and Xie, 2020); or consumer spending Baker,
Farrokhnia, Meyer, Pagel, and Yannelis (2020); Coibion, Gorodnichenko, and Weber (2020a) during this
period.
3
of the feasibility of workers to work from home, but they do so using task descriptions in the
ONET survey. Like us, Alon et al. (2020) construct a measure based on workers’ answers
to the American Time Use Survey (ATUS). Though the details of the construction differ,
the main idea is similar. Dingel and Neiman (2020) and Alon et al. (2020) mainly focus on
demographic differences among workers who can and cannot work remotely; by contrast, we
are interested in how the ability to work from home is related to outcomes. In this respect,
our work is closest to Mongey et al. (2020), who explore how the Dingel and Neiman (2020)
measure and a measure of occupations requiring high physical proximity relate to differential
outcomes across workers. Though the first part of our analysis also explores worker-level
outcomes, our primary focus is on firms. In this regard, we view our work as complementary.
1 Measuring Exposures of Production to Covid
We begin with a brief description of our construction of exposure measures from the data.
1.1 Workers’ ability to work remotely
Following the rapid increase in cases throughout the US, many employers as well as state and
local governments quickly imposed restrictions requiring that workers stay at home, leading
to what is essentially the largest global experiment in telecommuting in human history. Our
starting point is the simple premise that supply-side disruptions are likely to be more severe
in occupations/industries for which workers have had little to no ability to or experience with
telework in the past. Accordingly, our measure of industry exposure to work disruptions due
to Covid-19 builds on Alon et al. (2020), who use data from the Leave and Job Flexibilities
module of the American Time Use Survey (ATUS) in 2017 and 2018 (n = 10, 040), which
asks several questions about workers’ ability and past experiences with working from home.
Our preferred measure utilizes responses to two different questions, though the module also
includes several additional questions about reasons for and frequency of remote work. We
obtain the ATUS microdata from IPUMS.2
Crucially, the survey draws a distinction between workers who are able or unable to
telecommute, as opposed to actually regularly telecommuting, as the former is the relevant
metric during a pandemic.3 The output of this procedure is an occupation/industry-level
2
Sandra L. Hofferth, Sarah M. Flood, Matthew Sobek and Daniel Backman. American Time Use Survey
Data Extract Builder: Version 2.8 [dataset]. College Park, MD: University of Maryland and Minneapolis,
MN: IPUMS, 2020. https://doi.org/10.18128/D060.V2.8
3
For example, the Census’ American Community Survey asks whether workers worked from home last
week, which captures regular telecommuting behavior. According to the Census Bureau, around 4.3% of
4
metric of the fraction of workers in each occupation that should in principle be able to work
from home. The first major question asks for a yes/no reply to the prompt: “As part of your
(main) job, can you work at home?” Using the survey’s person weights, approximately 78%
of households answer yes to this question. Alon et al. (2020) construct a measure of telework
experience by identifying the share of workers in a given occupation that, according to their
answers to this question, are able to telecommute, and provide several facts about the share
of workers in various demographic groups which are employed in occupations which can be
performed from home.
Our preferred measure also makes use of responses to one additional question, “Are there
days when you work only at home?”, which is only asked to those who say that they are
able to work from home. Around 51% of households who are able to work from home also
indicate that they have worked days entirely from home. In our view, answering this question
in the affirmative provides a sharper classification of workers who will more likely be able to
perform the majority of their job responsibilities from home, as opposed to only a subset
of tasks (e.g., answering emails/phone calls) remotely. For example, 64% of computer and
information systems managers (occupation 110) say that they can work from home, and 47%
of them have worked days entirely from home. These same proportions are 59% and just
13%, respectively, for medical and health services managers (occupation 350), a group that,
in our view, is a part of a population which is more likely to be unable to work full-time from
home as effectively.4
In brief, we classify a surveyed worker as ‘able to work from home’ if they answer ‘yes’ to
both questions above. We then compute measures of industry and occupation exposure by
aggregating these survey responses across employees according to the industry and occupation
codes which appear in the ATUS. In brief, our measure of industry exposure is
Covid-19 Work ExposureI = 1 − % of workers able to work from homeI , (1)
which is computed as a weighted average of individual-level responses aggregated so as to be
nationally representative using the BLS’ person weights. If fewer than 5 survey respondents
workers worked from home according to this measure in 2010 (Mateyka, Rapino, and Landivar, 2012).
4
As further evidence of a sharper distinction, among households who provide non-missing replies to the
question “What is the main reason why you work at home?”, 60% of households who have worked days
only from home are about twice as likely to list “job requires working at home”, “reduce commuting time or
expense”, or “personal preference”, which is significantly higher than the 31% frequency for other workers (t
statistic on difference = 12.1). By way of contrast, in the group of workers who say that they have worked
from home but have never worked days only from home, the most common answer to this question is “finish
or catch up on work”, which is selected 36% of the time, a significantly higher than the 14% of the time for
those who have worked days from home (t=7.3).
5
are directly employed in a given Census industry, we instead extrapolate using a weighted
average of occupation-based measures.5 Given that an individual’s occupation likely provides
a more accurate description of the types of tasks that a worker performs on a daily basis,
we elect to use the occupation-based measure in person-level regressions below—though we
obtain similar results if we use the industry-based measure. In general, we also find similar
results if we use the more inclusive Alon et al. (2020) measure instead.
As most of our outcomes by industry use NAICS-based classifications, we crosswalk
between the ATUS industry codes and NAICS industries.6 We aggregate most outcomes,
when available, to the 4-digit NAICS level given that the industry codes available in the
IPUMS ATUS extract roughly correspond with this level of aggregation.
We find that most industries are highly exposed; the mean value of the measure is
approximately 85%. Yet, there is considerable dispersion across NAICS industries—the
cross-sectional standard deviation of our measure across 4-digit NAICS industries is approxi-
mately 17%. Clearly, some industries are more exposed than others. For example, General
Merchandise Stores (NAICS 4523) and Meat Production (NAICS 3116) have an exposure
measure of close to 1, since almost none of the workers in that industry report that they can
work from home. By contrast, Software Publishers (NAICS 5112) is an industry in which the
vast majority of workers report they can work from home, and accordingly has an exposure
of just 0.38.
In addition, we perform some manual adjustments to the ATUS work exposure measure.
For certain industries, we set their work from home exposure to 1, giving them “full exposure.”
These are largely industries that have been almost completely shut down and thus, even if
some of their workers have the flexibility to be able to work from home in normal times,
they cannot continue business as usual during lockdowns in a pandemic. These industries
include: 4811 Scheduled Air Transportation, 5121 Motion Picture and Video Industries, 5151
Radio and Television Broadcasting, 5615 Travel Arrangement and Reservation Services, 7111
Performing Arts Companies, 7112 Spectator Sports, 7113 Promoters of Performing Arts,
Sports, and Similar Events, 7114 Agents and Managers for Artists, Athletes, Entertainers,
and Other Public Figures, 7131 Amusement Parks and Arcades.
5
Specifically, we use data from the (considerably larger) 2016 American Community Survey to estimate
the share of employees for each occupation by industry. We choose 2016 so that the occupation and industry
codes coincide with those used in ATUS. American Community data are also taken from IPUMS: Steven
Ruggles, Sarah Flood, Ronald Goeken, Josiah Grover, Erin Meyer, Jose Pacas and Matthew Sobek. IPUMS
USA: Version 10.0 [dataset]. Minneapolis, MN: IPUMS, 2020. https://doi.org/10.18128/D010.V10.0
6
We make use of the crosswalks created by Evan Soltas, available here, which we update to use
the 2017 NAICS code system: https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi:
10.7910/DVN/O7JLIC
6
1.2 Critical Industries
While policy responses to the Covid-19 crisis have imposed severe restrictions on interactions
between individuals, governments have found it necessary to make some exceptions, thus
creating a need to classify industries as essential or non-essential. The Cybersecurity and
Infrastructure Security Agency (CISA) provided guidelines to states about what kinds
businesses should remain open. Keeping track of essential industries can enhance our analysis
of work from home exposure because if a business is allowed to stay open, then it may be
irrelevant if their workers can telecommute. We start with Pennsylvania’s guidelines largely
because they provide a list, based on the CISA guidance, of essential industries at the 4 digit
NAICS (2017) level.7 These industries primarily correspond to the production and sale of
food and beverages; utilities; pharmacies; transportation; waste collection and disposal; and
some healthcare and financial services.
While CISA’s essential industry classification is a good place to start, we alter it for
two basic reasons. First the classification of essential industries seems too coarse in certain
cases. For instance, the NAICS code 4831 is “Deep Sea, Coastal, and Great Lakes Water
Transportation” and is listed as essential by Pennsylvania. While some of the firms in
this industry are shipping companies, the largest firms in this industry are cruise ships,
which clearly have not been permitted to maintain business as usual. Second, the CISA
list was somewhat inclusive in what types of industries were classified as providing critical
infrastructure. We therefore removed from the critical list industries that faced severe
restrictions to operating at a scale close to or even higher than pre-pandemic baselines. For
example, restaurants, restaurant suppliers, and airlines have been permitted to remain open,
however their patronage has declined dramatically due to quarantines and heavy restrictions
on domestic and international passenger travel. While some restaurants remained open
for curbside and/or takeout service, it is apparent that they were not classified as critical.
Accordingly, we have also classified industries that are auxiliary to such industries (such as
firms in NAICS code 4244 , Grocery and Related Product Merchant Wholesalers, which
largely deliver food supplies to restaurants) as non-critical.
Given that we exclude the list of critical industries from the bulk of our empirical analysis,
our choices of narrowing down the critical industry list can be viewed as conservative.8
7
The CISA guidance is available at https://www.cisa.gov/publication/
guidance-essential-critical-infrastructure-workforce, and the full Penn-
sylvania list is available here: https://www.scribd.com/document/452553026/
UPDATED-5-45pm-March-21-2020-Industry-Operation-Guidance
8
As there are 311 industries represented by NAICS-4 codes, we will not to attempt to explain our
justification by industry, but hope to provide some clarification on our thought process, which is primarily
disciplined by the CISA guidance. In the case of manufacturing, for example, the CISA guidelines state that
7
Appendix Table A.1 provides the list of critical industries.
1.3 Validations using Foot Traffic Data
We validate our critical/non-critical industry classification and remote work exposure measures
using weekly foot traffic data from SafeGraph.9 SafeGraph collects anonymized information
on location activity from a large panel of mobile devices. For more than 5 million points of
interest (POIs), we can observe aggregated daily activity of users in the panel. As activity
exhibits strong within-week seasonal patterns, we aggregate the data to the weekly level.
Given our focus on workers’ ability to telecommute, we focus on visits to POIs that last for 4
hours or more, which we will refer to as “worker foot traffic”.10 For each 4-digit industry,
we compare activity in week t with a baseline level of average activity: the mean of weekly
activity over the period beginning with the week of January 6-12 and ending with the week
of February 10-16. In the baseline period, the median 4-digit NAICS industry receives 18,705
worker visits per week, and the distribution of activity is highly right skewed, with the 90th,
95th and 99th percentiles having 560 thousand, 1.03 million, and 5.60 million hits respectively.
If our critical industry classification is accurate, we would expect to see a significantly
smaller decline in foot traffic in critical industries than in non-critical ones —starting from
the middle of March 2020 when Covid-19 awareness and state-mandated lockdowns went into
effect. To test this hypothesis, we then aggregate foot traffic across all POIs to the 4-digit
the following workers should be deemed critical: ”Workers necessary for the manufacturing of metals (including
steel and aluminum), industrial minerals, semiconductors, materials and products needed for medical supply
chains and for supply chains associated with transportation, aerospace, energy, communications, information
technology, food and agriculture, chemical manufacturing, nuclear facilities, wood products, commodities used
as fuel for power generation facilities, the operation of dams, water and wastewater treatment, processing
and reprocessing of solid waste, emergency services, and the defense industrial base... ” (p. 17)
As steel and aluminum are mentioned explicitly and directly correspond to NAICS-4 codes 3312 (”Steel
Product Manufacturing from Purchased Steel”) and 3313 (”Alumina and Aluminum Production and Process-
ing”), we have included those as critical. We also deemed Soap, Cleaning Compound, and Toilet Preparation
Manufacturing (3256), Pharmaceutical and Medicine Manufacturing (3254), Plastics Product Manufacturing
(3261), Agriculture, Construction, and Mining Machinery Manufacturing (3331), and Medical Equipment
and Supplies Manufacturing (3391) as critical, as these clearly fall under the jurisdiction of the quoted CISA
guideline. However, as most other manufacturing NAICS-4 codes represent industries that manufacture both
products that would fit into the above categories as well as those that would not, we have largely chosen to
designate the remainder of the manufacturing industries as non-critical.
9
We are grateful to SafeGraph for making the data freely available to our research team and the
broader community for Covid-19 related work. Further information about the data are available here:
https://docs.safegraph.com/docs/weekly-patterns. See also, e.g., Mongey et al. (2020) and Farboodi,
Jarosch, and Shimer (2020) for additional applications of SafeGraph data to the study of Covid-19.
10
Specifically, SafeGraph makes available data on number of visits by “bucketed dwell time”, which captures
the length of time that a device is at each POI. Consistent with the recommendations of SafeGraph researchers,
we scale total weekly activity by a moving average the total number of devices in the panel over the prior
week for each point of interest prior to aggregating.
8
NAICS level so as to match our critical industry classification. Specifically, we estimate
I ( Total Worker Foot Traffic in week t in Industry I
0g
= 6, Critical Indust .
Pre-period Average Worker Foot Traffic of Industry i) de FOe Urttical Mndustry TE r.t
(2)
When estimating (2), we weigh observations by total industry employment, and, to minimize
the effect of outliers, we winsorize these industry indices at the lst and 99th percentiles.'!
Figure 2 plots the results. Panel A reports average levels of our activity indices for
critical and non-critical indices, respectively, along with 95% confidence intervals. Both
sets of industries followed essentially identical trends in January and early February, then
experienced a significant decline in foot traffic which dramatically accelerated in the middle
of March of 2020. Though the Covid-19 pandemic affected almost all establishments, we can
see that the decline is smaller for industries classified as critical. Panel B compares differences
in traffic between the two industries by plotting the estimated coefficient 6; in equation (2)
above. It shows that this decline was significantly smaller, by around 20%, for industries that
we identified as providing critical infrastructure. We interpret these results as supporting our
industry classification into ‘critical’ and ‘non-critical’ industries.
Our foot traffic data also permit a simple validation of our remote exposure measure.
Our hypothesis is that, even in critical industries, workers would prefer to work from home
to minimize their exposures to the disease to the extent that such activities are feasible.
Whereas workers in non-critical industries are essentially prevented from going to their
workplaces during, the decision to telework is more discretionary in nature for workers in
critical industries. Accordingly, we would expect to see larger declines in worker foot traffic
(ie increased telecommuting) in critical industries for which workers are more easily able to
telework. To test this premise, we estimate the following specification for critical (panel C)
and non-critical (panel D) industries,
I ( Total Worker Foot Traffic in week t in Industry I
)
= Covid E .
Pre-period Average Worker Foot Traffic of Industry i) CFT COVE Exposures FEr.t
(3)
where our prediction is that 7, > 0 since workers in critical, high exposure industries will be
more likely to go to work relative to their counterparts who can telecommute more easily.
Panel C provides strong evidence in support of this hypothesis: a 1% increase in the fraction
of workers who cannot work remotely is associated with between 1-1.5% higher foot traffic
'1Some industries have very sparse coverage in the pre-period, so drop the approximately 5% industries with
fewer than 30 average worker hits per week, though our results are insensitive to increasing this threshold or
eliminating it entirely. Total NAICS4 employment is taken from the Census’ 2017 Statistics of United States
Businesses tables: https: //www.census.gov/data/tables/2017/econ/susb/2017-susb-annual . html
throughout the month of April. In contrast, exposure is not correlated with changes in foot
traffic in non-critical industries, which is consistent with the non-critical distinction requiring
all workers to work remotely regardless of feasibility constraints.
1.4 Descriptive Facts
Figure 1 plots the distribution of the measure across NAICS4 industries. We report results
separately for critical and non-critical industries. We note that most industries have relatively
high values of Covid-19 work exposure; the average is equal to approximately 0.85 for critical
and 0.87 for non-critical industries. More importantly, however, we see there is considerable
dispersion in our measure—the cross-sectional standard deviation is approximately 17.4%.
The bulk of our paper explores the extent to which these cross-sectional differences are
predictive of differential economic outcomes during the pandemic.
Table 1 illustrates how our industry Covid-19 work exposure measure is related to firm
characteristics; due to data availability, we restrict attention to publicly-listed firms. Panel A
shows results for all industries, while Panel B restricts to the non-critical industry sub-sample.
Two patterns stand out. First, we see that firms in the high-exposure category tend to be
larger. For instance, focusing on the sub-sample of non-critical industries, the median firm
in the top quartile of the Covid-19 work exposure measure has 4.9 thousand employees; by
comparison, the median firm in the bottom quartile has 1.5 thousand employees. Second,
there is some evidence that firms in the least exposed industries have more intangibles than
firms in the most exposed industries: these firms have higher market valuations (median
Tobin’s Q of 2.3 for the most exposed vs 1.3 for the least exposed); have lower ratios of
physical capital (PPE) to book assets (0.16 vs 0.58); and spend significantly more on R&D
and SG&A than firms in the most exposed industries.12 This is consistent with the view that
firms with more intangibles have more jobs that can be done remotely than firms that rely
more on physical assets. Last, firms in the most exposed industries are more profitable in
terms of bottom-line accounting measures (return-on-assets) but this difference is driven by
higher spending on intangibles and/or fixed costs by firms in the least exposed industries.
12
Corrado, Hulten, and Sichel (2009) and Eisfeldt and Papanikolaou (2013) argue that higher SG&A
expenditures are associated with higher investment in intangibles. That said, Eisfeldt and Papanikolaou
(2013) focus on within-industry differences, whereas these results reflect between-industry patterns. As such,
these differences in SG&A could also reflect differences in operating leverage (importance of fixed costs) or
differences in accounting practices.
10
2 Covid-19 Work Exposure and Economic Performance
So far, we have constructed a measure of exposure to Covid-19 work disruptions using data
from the 2017 and 2018 ATUS. Here, we explore the extent to which this Covid-19 work
exposure measure is able to predict differences in economic performance across industries
and individual workers. Our outcome variables include data on employment outcomes at the
industry and worker level, as well as forward-looking variables that are indicative of future
economic performance—stock returns; analyst forecasts; and default probabilities.
Each of these variables has distinct advantages and disadvantages. Data from financial
markets and financial analysts are forward-looking, but are only available for public firms—
which may be less exposed to supply-side disruptions than smaller private firms. Employment
data are more representative of the universe of firms, but are available with a lag and may
not accurately reflect outcomes for firm owners.
2.1 Employment
We begin by exploring the extent to which our measure of production disruption due to
the pandemic can predict differences in employment declines across industries. We use two
sources for data on employment.
First, we focus on aggregated employment data from the Bureau of Labor Statistics (BLS).
Each month, the BLS aggregates data from large surveys of firms’ establishments in order to
construct estimates of total employment by industry. We make use of data from Table B-1A
of the April employment report, which provides estimates of seasonally-adjusted employment
at the 3-digit NAICS level for the vast majority of non-agricultural industries in the US
economy.13 While BLS reports monthly data for more detailed NAICS industries, measures
for April 2020 are suppressed for a large number of the more disaggregated categories. As
such, we conduct our analysis at the 3-digit level. For purposes of this analysis, we define a
NAICS3 industry as critical if more than 50% of its pre-pandemic employment is associated
with our list of 4-digit NAICS critical industries.
We estimate the following specification
Employment GrowthI = a + β Covid-19 Work ExposureI + εI . (4)
Here, the unit of observation is a 3-digit NAICS industry and the outcome variable is
employment growth. We examine both monthly but also annual changes—that is, between
13
Data were accessed here: https://www.bls.gov/web/empsit/ceseeb1a.htm
11
either March 2020 or April 2019 and April 2020. Since the size of these industries varies
greatly, depending on the specification, we also present results in which we weight industries
using total employment as of the baseline period used in each growth rate. Last, we report
results either using the full sample, or by excluding the list of critical industries in Table A.1.
Table 2 and Figure 3 present our results. Examining columns (1) through (8) of the table,
we see an economically and statistically significant correlation between our exposure measure
and industry-wide declines in employment that is robust across different specifications. In
particular, a one-standard deviation increase in our Covid-19 work exposure measure is
associated with a 6.2 to 23.0 percent greater decline in employment. The magnitudes are
comparable when we focus on monthly or annual growth rates and are larger when we exclude
critical industries and weigh industries by their total employment. These magnitudes are
quite significant and comparable to either the average decline in employment during this
period (14 percent) or the cross-sectional standard deviation (16 percent). In panel B, we
also report the differences in average employment growth rates between industries which
we classify as critical vs non-critical. On an employment-weighted basis, monthly declines
in employment are more than 3 times larger in non-critical industries (-22.8%) relative to
critical ones (6.0%).
Figure 4, Panel A, provides a more aggregated summary of the relationship between
our exposure measure and BLS employment year-on-year employment growth. Specifically,
we aggregate across non-critical industries by major 2-digit NAICS sector. Even at this
level of aggregation, one can observe a strong relationship between exposure and changes in
employment. Unsurprisingly, the three hardest-hit sectors are retail, hotels, and entertainment,
which are also the three most highly exposed sectors according to our measure.
The recent availability of CPS data allows us to explore outcomes at the level of individual
workers. The advantage of doing so is twofold. First, by observing the characteristics of
individual workers we can control for some variables that may be correlated with our Covid-19
work exposure. For instance, workers that can work remotely tend to be in white collar
occupations (see e.g., Dingel and Neiman, 2020). By including controls for workers’ level of
education, or past earnings, we can ensure that we are comparing otherwise similar workers.
Further, analyzing outcomes for individual workers, and how these outcomes vary with
observables, reveals a fuller picture of the effects of the Covid-19 on the supply side.
We use the April version of the Current Population Survey (CPS) which contains employ-
ment information for individual workers.14 We restrict the sample to adults of working age
14
There are some possible selection issues. IPUMS notes that: “Interviews for April were conducted
exclusively by phone. The two Census Bureau call centers that usually assist with the collection of CPS data
remained closed. Response rates continued to be low in April, over 10 percentage points below the average
12
(25 to 60 years) that are present in the March 2020 survey and who report that they were
“at work last week” as of March 2020. Our sample contains 23,984 workers. In April of 2020,
19,664 out of these workers report they are at work; 2,144 report they are out of a job; while
1,257 report they have a job but were not at work last week. We include the first and last
groups in our definition of employment for our regressions below.
We impose one other substantial restriction of the sample. As of the start of the year, the
BLS changed its coding of occupations relative to prior years, and a crosswalk between the
older vintage of occupation codes from ATUS and the newer codes in the 2020 CPS has not
yet been developed. In part due to this technical reason and also given that several researchers
have already documented a disproportionate increase in unemployment for lower income
workers, we restrict attention to the subsample of workers who appear in the March/April
2020 core CPS and March 2019 Annual Social and Economic (ASEC) CPS surveys. This
enables us to construct controls for an individual’s prior income and, motivated by evidence
on potentially concentrated adverse effects for small businesses, firm size. For each worker,
we assign an exposure measure based on her March 2019 occupation which is equal to one
minus the fraction of ATUS respondents (weighted using the ATUS sampling weights) who
had worked a full day from home.15
We estimate the following specification
NotEmployedi = a + β Covid-19 Work Exposurei + c Zi + εi . (5)
Our main outcome variable is a dummy variable that takes the value of 1 if the worker
was not ‘at work last week’ in the April 2020 CPS survey—which therefore includes both
unemployed and furloughed workers. For ease of interpretation, we multiply the dependent
variable by 100 to express probabilities in percentage points in tables and figures.
Since the CPS contains information on workers’ occupation, we use the occupation-level
measure of our Covid Work exposure rather than the industry aggregate—though, using
industry aggregates produces largely comparable results. Depending on the specification, the
vector of worker-level controls and interaction terms Zi includes gender dummies; college
education; whether the worker has children younger than 14 years old; education (a college
graduate dummy); worker age; worker earnings as of 2019 (quartile dummies); and firm size
(see below). Response rates continued to be particularly low among rotation groups one and five who would
have normally received a visit from the enumerator. Month-in-sample one and five response rates were similar
to those same groups in March. Additionally, those households that entered the survey for the first time in
March had a similarly low response rate for their second interview in April.”
15
Note that we also assigned an alternative measure based on each worker’s March 2020 industry and
obtain similar results.
13
dummies. We cluster the standard errors by occupation and industry (using Census codes).
Table 3 shows results.
Examining Table 3, several facts stand out. First, as we see in the first three rows
of the table, workers in occupations that are less able to work remotely (higher Covid-19
work exposure) are more likely to stop working. These effects are not absorbed by worker
characteristics—as we see in columns (2) and (3)—suggesting that our Covid-19 work exposure
measure is not simply capturing differences in the composition of the work force. Further, the
point estimates are somewhat larger for non-critical industries relative to critical industries.
Focusing on column (3)—which includes controls for worker gender, age, education, prior
earnings, and firm size—a one-standard deviation increase in our covid-19 work exposure
measure is associated with a 4.5 percentage point increase in the probability of a worker
in a non-critical industry being without employment in April 2020, compared to a 1.0
percentage point increase for a worker in a critical industry, that is however statistically not
significantly different from zero. Moreover, relative to workers in non-critical industries with
similar characteristics, employees in critical industries are 8 percentage points less likely to
remain employed. Hence, in the remainder of the paper, we restrict attention to non-critical
industries.
One potential concern with these results is that they could simply reflect differential
employment trends in white-collar versus blue-collar occupations. Column (4) of Table 3
performs a placebo exercise, in which we repeat the analysis for the February CPS sample—
well before the effects of the pandemic became apparent. Examining the results we note that
none of the coefficients of interest are statistically different from zero.
We next allow the coefficient β on the Covid-19 work exposure measure in equation (5) to
vary with worker characteristics Zi . In all specifications with interactions, we always include
dummies for all levels of the categorical variable of interest as controls. Doing so reveals
how the employment status of different workers varies in response to the same level of work
disruption due to Covid-19.16
Figure 5 summarizes our findings. In Panel A of the figure, we see an economically and
statistically significant difference in how the employment status of workers of different income
levels is related to our Covid-19 exposure measure. A one-standard deviation increase in our
work exposure measure is associated with a 9.2 percentage point increase in the likelihood of
non-employment for workers at the bottom quartile of the earnings distribution—compared
to just a 1.6 percentage point increase for workers in the top quartile. The fact that we
16
In additional specifications, we verify that many of these patterns persist even if we include occupation
and/or industry fixed effects, implying that we can detect similar differences even within occupations/industries.
These results are suppressed for brevity and are available upon request.
14
see such a differential response of employment status to the feasibility of remote work for
employees with different earnings levels has several interpretations. One possibility is that
firms try to retain their most able—and highest paid—employees and reduce employment for
workers of lower skill levels first.
Another possibility is that these responses reflect differences in age or the gender pay
gap. Panel B of the Figure shows that there is some evidence that the employment status
of younger workers responds more by approximately a factor of two than older workers to
Covid-19 work disruptions, but the coefficients are imprecisely estimated so the difference is
not statistically different from zero. Panels C and D condition on gender as well as college
attainment and whether or not the household has at least one child under the age of 14. In
both cases, we see that estimated sensitivities for women are larger relative to men, though
these differences are not always statistically significant. Workers without college degrees have
higher estimated coefficients. In panel D, we observe that whereas men with young children
and work in non-critical industries exhibit have smaller coefficients than their men without
young children, the opposite appears to be the case for women with young children. The
7.7 percentage point difference in coefficients between men and women with young children
is highly significant (t=3.4).17 Panel E shows that these gender differences do not merely
reflect the gender pay gap. More importantly though, Panel E reveals that the employment
status of female workers is more sensitive to work disruptions than the status of male workers,
for all income levels. Panel F shows that the workers most adversely affected by the work
disruptions due to Covid-19 are female workers with young children and without a college
degree. For this group, a one standard deviation increase in our Covid-19 work exposure
measure is associated with about a 15 percentage point increase in non-employment—which
is three times higher than the average response among all workers.
In sum, we find that our Covid-19 work exposure measure is an economically significant
predictor of employment. Workers in occupations that cannot work remotely are significantly
more likely to have lost their employment status in April 2020. This correlation is evident
both in cross-industry comparisons and in individual worker regressions.
2.2 Revisions in Revenue Forecasts
We next examine the ability of our measure to predict revisions in analyst forecasts regarding
the economic performance of these industries. Specifically, we use consensus forecasts of
firms’ revenues over various horizons from Capital IQ as of various points in time. We begin
17
This gap remains sizable at 5.5 percentage points when we absorb occupation fixed effects.
15
by comparing the change in revenue forecasts for the same accounting period (e.g., 2020Q2)
across two points in time, February 14 and May 15, 2020. Later, we illustrate how these
forecasts evolved on a weekly basis. We focus on forecasts horizons up to three years ahead,
that is, Q2 through Q4 of 2020, and full year forecasts for 2020–22. We use the Capital IQ
‘consensus forecast’; to minimize measurement error, we require that each firm has at least 3
individual forecasts. Due to this restriction as well as the fact that underlying forecasts are
less sparsely populated at longer horizons, we lose some industries for longer run forecasts,
especially 2022. For each firm f in the sample, we compute the percent change in revenue
forecast from February to May 2020,
Revenue forecast for Period τ , as of May 2020f
− 1. (6)
Revenue forecast forPeriod τ , as of February 2020f
For purposes of these regressions, we winsorize firm-level measures at the 1st and 99th
percentiles, then aggregate the non-missing firm-level revenue forecasts to the 4-digit NAICS
level, weighted by the corresponding level of the firms’ February 2020 forecasts for the same
period,
P
f ∈I Revenue forecast for Period τ , as of May 2020f
Forecast RevisionτI = P − 1. (7)
f ∈I Revenue forecast forPeriod τ , as of February 2020f
We re-estimate equation (10), but we now replace the dependent variable with the revision
in analyst forecasts in (7)
Forecast RevisionτI = aτ + βτ Covid-19 Work Exposurei + ετi . (8)
Our dependent variable is the revision in analysts’ revenue forecasts for period τ between
February and May 2020, aggregated to industry i. To conserve space, we focus our preferred
specification in which we restrict the sample to non-critical industries and weigh observations
by the number of Compustat employees in each industry; using equal weights or weighting
by the number of firms leads to quantitatively similar results. Standard errors are robust to
heteroskedasticity (White, 1980).
Columns (2) to (7) of Table 4 present our findings. Column (2) shows the effect of the
exposure measure on revenue forecasts revisions for Q2 2020. A one-standard deviation
increase in the exposure measure is associated with a 7.7 percent decline in analysts’ revenue
forecasts. Importantly, the effect is quite persistent, though the magnitudes do decline with
the forecast horizon τ , as we can see from comparing the magnitudes across columns. Figure 7
16
plots the estimated coefficients as a function of the forecast horizon τ . We see that financial
analysts expect the direct economic cost of the pandemic to subside significantly over the
course of 2020: a one-standard deviation increase in the exposure measure is associated with
a 3 percent decline in revenue for Q3, while the estimate for Q4 2020 is equal to -1.4 percent.
The overall estimate for 2020 is equal to 2.6 percent. Extending our analysis to 2021 and 2022,
we note that our estimates are significantly weaker at 2 and 1.5 percent, respectively—though
still statistically significant—suggesting that analysts expect the costs of these supply side
disruptions to persist for several years. Figures 4 and 6 provide scatter plots for non-critical
industries aggregated by major sector and across all 4-digit industries, respectively.
In Panel B, we also report (Compustat 2019 employment-weighted) averages of these
revenue forecasts for critical versus non-critical industries. We note that projected declines
are roughly three times larger for firms in non-critical industries relative to their counterparts
critical ones. In both cases, the largest expected declines in revenue are expected for 2020Q2
(-26% and -8% for non-critical and critical, respectively), though revenue forecasts suggest
that non-critical industries are expected to experience substantial long-run declines in output.
In particular, non-critical industries are projected to experience revenue declines of 10 percent
and 8 percent in 2021 and 2022, respectively.18
2.3 Default Probabilities
So far, we have seen that our Covid-19 work exposure measure predicts declines in both
employment declines as well as expected revenue across industries. We next examine another
forward-looking measure, expected probabilities of default.
We obtain estimates of default probabilities from the Risk Management Institute (RMI)
of the National University of Singapore. RMI generates forward-looking default probabilities
for issuers on a daily basis for maturities of 1, 3, 6, 12, and 24 months ahead using the
reduced form forward intensity model of Duan, Sun, and Wang (2012). These measures have
been shown to work well in forecasting applications and are available for a very wide array of
firms (over 70,000 publicly listed firms worldwide).19 We next compute the first difference
of the default probability over our the May–Feb 2020 period and aggregate these default
probabilities at the industry level (4-digit NAICS) by averaging across firms (weighted by
18
See, e.g., Landier and Thesmar (2020) and Barrero et al. (2020) for related evidence on projected
aggregate declines.
19
For example, Gallagher, Schmidt, Timmermann, and Wermers (2020) use these probability estimates to
study changes in portfolio risk of international portfolios of securities owned by US money market funds, and
illustrate that the default probabilities closely track CDS spreads. For additional details on the database, see
https://rmicri.org/en/.
17
employment).
We estimate the following specification,
Default ProbabilityτI = aτ + βτ Covid-19 Work ExposureI + ετi . (9)
The outcome variable is the industry-level probability of default over the next τ months. As
before, we restrict the sample to non-critical industries and weigh observations by the number
of employees in each industry.
Table 5 and Figure 8 present the results. Examining Table 5, we see that differences in our
Covid-19 work exposure measure are associated with increased probabilities of default. This
correlation is both statistically and economically significant and is monotonically increasing
in the forecast horizon τ . For instance, a one-standard deviation increase in our measure is
associated with a 0.22 percentage point increase in the probability of default over the next 6
months and a 0.44 percentage point increase over the next 2 years. Given that the average
probability of default over these horizons is 0.26 percent and 1.38 percent, respectively, these
magnitudes are quite substantial. Consistent with earlier evidence, we also observe that these
increases in default risk is considerably higher for non-critical industries relative to critical
industries.
2.4 Stock Market Performance
So far, we have seen that differences in our Covid-19 work exposure measure are correlated
with differences in analyst forecasts of revenue growth and default probabilities. Though
forward looking, both of these measures are somewhat subjective and may not necessarily
reflect the ‘average’ beliefs in the economy. Our next outcome variable partly addresses this
concern, as it focuses on differences in stock market valuations across industries.
We use data on stock returns from the Compustat Daily update database to holding
period returns over various horizons. We restrict attention to common stocks of firms that
have headquarters in the United States; are traded on NYSE, AMEX, or NASDAQ; and
eliminate financial and real estate firms (NAICS codes 52 and 53).20 Almost all of the
industries performed negatively during this period—the cross-sectional average decline is
equal to approximately 26 percent. Yet, there is considerable cross-industry dispersion—
20
We also exclude firms in the education sectors (6111–6114), because the remote exposures of employees
of the listed firms employed in these sectors are likely to be quite different from those employed in the
non-profit/government enterprises which comprise the vast majority of employment in these sectors. Likewise,
we exclude fast food restaurants (NAICS 722513) from the sample, as their exposures are likely to be quite
different from other firms in the restaurant industry.
18
the cross-sectional standard deviation is approximately equal to 18 percent. A select few
industries experienced price appreciation: firms in the grocery store industry (NAICS 4451)
saw a 21 percent increase in stock market values; firms in electronic shopping (NAICS 4541)
appreciated by approximately 13 percent. Others such as airlines (NAICS 4811) experienced
stock market declines of more than 65 percent.
Figure 9 illustrates our main results graphically via a scatter plot of our exposure measure
versus the stock market performance of these industries. Essential industries are marked in
red, while the trend line is estimated in the sample of non-essential industries. We note some
significant outliers that do not fit the line. For example, NAICS code 5414 corresponds to
interior designers; these workers can work remotely in normal times, but likely face significant
impairments to implementing these designs in the current environment.
We relate differences in our work exposure measure to differences in cumulative stock
market performance using the following specification,
Stock Returnsi = a + β Covid-19 Work Exposurei + εi (10)
The dependent variable corresponds to cumulative stock returns from February 14th through
May 15, 2020. As before, we restrict the sample to non-critical industries. To ensure that our
results are not driven by small industries with high idiosyncratic volatility, we estimate (10)
using weighted least squares—where we weigh observations by the total number of employees
(across firms in our Compustat sample) in each industry. Weighting by the number of firms
in each industry leads to quantitatively similar findings. Standard errors are robust to
heteroskedasticity (White, 1980).
Column (1) of Table 4 present our estimates. Examining the table, we see that our
industry exposure measure is economically and statistically significant in accounting for
differences in the economic performance across industries. The magnitudes are considerable:
a one-standard deviation increase in the lockdown measure is associated with a 6.8 percent
decline in stock prices during this period. Given that the cross-sectional standard deviation
in industries’ stock market performance during this period is approximately 16.6 percent, our
measure captures a significant share of the overall dispersion. Figures 4 and 9 provide scatter
plots for non-critical industries aggregated by major sector and across all 4-digit industries,
respectively. In Panel B, we also note that stock market declines are considerably larger
(26%) in non-critical industries relative to critical ones (8%).
In sum, we find that a measure of disruption exposure constructed with pre-2019 data
is a statistically significant predictor of cross-industry heterogeneity in returns during the
19
Covid-19 pandemic. That said, however, these estimates are somewhat at odds with the
impact on stock returns we discuss above. Specifically, as we compare Columns (1) through
(8) of Table 4, we see that a one-standard deviation increase is associated with 6.7 percent
decline in stock prices, whereas the expected decline in revenues over 2020 and beyond is
much smaller and short-lived. In reconciling these differences, a few possibilities emerge.
First, firms profits may be substantially more affected than revenues; second, the differential
stock return performance could also driven by differential increases in systematic risk during
this period. After all, the January to May period of 2020 is a period of increased uncertainty
(see e.g., Baker, Bloom, Davis, and Terry, 2020).
3 Real-time estimates of the supply-side disruption
In our analysis so far, we have developed a measure of industry exposure to production
disruptions due to Covid and shown that heterogeneity in risk exposures is associated with
cross-industry differences in economic performance during the pandemic period. Our analysis
was essentially cross-sectional in nature as we focused on cumulative outcomes during the
entirety of the pandemic period. However, our measure can also be used to construct a
real-time indicator of news associated with these production disruptions. To do so, we exploit
the high-frequency information in data from the financial markets.
In particular, we re-estimate equation (10) for the subsample of non-critical industries,
while allowing the slope coefficient to vary over time,
Stock Returnsi,t = a + βt Covid-19 Work Exposurei + εi,t . (11)
The realized time-series of the slope estimates βt have a portfolio return interpretation.
That is, they are a noisy (to the extent that our firm-level exposures are mis-measured)
estimate of the ‘Covid-19 return factor’, that is, the source of common variation in returns
that is related to the production disruptions as a result of the pandemic. (see, e.g. Fama
and MacBeth, 1973; Cochrane, 2009, for more details). Intuitively, this factor is a long-short
portfolio of industries based on their disruption exposure: it loads positively on industries in
which a lower fraction of workers report that they can telecommute, while loading negatively
on industries in which a higher fraction of workers report that they can work remotely.
Specifically, the return on the Covid-19 factor can be expressed as
N n
X 1 X empi,0 X
β̂t = wi (Ri,t − R̄temp ), where R̄temp ≡ Ri,t ≡ ei Ri,t , (12)
i
N i=1 emp0 i=1
20
where e; is the employment share of industry 7. The portfolio weight industry 7 receives in
the Covid-19 portfolio can be written as
N —
€; IX; — vie eiX)| e; LX; _ Xemp]
Ow N 2 SAN empl 2? (13)
j=l ej Es — kat enX;| doj=1 ej LX; —xX P|
where X; is our (predetermined) measure of exposure for firm 7, so the portfolio holds long
positions in stocks with exposure above the employment-weighted average exposure and short
positions otherwise.
Figure 11 summarizes how these portfolio weights vary across different broad sectors
(2-digit NAICS codes). The portfolio overweights industries whose workers can perform their
tasks from home, and underweight sectors where workers cannot work remotely. Consistent
with the discussion so far, we see that firms in retail trade, and accomodation and food
services receive large positive weights; by contrast firms in professional services and information
technologies receive negative weights. Appendix Table A.2 contains the full list of weights on
NAICS 4-digit industries.
Figure 10 plots these factor realizations in blue, along with returns on the market portfolio
in red. First, we note that the correlation between our factor and the market portfolio in
the three months before the pandemic period is significantly negative (approximately -0.4),
which suggests that the industries most exposed to the covid pandemic are industries with
low systematic risk during normal times. However, given the dominance of covid-related
news during this period, the correlation with the market during the pandemic period is
positive, though modest (it ranges from 0.1 to 0.3 depending on how the factor is constructed).
Further, we see in Figure 10 that extreme daily realizations of our factor are associated with
significant news about the pandemic.
In brief, we see that realizations of our ‘Covid-19 factor’ largely tracks the unfolding of
news related to the pandemic. As of May 15 2020, this long-short portfolio had lost roughly
50 percent of its value since the beginning of the year—compared to 10 percent for the broad
market index. This pattern supports the view that Covid-19 is primarily a reallocation shock
as not all sectors were symmetrically affected—which also echoes the views in Barrero et al.
(2020). These results speak to how investors could construct an investment strategy that
could significantly hedge future covid-19 related uncertainty. They could do so by taking a
short position in our Covid-19 factor, which would involve taking the opposite position to
the one described in Appendix Table A.2 .
Last, we also examine the real-time response of our two other forward-looking variables—
21
revisions in analyst forecasts and estimated default probabilities. Figures 12 and 13 summarize
our findings. Similar to our stock return evidence above, we see an increased divergence
in revenue forecasts and expected default probabilities across high- and low-Covid 19 work
exposure industries by the end of March and beginning of April 2020. Comparing the two
figures, we see that default probabilities appear to respond a bit earlier than analyst revenue
forecasts, and they seem to recover somewhat from their peaks by the end of April. That
said, these differences could simply be due to the fact that analyst forecasts are released with
some delay.
4 Conclusion
We characterize the supply-side disruptions associated with Covid-19 exploiting differences in
the ability of workers across industries to work remotely. Our results uncover economically
sizeable differences in economic outcomes for both workers and firms depending on the
ability to work remotely. Sectors in which a higher fraction of the workforce is not able to
work remotely experienced significantly greater declines in employment; significantly more
reductions in expected revenue growth; worse stock market performance; and higher expected
likelihood of default. Lower-paid workers, especially female workers with young children,
were significantly more affected by these disruptions.
References
Alon, T. M., M. Doepke, J. Olmstead-Rumsey, and M. Tertilt (2020, April). The impact of
covid-19 on gender equality. Working Paper 26947, National Bureau of Economic Research.
Baker, S. R., N. Bloom, S. J. Davis, and S. J. Terry (2020, April). Covid-induced economic
uncertainty. Working Paper 26983, National Bureau of Economic Research.
Baker, S. R., R. A. Farrokhnia, S. Meyer, M. Pagel, and C. Yannelis (2020, May). Income,
liquidity, and the consumption response to the 2020 economic stimulus payments. Working
Paper 27097, National Bureau of Economic Research.
Barrero, J. M., N. Bloom, and S. J. Davis (2020, May). Covid-19 is also a reallocation shock.
Working Paper 27137, National Bureau of Economic Research.
Bartik, A. W., M. Bertrand, Z. B. Cullen, E. L. Glaeser, M. Luca, and C. T. Stanton (2020,
April). How are small businesses adjusting to covid-19? early evidence from a survey.
Working Paper 26989, National Bureau of Economic Research.
22
Bonadio, B., Z. Huo, A. A. Levchenko, and N. Pandalai-Nayar (2020, May). Global supply
chains in the pandemic. Working Paper 27224, National Bureau of Economic Research.
Borjas, G. J. and H. Cassidy (2020, May). The adverse effect of the covid-19 labor market
shock on immigrant employment. Working Paper 27243, National Bureau of Economic
Research.
Cajner, T., L. D. Crane, R. A. Decker, J. Grigsby, A. Hamins-Puertolas, E. Hurst, C. Kurz,
and A. Yildirmaz (2020, May). The u.s. labor market during the beginning of the pandemic
recession. Working Paper 27159, National Bureau of Economic Research.
Campello, M., G. Kankanhalli, and P. Muthukrishnan (2020, May). Corporate hiring under
covid-19: Labor market concentration, downskilling, and income inequality. Working Paper
27208, National Bureau of Economic Research.
Cochrane, J. (2009). Asset Pricing: Revised Edition. Princeton University Press.
Coibion, O., Y. Gorodnichenko, and M. Weber (2020a, May). The cost of the covid-19 crisis:
Lockdowns, macroeconomic expectations, and consumer spending. Working Paper 27141,
National Bureau of Economic Research.
Coibion, O., Y. Gorodnichenko, and M. Weber (2020b, April). Labor markets during the
covid-19 crisis: A preliminary view. Working Paper 27017, National Bureau of Economic
Research.
Corrado, C., C. Hulten, and D. Sichel (2009). Intangible capital and u.s. economic growth.
Review of Income and Wealth 55 (3), 661–685.
Ding, W., R. Levine, C. Lin, and W. Xie (2020, April). Corporate immunity to the covid-19
pandemic. Working Paper 27055, National Bureau of Economic Research.
Dingel, J. I. and B. Neiman (2020, April). How many jobs can be done at home? Working
Paper 26948, National Bureau of Economic Research.
Duan, J.-C., J. Sun, and T. Wang (2012). Multiperiod corporate default predictiona forward
intensity approach. Journal of Econometrics 170 (1), 191 – 209.
Eisfeldt, A. L. and D. Papanikolaou (2013). Organization capital and the cross-section of
expected returns. The Journal of Finance 68 (4), 1365–1406.
Fairlie, R. W., K. Couch, and H. Xu (2020, May). The impacts of covid-19 on minority
unemployment: First evidence from april 2020 cps microdata. Working Paper 27246,
National Bureau of Economic Research.
23
Fama, E. F. and J. D. MacBeth (1973). Risk, return, and equilibrium: Empirical tests.
Journal of Political Economy 81 (3), 607–636.
Farboodi, M., G. Jarosch, and R. Shimer (2020). Internal and external effects of social
distancing in a pandemic. Technical report, National Bureau of Economic Research.
Gallagher, E. A., L. D. Schmidt, A. Timmermann, and R. Wermers (2020). Investor
information acquisition and money market fund risk rebalancing during the 2011–2012
eurozone crisis. The Review of Financial Studies 33 (4), 1445–1483.
Hassan, T. A., S. Hollander, L. van Lent, and A. Tahoun (2020, April). Firm-level exposure
to epidemic diseases: Covid-19, sars, and h1n1. Working Paper 26971, National Bureau of
Economic Research.
Landier, A. and D. Thesmar (2020, May). Earnings expectations in the covid crisis. Working
Paper 27160, National Bureau of Economic Research.
Mateyka, P. J., M. A. Rapino, and L. C. Landivar (2012, October). Home-based workers in
the united states: 2010. Working paper, United States Census Bureau.
Mongey, S., L. Pilossoph, and A. Weinberg (2020). Which workers bear the burden of social
distancing policies? Technical report, National Bureau of Economic Research.
White, H. (1980). A heteroskedasticity-consistent covariance matrix estimator and a direct
test for heteroskedasticity. Econometrica 48 (4), 817–838.
24
Figures and Tables
Figure 1: Distribution of Covid-19 Work Exposure Measure across Industries
Non−Critical Industries Critical Industries
4
3
Density 2
1
0
.2 .4 .6 .8 1 .2 .4 .6 .8 1
Covid−19 Work Exposure
The figure plots the distribution of the Covid-19 work exposure measure across critical (right
panel) and non-critical (left panel) industries.
25
Figure 2: Changes in Worker Foot Traffic: Critical and Non-Critical Industries
A. Weekly Worker Foot Traffic for B. Differences in Weekly Worker Foot Traffic
Critical and Non-Critical Industries (Critical vs Non-Critical)
20
50
Difference in average scaled worker foot traffic (%)
Estimate
95% CI
Average scaled worker foot traffic (%)
0
40
−20 30
20
−40
10
−60
Non−Critical
Critical 0
−80
Jan06 Jan13 Jan20 Jan27 Feb03 Feb10 Feb17 Feb24 Mar02 Mar09 Mar16 Mar23 Mar30 Apr06 Apr13 Apr20 Apr27 May04 May11 May18 Jan06 Jan13 Jan20 Jan27 Feb03 Feb10 Feb17 Feb24 Mar02 Mar09 Mar16 Mar23 Mar30 Apr06 Apr13 Apr20 Apr27 May04 May11 May18
Date Date
C. Slope on COVID-19 Work Exposure: D. Slope on COVID-19 Work Exposure:
Critical Industry Subsample Non-Critical Industry Subsample
200
Estimate
250
95% CI
Slope on Covid−19 remote exposure (%)
Estimate
95% CI
Slope on Covid−19 remote exposure (%)
200 100
150
0
100
−100
50
0
−200 Jan06
Jan13
Jan20
Jan27
Feb03
Jan06 Jan13 Jan20 Jan27 Feb03 Feb10 Feb17 Feb24 Mar02 Mar09 Mar16 Mar23 Mar30 Apr06 Apr13 Apr20 Apr27 May04 May11 May18
Feb10
Feb17
Feb24
Mar02
Mar09
Mar16
Mar23
Mar30
Apr06
Apr13
Apr20
Apr27
May04
May11
May18
Date Date
The figure documents the differential decline in average worker foot traffic between critical
and non-critical industries—given our classificationin Appendix Table A.1. Panel A plots the
Total Worker Foot Traffic in week t in Industry I
average change in weekly worker foot traffic—log Pre-period Average Worker Foot Traffic of Industry I
.
Panel B plots the estimated βt coefficients from equation (2) in the main document, which
capture the differences in weekly foot traffic (relative to pre-period averages) between critical
and non-critical industries. Panels C and D report δt coefficients from cross-sectional
regressions of industry-level changes in worker foot traffic on our Covid-19 Work Exposure
(equation (3)) measure for critical and non-critical industries, respectively. All regressions are
weighted by total NAICS4 employment, and pre-period averages are computed from January
6-February 16 of 2020. Dates on the horizontal axis correspond with the end of each calendar
week.
26
Figure 3: Employment Growth and Covid-19 work exposure
0 524 519 521 492
517 334
522 445486
Change in employment (%), April−−March 2020
518 511 322 622 533 452
562
325 211 623
515 444 447
425 531 423 335311
454443 484
493
424
324 333 321 237332
813
446 326
488
331 483
213
236
339 327 238
621 624
337 811
561 441
313
323314
−20 532
336
712
481
453
451
485
−40 315 721
711
442
722
512
812
448
−60 713
.2 .4 .6 .8 1
COVID−19 Work Exposure
20
Change in employment (%), Apr’19−−Apr’20
519
492
211
518 524 334 562
0 511 522
521 444 622 533 445486
452
517 325 531 322 623
493
423 311 447
425 515 335 237
484
424 321326
443
813 324 454446
333 488332
236 483
621 624 238
327331
339 811
337 441
561
−20 323 532
213 313
314
336 712
481
453
451 485
−40 721
315
512 711
442 722
812
713
−60 448
.2 .4 .6 .8 1
COVID−19 Work Exposure
Figure plots the correlation between employment growth and Covid-19 work exposure at
the 3-digit NAICS level. Top panel plots month-to-month changes (April vs March 2020)
while the bottom plots year-on-year changes (April 2020 to April 2019). The points in red
correspond to the critical industries listed in Appendix Table A.1.
27
Figure 4: Real activity of non-critical industries and Covid-19 work exposure by major sector
A. BLS employment growth (yoy %) B. Stock returns (%)
Mining −10
IT
0
Change in employment (%), Apr’19−−Apr’20
Wholesale
IT Real Estate −20
Construction
Manufacturing Bus Services Manufacturing
Health
Transportation
Support Services Support Services
−20 Wholesale
Stock returns
Other Services −30
Health
Hotels
Other Services
Construction
Retail
Retail −40 Mining Education
−40
Entertainment
Hotels −50
Entertainment
−60 −60 Transportation
.6 .7 .8 .9 1 .5 .6 .7 .8 .9 1
Covid−19 Work Exposure Covid−19 Work Exposure
C. 2020 Q2 revenue revisions (%) D. 2020 Q3 revenue revisions (%)
0 Bus Services Retail
Construction
−10
Construction IT
Bus Services Manufacturing
Wholesale
IT
2020 Qtr 2 revenue revisions 2020 Qtr 3 revenue revisions
−20
Wholesale Other Services
Support
Manufacturing Services Support Services
Other Health
Services
Retail −20
Health
−40 Mining
Hotels −30 Hotels
−60
Transportation
Mining Entertainment
−40
−80 Entertainment Transportation
.5 .6 .7 .8 .9 1 .5 .6 .7 .8 .9 1
Covid−19 Work Exposure Covid−19 Work Exposure
E. 2020 Q4 revenue revisions (%) F. 2022 revenue revisions (%)
0 Bus Services
−5 IT
Retail
Manufacturing Construction
Bus Services Wholesale Retail
Construction
IT Other Services
2020 Qtr 4 revenue revisions
−10 Hotels
Manufacturing
Wholesale Health
2022 revenue revisions
−10
Other Services
Support Services Entertainment
Support Services
−15
Hotels
−20 Entertainment
−20
Transportation Transportation
−30 −25
Mining Mining
.5 .6 .7 .8 .9 1 .5 .6 .7 .8 .9 1
Covid−19 Work Exposure Covid−19 Work Exposure
Panel A plots employment growth rates from April 2020 vs 2019 from the BLS against
exposure (aggregated across 3-digit industries using BLS total employment weights). Panels
B-F plot stock returns (panel A) and revisions in revenue forecasts (panels B-E) for non-critical
industries from mid-February to mid-May versus the Covid-19 work exposure measure. For
purposes of generating the graph, we average across 4-digit non-critical industries, weighting
by Compustat employment. 28
Figure 5: Employment and Covid-19 work exposure: Worker Heterogeneity
A. Income B. Age
15 10
Coefficient on COVID−19 Work Exposure Coefficient on COVID−19 Work Exposure
8
10
6
5
4
2
0
Below 25% 25% to 50% 50% to 75% Above 75% Under 30 30−39 40−49 50 and over
Income group Age group
Estimate 95% CI Estimate 95% CI
C. Gender and Education D. Gender, Education, and Family Status
15 15
Coefficient on COVID−19 Work Exposure Coefficient on COVID−19 Work Exposure
10 10
5 5
0 0
Women Men Women Men
College No College 95% CI Has Kids Under 14 No Kids Under 14 95% CI
E. Income and Gender F. Gender, Education, and Family Status
15 Women Men
20 20
Coefficient on COVID−19 Work Exposure
Coefficient on COVID−19 Work Exposure
15 15
10
10 10
5 5 5
0 0
0
College No College College No College
Below 25% 25% to 50% 50% to 75% Above 75%
Women Men 95% CI Has Kids Under 14 No Kids Under 14 95% CI
Figures plot how the coefficients of a regression of non-employment status on our Covid-19
work exposure measure vary with workers characteristics.
29
Figure 6: Revenue forecast revisions and Covid-19 work exposure
2020 Q2 Revenue forecast revision (%)
4442
4812
4541 2379
3111 3114 4539
0 5112 5173 3112
6214 4451
4523
2211 3222 4461
2213 6231
4242 3119 4441
5611 3113
5415 5412 5413 3118 49216216
5612 3324 3221
3254 4249 3344 3332 2362
6222
4842 3116
3261
5411 5152 3256 3366 3274
5616
Revenue forecast revision (%)
3342 3341 42395614 5617 4236
33232361 4532
4831
4413
5174 4247 62238121
2371
4885 2212
3132
4869
3325
5182 5191 31213345 4412
4431 2383
3312 4861
4862
48415621
2382 3273
3343 5179 8122 3259 3359
3351 3322
3272
8123
2123
5419 5416
4246
5417 4234 4832
4237
3219 4483
33134511 3321
33644233 5111
3333 3251 3399 6221
3255 4238 3334 231
3371 5613 3311
3353 3329
5418 3252
6215 3339 6211 33313211 5622
3279
5151
3365 3335 2121
33914251
3336 6219 3231
3352
4421 3212
4482
4244 4522
4471
2131 33693141
4481
3362 3326
2111 3379 3262
4411
6213 4422 7225
31528129 7223
4543
−50 5619
3363 3169 4853
3241 3361
7139
6244
7211
2372
5615
5121
7132
7131
4811
7113
−100
.2 .4 .6 .8 1
Covid−19 Work Exposure
2020 calendar year revenue forecast revision (%)
20
2379
Revenue forecast revision (%)
4541
3111 3114 6214 4442
0 5112 3254 51734242 5411 4461
4812
44314421
3222 4422
4441 44824511 4451
4481
2373 4539
45224523
4483
3274
3119
3112
3118
5612
2213 6222 33666231
3113
3116
5415 4249
3341 22114543 4921 3256
6216 2362
6223
4842
5616 3221
8123
4532
5413 3344 3332 33242371
56113323 3261
8121 4413
3342 5412 3121 4234 5111 42395614
3345 5152
3259 3152 56174236 3312 3325
2382
3351 4869
8122 4412 48415621
3322
4831
5182 5191 3343 5174 5417 4233 3335
6215 33713399 62214885 3329
3321
3132
3273
517932555416 4251 62193251 3359
4244
4231 33533219 3272
4237
42463391 3333 4238 3339 2383
3169
7132
3311 5151
4862
2212
4832
3334 3313 2361 4411
3364 5418 3231 5613
3352 3252 3262
3279
5622
33692123
3211
3365 5419 3336 4471
33312121
3362 3212 7225 3141
−20 4247 3379 3361 7223
4861
3363
56196211
3326
6213
7139
4853
2131 2111
8129 6244
2372 7112
5121
3241
−40 7211
5615
4811
7131
−60
7113
.2 .4 .6 .8 1
Covid−19 Work Exposure
The top panel of this figure plots the revisions in revenue forecasts for Q2 2020 between
mid-February and mid-May 2020 versus the Covid-19 work exposure measure. The bottom
panel plots the corresponding revisions for the year 2020. The points in red correspond to
the critical industries listed in Appendix Table A.1.
30
Figure 7: Revenue forecast revisions relative to February 2020 forecast as a function of Covid-19 work exposure
0
% Change in Revenue Forecast relative to January 2020 forecast
−1
−2
−3
−4
−5
−6
−7
31 −8
−9
−10
−11
−12
2020Q2 2020Q3 2020Q4 2021 2022
The Figure plots the estimated coefficients from equation (8) in the main document. Specifically, we plot the relation between
revisions in industry-level revenue forecasts (from mid-February to mid-May 2020) as a function of their Covid-19 work exposure in
the sub-sample of non-critical industries. The plotted coefficients correspond to columns (2)-(4) and (6)-(7) of Panel B of Table 4.
Coefficients are standardized to a unit standard deviation change in the Covid-19 work exposure measure. Shaded areas correspond
to 95% confidence intervals.
Figure 8: Default Probabilities and Covid-19 work exposure
3m 6m
6 7139
Change in default probability (%), Feb−−May 2020 Change in default probability (%), Feb−−May 2020
6
7139
4
4
2111 4512
3222 4461 4481
2111 3231
4481 5615
2 3222 4461 6233 4522
4512 2 5414
2123
3231 5111 4422 2379
56193323
6233 5615
2123 48836231 4811
2121
5414 4249 5111 4422 3323 4522 44824511
5619 2379 2121
4511 5179
4482 48836231 4811 2131 3313
8121 4831
5151
5179
4233
2131 33138121 5151
4831 4249 4233 3333 8129 2373
7132 3273
3365 54153254 3364
3341 3333
5418 4247614 31528129
3399 6221
4236 373 3261
2362 33132
169 3262
4237
7225
44713273
5121
7224
8121
7131 3365 54153254 3364 5418 4247
5614 31523256
3399
3361
6221
3169
211
4236
2362 3262
4237
3261
72257224
7131
8121
5121
0 5182 5112
3364 3342
3343 5174
5419 4246
5412
3255
5417 5413
5416 4241
51735411
3364
4251336
4234
45 3241 5111
6219
8122
4239 4543
4238
3251
4239
3344
3335
3259
5152
211 3379
3222
3352
4412
3371
4421
3372
3339
6116 3361
3399 5613
3256
5612
5611
3332
3334
4239 6213 42443312 3363
3362
4231
5617 211
3219 4841
399
3252
2361
2371
5616
721133113321
6222
2372
359
4881
3212
2382
4889
24
351
353
2383
4243
4461 3329
3325
3327 3141
39
4411
369
7223
81117113
3222
3272
331
6244711164532
3322
3219
3262
3221
3326
3279
4832
3274
4862
3132
3329
4869
3211
112
8123
4483
114 4246
5416 3341
5413 336
5411 3241 6219
5111 4543
4238 3379
3222 3362
3399 3219
2361
2371
3399
3252 3141
333139
5616
4471
3329
3325
7113
4411
369
3364 3342
5173 42423391
3121 3112
6215
119
3111
3118 6214
4921
4441
4541 2213
6216 6223
4885
31154842 5622
4859 3116
56214853
4523
4245
4413
3113 5412 5173 4234 3352 7211
5613 3311
4841
7223
33313322
3222
3114 4 812 4451
4523
4539
0 5182 5112 3343 54195174
5173 4242
5417
3255 3391 3345
3364
4251 8122
4239
4239 211
3335
5152
3112
6215
3119 4239
3251
3344
3259
6213 371
372
4421
3339
4412
6116 4231
5612
5611
3332
33345617
4921
62143363
2213
4441 4461
4243
6216 6222
2372
4881
3212
359
6223
4889
2382
24
351
2383
353
48853327
3312
4859
4842 2212
3219
3326
63221
4862
3279
4853
4832
4532
3262
3321
3274
3132
8111
62447111
4245
3329
4869
3211
112
8123
4483
4523
5622
3116
4413
5621
4244 4539114
3121 3118
3111 4541 3115 3113
4451
4523
5191 4861 4241 3114 4812
5191 3272
4431 4861
4431
−2 −2
.2 .4 .6 .8 1 .2 .4 .6 .8 1
Covid−19 Work Exposure Covid−19 Work Exposure
12m 24m
10 7139
Change in default probability (%), Feb−−May 2020 Change in default probability (%), Feb−−May 2020
4512
4522
5414 4811
5 4461
2111 6233 5615
5111 4883
7139 3222 4481
3231 4831
2379 2123
5619
4512
5 2131
4422 4482
332373 6231 3273
3313
2111
4461
3231 5615
4522 8129 7132 5151
7131
7224
3222 3364 32223361 31694237
5414 6233 4481 4233 4247
3333 56143152 3399 6221 72257113
4238 3379 3219
62115616
7211
4881 7223
33693141
5111 4811
2379 2123 5415 5416
5412 5413 5418
4234
5411 3241 4543
3352
3372 42362371
4889
6223
2361 4411
3325
3329
8121
333139
3326
3219
2121
5619 4883
4422 4242 3341 5111
6219
42396116 5613
3362
371
3399
33394231
3256 3252
3212
2362
6222
44612372
2382
3399 3262
3222
3279
3331
4832
3221
4853
3322
12
3274
3323 5182
5179
5419
5173 5417
4246
32553391 3364
3345 8122 3335
3112
211
6215 3344
6213
3259
5152 4421
4921
3334
5617
33322213
5611
6214
5612
4441 3242383
3351
3353
3359
4243
3312
3366
4859
6244
5622
3132
4869
3116
111
4482 6231 0 3364 3254
5112 3342 5174 4251
3121
5173 4239
3119
3118
3111 6216 8121
4842
3327
4885 3211
4245
8123
4413
5621112
4483
3329
4523
3113
3262
8111
4831 3365 3343 3336 4239 4541
3311 3329
5121
4471
4523
4862
3261
3313 2373 2121 5191 3251 311548414451
5179
2131
8129 6221 7132 5151
3273 4412 4244
3114 4511 33217114
4233
33645411 333362194247
5614 3152 3399 8121
3169
4511
6211 4237
7224
7131
7225
3262
3365 54153254 5416
4246 3341
5413 5418
3241 3222
4238
4543 33793361
3362
3399 4236
32192362
5616 7113
3325
333139
2371 8121
3141
3329
4411
3369
7223 3363
5182 5174 5412
4242 5173 4234 5111 3352
4239 3372
3371
33393256
4421 7211
2361
3252
5613
3399
4881
6222
6223
44612372
3212
4889 3261
3222
5121
3322
3219
3331
3326
3221
3279
3329
4471
2212
4853 4539
4532
0 3364 3342
5112 3343 54195173 5417
3255 3391 3336
3364 4239
3345
4251
3121 8122
4239
6215
3119
3118 3335
2211
3112
3111 3344
3259
5152
3251 6116
6213
3114 3332
3334
6214
4412
4541 4231
4921
5612
5611
5617
2213
4441
6216 3359
3242382
3351
2383
3311
3312
3353
4243
3115 66
62447111
4859
3327
4842
4841
4885 4832
3274
3132
5622
4869
8111
4245
3262
3211
3329
8123
112
4483
4862
4523
3116
4413
5621
3113
3321
4451
4523
114
4532 4812
5191 4244 4539
3363
4812
4241
4249 4861
4241 4861
3272 4249
4431
−5
3272
−5 4431
.2 .4 .6 .8 1 .2 .4 .6 .8 1
Covid−19 Work Exposure Covid−19 Work Exposure
Panels of this figure plot changes in 3, 6, 12, and 24 month default probabilities for Q2 2020
between mid-February and mid-May 2020, respectively, versus the Covid-19 work exposure
measure. The points in red correspond to the critical industries listed in Appendix Table A.1.
32
Figure 9: Stock Price change and Covid-19 work exposure
7114
4451
20
6219 4812
Stock Price Performance, Feb−−May 2020
4541 4539
3114
3254 31154842 4523
3111 6214 4413
0 5112 4249
3262
4441 6211 4841 4483
5191 3342 3119
3112
6215
4239
5182 5179 3345 5614 6216 4885 4859
3344 3256 4471
4242 4431 5617 3212 4237 4532
3221
3343 3255
517354173391 3336 4412 3361
3118 3222 5611 4442 3353 32613116
5415 3121 5152
3251 3352
3259 4461
2213 3113
−20 3341 4921
3332 33246222 7225
4251 2211 33624243 3327
5612
3339
5621
3322
4853
3311 8123
3329
5419 5412 4234 3333
3241
5111 3335 3334
4231
3363
3399 23833331
42366223
3325
5151
4861
3211
3365 5416 42415411 8122 4238 3371
2382
3359
2362
6231
4411
33694245
3366
5616
2361 3321
7111
5613 4869
4862
4247 3252 3219
3312
3351
4511 8121
5413 4421 62214481 3132
3279
5622
5174 4233 71397112
5121
4832
5418 3323 4881 8111
2123
−40 2111 6116 4244 2371 3274
3152 7211
3364 3379 62447223
4246 3231 4482 4889 3141
6213 7113
5615
4543 3372 6233 3326
3169
7132 3273
2373
44223313 2379 2121
2131 5619 3272
7131
2372
−60 8129 45227224
4512
4811
4831
4883
5414
−80
.2 .4 .6 .8 1
Covid−19 Work Exposure
The figure plots the share price performance of different industries between mid-February to
mid-May 2020 versus the Covid-19 work exposure measure. The points in red correspond to
the critical industries listed in Appendix Table A.1.
33
Figure 10: COVID-19 factor vs market
1.2 Violent protests
CDC expects community complicate
spread of coronavirus, reopening plans
warns of severe disruptions Murray model predicts
1.1 earlier peaks in US
Positive initial
test results for
WHO Declares Moderna vaccine
1 COVID-19 Negotiations
a Pandemic stall over new
aid package States prepare
to ease lockdown
0.9
Fed announces
unlimited Positive
First confirmed
bond purchases vaccine
case outside China Covid death toll
0.8 jumps in China
Cases news,
spread outside US jobless lockdown
China claims higher loosen
than expected
0.7
States set up
curfew guidelines
34 0.6
NYC Businesses
To keep half of
workforce at home
0.5
Murray model predicts
Study finds death higher death toll for US
rate in Wuhan lower
0.4 than initial estimates
0.3
Jan Feb Mar Apr May Jun Jul
2020
COVID19 Factor Market Portfolio
The figure shows the time-series of the estimated slope coefficients in equation (11) in blue. As noted by Fama and MacBeth (1973)
the realizations of these slope coefficients have a portfolio interpretation, here labeled as the ‘Covid-19 return factor’. These implied
portfolio returns are accumulated since beginning of the year. The red line represents cumulative returns to the market portfolio
during this period. Events are dated as if they occurred at the beginning of the trading day on the date of their announcement.
Figure 11: COVID-19 factor: portfolio composition (NAICS 2-digit Industry)
44−45: Retail Trade
72: Accommodation and Food Services
56: Administrative and Support and Waste Management and Remediation Services
48−49: Transportation and Warehousing
23: Construction
71: Arts, Entertainment, and Recreation
81: Other Services (except Public Administration)
62: Health Care and Social Assistance
31−33: Manufacturing
42: Wholesale Trade
61: Educational Services
21: Mining, Quarrying, and Oil and Gas Extraction
54: Professional, Scientific, and Technical Services
51: Information
−100 −50 0 50
Fama−Macbeth portfolio weight (%)
The figure the loadings of our ‘Covid-19 return factor’ on broad industry sectors (defined at
the 2-digit NAICS level). See Appendix Table A.2 for portfolio weights at a more detailed
industry classification.
35
Figure 12: Revenue forecast revisions and Covid-19 work exposure over time
A. 2020 calendar quarter forecast revisions (%)
0
Coefficient on Covid Exposure
−20
−40
2020Q2
2020Q3
−60 2020Q4
Feb21 Feb28 Mar06 Mar13 Mar20 Mar27 Apr03 Apr10 Apr17 Apr24 May01 May08 May15
Consensus forecast date
B. Longer horizon forecast revisions (%)
0
Coefficient on Covid Exposure
−10
−20
2020
2021
2022
−30
Feb21 Feb28 Mar06 Mar13 Mar20 Mar27 Apr03 Apr10 Apr17 Apr24 May01 May08 May15
Consensus forecast date
Figure plots the Fama-McBeth slope coefficients of a regression of forecast revisions on the
(non-critical) industry’s Covid-19 work exposure at a weekly frequency. The top panel of
this figure plots the slope coefficient for revisions in revenue forecasts for Q2–Q4 2020. The
bottom panel plots the corresponding revisions for the years 2020 through 2020.
36
Figure 13: Default Probabilities and Covid-19 work exposure over time
4
Coefficient on Covid−19 work exposure (%)
3
2
1
0
01jan2020 01feb2020 01mar2020 01apr2020 01may2020
12M 3M 6M 2Y
Figure plots the Fama-McBeth slope coefficients of a regression of default probabilities on
the (non-critical) industry’s Covid-19 work exposure at a daily frequency. The dependent
variable is expressed in percentage points so that the slope coefficient can be interpreted as
the increase, in basis points, of the probability of default associated with a 1 percentage point
increase in exposure. Regressions are weighted by employment. Different colors correspond
to different forecast horizons.
37
Table 1: Firm summary statistics by quartiles of Covid-19 work exposure
A. All industries B. Firms in non-critical industries
Firm characteristic (median) Covid work exposure bin: Covid work exposure bin:
Lowest 2 3 Highest Lowest 2 3 Highest
(1) (2) (3) (4) (5) (6) (7) (8)
Covid-19 work exposure 0.45 0.64 0.82 0.98 0.32 0.67 0.86 0.97
Sales ($ M) 41 471 1308 1479 414 769 1323 1360
Employment (thousands) 0.18 1.14 2.67 4.79 1.5 1 4.3 4.87
Market capitalization ($ M) 774 933 2059 1594 1914 1129 1599 1403
Tobin’s Q 2.7 1.76 1.44 1.37 2.28 1.52 1.56 1.34
Book Debt to Book Assets (%) 40.2 52.7 57 61.1 53.3 52.9 55.5 62.3
R&D / Sales(%) 31.4 8.6 2.1 0.1 15.5 7.4 2.5 0.3
SG&A / Sales 60.9 32.5 19.7 15.5 49.4 23.5 20.6 16.3
PPE to Assets 12.2 25.2 52.9 62.1 16.4 32.5 36.1 58
Gross Profitability to Assets 11.9 29.7 23.8 23.3 34.7 24.1 29.6 23.1
Return on assets (ROA) -19 7.8 10.6 11.2 6.1 9.3 10.8 11.4
Table reports median characteristics of firms in the Compustat database, grouped into
quartiles based on our industry-based measure of Covid-19 work exposure. The left panel (A)
summarizes these characteristics for all firms, including those in critical industries. The right
panel (B) repeats the analysis for the subsample of firms which we classify as non-critical.
The list of critical industries at the 4-digit NAICS level is in Appendix Table A.1. Gross
profitability is sales (Compustat: sale) minus costs of goods sold (Compustat: cogs); Tobin’s
Q is computed as the sum of the book value of assets (Compustat: at) plus the market value
of equity (Compustat: prcc f times csho) minus the book value of equity (Compustat: ceq)
minus deferred taxes (Compustat: txdb). PPE is Property Plant and Equipment (Compustat:
ppegt).
38
Table 2: Covid-19 Work Exposure and Employment Growth
Weighted by Mar’20 employment Equal weighted
A. Month-on-month
growth (%) All Industries Non-Critical All Industries Non-Critical
(1) (2) (3) (4)
Covid-19 Work Exposure -61.9∗∗ -104.3∗∗∗ -34.7∗∗∗ -49.6∗∗∗
(-2.39) (-3.24) (-3.30) (-3.17)
Number of observations 73 50 73 50
R2 0.151 0.325 0.079 0.139
Weighted by Mar’19 employment Equal weighted
B. Year-on-year
growth (%) All Industries Non-Critical All Industries Non-Critical
(5) (6) (7) (8)
Covid-19 Work Exposure -64.2∗∗ -110.3∗∗∗ -36.0∗∗∗ -57.0∗∗∗
(-2.37) (-3.33) (-3.28) (-3.49)
Number of observations 73 50 73 50
R2 0.149 0.333 0.076 0.166
Month-on-month growth Year-on-year growth
C. Average, all
industries Weighted Unweighted Weighted Unweighted
(9) (10) (11) (12)
Critical Industries -6.02∗∗∗ -6.18∗∗∗ -4.91∗∗ -5.79∗∗∗
(-2.95) (-3.74) (-2.54) (-3.08)
Non-Critical Industries -22.75∗∗∗ -18.10∗∗∗ -22.60∗∗∗ -17.98∗∗∗
(-4.88) (-7.68) (-4.64) (-7.26)
Diff: Non-Critical-Critical -16.73∗∗∗ -11.92∗∗∗ -17.70∗∗∗ -12.19∗∗∗
(-3.29) (-4.14) (-3.38) (-3.92)
Number of observations 73 73 73 73
Panels A and B of table reports the coefficients estimates from equation (4) in the document
using employment-weighted and unweighted specifications (t-statistics in parantheses). Odd-
numbered columns repeat the analysis for the subsample of non-critical industries. The
dependent variable is expressed in percentage points and constructed using private sector
employment data from the BLS’ April 2020 employment report at the 3-digit NAICS level.
Standard errors are robust to heteroskedasticity (White, 1980). The list of critical industries
at the 4-digit NAICS level is in Appendix Table A.1. We classify a 3-digit NAICS industry
as critical if at least 50% of its employment is classified as critical based on our more
disaggregated measure. Panel C reports the average declines in the dependent variable for
critical and non-critical industries, respectively, as well as the difference between the two
averages. The cross-sectional standard deviation across BLS industries is 12.5%.
39
Table 3: Covid-19 Work Exposure and Worker Employment Outcomes
Probability (%) worker is not employed as of... April 2020 February 2020
(conditional on being employed as of March 2020) (1) (2) (3) (4)
∗∗∗ ∗∗∗ ∗∗∗
Critical industry -9.2 -9.2 -8.1 -0.2
(-4.87) (-5.32) (-5.01) (-0.63)
Non-critical industry × Covid-19 work exposure 50.6∗∗∗ 40.8∗∗∗ 36.0∗∗∗ -0.9
(6.28) (5.24) (4.77) (-0.59)
Critical industry × Covid-19 work exposure 28.4∗∗∗ 18.1∗∗ 10.8 -4.1
(3.22) (2.11) (1.43) (-1.56)
Controls
College Graduate Dummy -6.3∗∗∗ -3.8∗∗∗ -0.4
(-5.34) (-3.27) (-0.88)
Female Dummy 4.6∗∗∗ 3.2∗∗ 1.5∗∗∗
(3.14) (2.16) (2.83)
Kids under 14 -2.3∗ -1.7 -0.1
(-1.65) (-1.30) (-0.08)
Female Dummy × Kids under 14 4.0∗∗ 2.8 -0.4
(2.24) (1.58) (-0.56)
College Graduate Dummy × Kids under 14 -0.1 0.6 -0.7
(-0.09) (0.45) (-0.94)
Age 30 to 39 -2.5∗ -0.7 -0.5
(-1.79) (-0.56) (-0.50)
Age 40 to 49 -3.4∗∗∗ -1.2 -0.5
(-2.79) (-0.90) (-0.56)
Age 50+ -2.7∗∗ -0.2 -0.7
(-2.23) (-0.19) (-0.82)
2019 earnings in bottom quartile 12.5∗∗∗ 1.8∗∗∗
(6.88) (3.53)
2019 earnings in 2nd quartile 6.2∗∗∗ 0.5
(4.76) (1.20)
2019 earnings in 3rd quartile 2.0∗∗∗ 0.1
(2.87) (0.49)
Firm size: Under 10 2.4 0.5
(1.27) (0.71)
Firm size: 10 to 99 1.0 -0.3
(0.58) (-0.63)
Firm size: 100 to 999 -0.8 0.1
(-0.74) (0.23)
Constant 0.179*** 0.205*** 0.121*** 0.0135
(10.19) (10.09) (6.20) (1.53)
Number of observations 9,660 9,273 9,273 6,193
R2 0.040 0.056 0.075 0.010
Table reports the coefficients estimates from equation (5) in the document. Standard errors
are double-clustered based on occupation and industry (t-statistics in parantheses). The
dependent variable is an indicator which equals one if an individual does not have a job in
April (columns 1-3) or February (column 4) 2020. To be included in the sample, workers
need to appear in the March 2020 CPS and March 2019 ASEC survey, and either the April
and/or February core CPS sample. The list of critical industries at the 4 digit NAICS level is
in Appendix Table A.1. We define critical Census industries as those for which at least 50%
of employment is associated with critical NAICS industries. The cross-sectional standard
deviation of the exposure measure for workers in our sample is 12.4%.
40
Table 4: Covid-19 Work Exposure, Stock Returns and Analyst Forecasts
Stock Analyst Forecast Revisions (%)
A. Regressions for Returns (%) 2020Q2 2020Q3 2020Q4 2020 2021 2022
Non-Critical Industries
(1) (2) (3) (4) (5) (6) (7)
Covid-19 Work Exposure -38.9∗∗∗ -43.9∗∗∗ -17.1∗∗ -8.3∗ -15.1∗∗ -11.6∗∗∗ -8.5∗∗∗
(-5.37) (-4.32) (-2.45) (-1.78) (-2.13) (-3.35) (-2.96)
Observations 141 126 127 127 127 127 92
R2 0.295 0.229 0.099 0.053 0.077 0.151 0.134
B. Averages, All Industries
Critical industries -8.0∗∗ -6.8∗∗∗ -4.5∗∗∗ -2.7∗∗ -3.2∗∗∗ -2.5∗∗ -2.3∗∗
(-2.02) (-3.08) (-3.17) (-2.54) (-2.71) (-2.15) (-2.49)
Non-critical industries -25.6∗∗∗ -26.5∗∗∗ -15.3∗∗∗ -9.4∗∗∗ -12.5∗∗∗ -9.9∗∗∗ -7.9∗∗∗
(-10.53) (-8.67) (-8.58) (-7.90) (-6.82) (-10.08) (-9.57)
Diff: Non-critical - Critical -17.6∗∗∗ -19.7∗∗∗ -10.7∗∗∗ -6.7∗∗∗ -9.2∗∗∗ -7.4∗∗∗ -5.6∗∗∗
(-3.78) (-5.24) (-4.69) (-4.22) (-4.23) (-4.84) (-4.52)
Number of observations 203 181 182 182 184 183 140
Table, panel A, plots the coefficients estimates from equations (10) and (8) in the document
for non-critical industries, where the dependent variables are constructed using firm data
from Compustat and Capital IQ, aggregated to the 4-digit NAICS level. The regression is
weighted by total Compustat 2019 employment in each industry. Standard errors are robust
to heteroskedasticity (White, 1980). Panel B reports averages of each outcome for critical and
non-critical industries, as well as differences between the two. The list of critical industries is
in Appendix Table A.1. The cross-sectional standard deviation of exposure across critical
industries is 17.4%
41
Table 5: Covid-19 Work Exposure and Default Probabilities
Change in probability of default (%) in next
A. Regressions for
Non-Critical Industries 3 months 6 months 12 months 24 months
(1) (2) (3) (4)
Covid-19 Work Exposure 0.77∗ 1.27∗∗ 1.91∗∗∗ 2.52∗∗∗
(1.81) (2.15) (2.62) (3.05)
Observations 155 155 155 155
R2 0.071 0.086 0.102 0.111
B. Averages, All Industries
Critical Industries 0.103∗∗∗ 0.195∗∗∗ 0.306∗∗∗ 0.307∗∗
(2.96) (3.03) (3.04) (2.50)
Non-Critical Industries 0.509∗∗∗ 0.858∗∗∗ 1.263∗∗∗ 1.483∗∗∗
(2.77) (3.28) (4.07) (4.71)
Diff: Non-critical - Critical 0.405∗∗ 0.663∗∗ 0.957∗∗∗ 1.176∗∗∗
(2.17) (2.46) (2.94) (3.48)
Number of observations 221 221 221 221
Table plots the coefficients estimates from equation (9) in the document for non-critical
industries, where the dependent variables are constructed using firm data from Compustat
and the RMI Credit Research Database, aggregated to the 4-digit NAICS level (t-statistics
in parantheses). The dependent variable is expressed in percentage points so that the slope
coefficient can be interpreted as the increase, in basis points, of the probability of default
associated with a 1 percentage point increase in exposure. The regression is weighted by total
Compustat 2019 employment in each industry. Standard errors are robust to heteroskedasticity
(White, 1980). The list of critical industries is in Appendix Table A.1. Panel B reports
average changes in default probabilities, in percentage points, for critical and non-critical
industries, respectively, as well as differences between the two means.
42
Appendix Tables and Figures
Table A.1: List of Critical Industries
NAICS Code Industry Name
1111 Oilseed and Grain Farming
1112 Vegetable and Melon Farming
1113 Fruit and Tree Nut Farming
1119 Other Crop Farming
1121 Cattle Ranching and Farming
1122 Hog and Pig Farming
1123 Poultry and Egg Production
1124 Sheep and Goat Farming
1129 Other Animal Production
1141 Fishing
1142 Hunting and Trapping
1151 Support Activities for Crop Production
1152 Support Activities for Animal Production
2121 Coal Mining
2122 Metal Ore Mining
2123 Nonmetallic Mineral Mining and Quarrying
2131 Support Activities for Mining
2211 Electric Power Generation, Transmission and Distribution
2212 Natural Gas Distribution
2213 Water, Sewage and Other Systems
2373 Highway, Street, and Bridge Construction
3111 Animal Food Manufacturing
3112 Grain and Oilseed Milling
3113 Sugar and Confectionery Product Manufacturing
3114 Fruit and Vegetable Preserving and Specialty Food Manufacturing
3115 Dairy Product Manufacturing
3116 Animal Slaughtering and Processing
3117 Seafood Product Preparation and Packaging
3118 Bakeries and Tortilla Manufacturing
3119 Other Food Manufacturing
3121 Beverage Manufacturing
3254 Pharmaceutical and Medicine Manufacturing
3256 Soap, Cleaning Compound, and Toilet Preparation Manufacturing
3261 Plastics Product Manufacturing
3312 Steel Product Manufacturing from Purchased Steel
3313 Alumina and Aluminum Production and Processing
3331 Agriculture, Construction, and Mining Machinery Manufacturing
43
Table A.1: List of Critical Industries (cont)
NAICS Code Industry Name
3391 Medical Equipment and Supplies Manufacturing
4242 Drugs and Druggists’ Sundries Merchant Wholesalers
4245 Farm Product Raw Material Wholesalers
4413 Automotive Parts, Accessories, and Tire Stores
4441 Building Material and Supplies Dealers
4451 Grocery Stores
4452 Specialty Food Stores
4453 Beer, Wine, and Liquor Stores
4461 Health and Personal Care Stores
4471 Gasoline Stations
4523 General Merchandise Stores, including Warehouse Clubs and Supercenters
4539 Other Miscellaneous Store Retailers
4541 Electronic Shopping and Mail-Order Houses
4812 Nonscheduled Air Transportation
4841 General Freight Trucking
4842 Specialized Freight Trucking
4851 Urban Transit Systems
4852 Interurban and Rural Bus Transportation
4853 Taxi and Limousine Service
4859 Other Transit and Ground Passenger Transportation
4861 Pipeline Transportation of Crude Oil
4862 Pipeline Transportation of Natural Gas
4885 Freight Transportation Arrangement
4911 Postal Service
4921 Couriers and Express Delivery Services
4922 Local Messengers and Local Delivery
4931 Warehousing and Storage
5173 Telecommunications Resellers
5179 Other Telecommunications
5211 Monetary Authorities-Central Bank
5221 Depository Credit Intermediation
5222 Nondepository Credit Intermediation
5223 Activities Related to Credit Intermediation
5231 Securities and Commodity Contracts Intermediation and Brokerage
5232 Securities and Commodity Exchanges
5239 Other Financial Investment Activities
5241 Insurance Carriers
5242 Agencies, Brokerages, and Other Insurance Related Activities
5251 Insurance and Employee Benefit Funds
44
Table A.1: List of Critical Industries (cont)
NAICS Code Industry Name
5259 Other Investment Pools and Funds
5411 Legal Services
5412 Accounting, Tax Preparation, Bookkeeping, and Payroll Services
5621 Waste Collection
5622 Waste Treatment and Disposal
5629 Remediation and Other Waste Management Services
6111 Elementary and Secondary Schools
6211 Offices of Physicians
6214 Outpatient Care Centers
6215 Medical and Diagnostic Laboratories
6216 Home Health Care Services
6219 Other Ambulatory Health Care Services
6221 General Medical and Surgical Hospitals
6223 Specialty (except Psychiatric and Substance Abuse) Hospitals
6231 Nursing Care Facilities (Skilled Nursing Facilities)
6233 Continuing Care Retirement Communities and Assisted Living Facilities for the Elderly
9211 Executive, Legislative, and Other General Government Support
9221 Justice, Public Order, and Safety Activities
9231 Administration of Human Resource Programs
9241 Administration of Environmental Quality Programs
9251 Administration of Housing Programs, Urban Planning, and Community Development
9261 Administration of Economic Programs
9271 Space Research and Technology
9281 National Security and International Affairs
45
Table A.2: Covid-19 Factor, portfolio weights (4-Digit NAICS)
Portfolio
Industry (NAICS) Weight (%)
4481: Clothing Stores 19.67
5613: Employment Services 15.95
7225: Restaurants and Other Eating Places 14.62
7211: Traveler Accommodation 11.23
4811: Scheduled Air Transportation 11.05
5151: Radio and Television Broadcasting 8.19
4522: Department Stores 6.12
7223: Special Food Services 5.99
3363: Motor Vehicle Parts Manufacturing 5.59
4831: Deep Sea, Coastal, and Great Lakes Water Transportation 5.57
3361: Motor Vehicle Manufacturing 5.51
3344: Semiconductor and Other Electronic Component Manufacturing 4.50
3329: Other Fabricated Metal Product Manufacturing 3.66
4411: Automobile Dealers 3.04
3339: Other General Purpose Machinery Manufacturing 2.76
7131: Amusement Parks and Arcades 2.75
3221: Pulp, Paper, and Paperboard Mills 2.66
5617: Services to Buildings and Dwellings 2.45
4244: Grocery and Related Product Merchant Wholesalers 2.28
5121: Motion Picture and Video Industries 2.10
3262: Rubber Product Manufacturing 2.02
3222: Converted Paper Product Manufacturing 1.96
4511: Sporting Goods, Hobby, and Musical Instrument Stores 1.90
3311: Iron and Steel Mills and Ferroalloy Manufacturing 1.88
2382: Building Equipment Contractors 1.63
2371: Utility System Construction 1.60
3252: Resin, Synthetic Rubber, and Artificial and Synthetic Fibers and Filaments Manufac- 1.48
turing
5616: Investigation and Security Services 1.44
3219: Other Wood Product Manufacturing 1.41
4431: Electronics and Appliance Stores 1.32
3353: Electrical Equipment Manufacturing 1.29
4482: Shoe Stores 1.29
4231: Motor Vehicle and Motor Vehicle Parts and Supplies Merchant Wholesalers 1.24
7132: Gambling Industries 1.19
7139: Other Amusement and Recreation Industries 1.14
3399: Other Miscellaneous Manufacturing 1.13
3141: Textile Furnishings Mills 1.12
46
Table A.2: Covid-19 Factor, portfolio weights (cont)
Portfolio
Industry (NAICS) Weight (%)
2362: Nonresidential Building Construction 1.10
5152: Cable and Other Subscription Programming 1.09
8123: Drycleaning and Laundry Services 1.02
3152: Cut and Sew Apparel Manufacturing 1.02
3324: Boiler, Tank, and Shipping Container Manufacturing 1.01
4236: Household Appliances and Electrical and Electronic Goods Merchant Wholesalers 1.00
2379: Other Heavy and Civil Engineering Construction 0.99
2361: Residential Building Construction 0.95
4532: Office Supplies, Stationery, and Gift Store 0.94
3359: Other Electrical Equipment and Component Manufacturing 0.90
3334: Ventilation, Heating, Air-Conditioning, and Commercial Refrigeration Equipment 0.90
Manufacturing
3366: Ship and Boat Building 0.87
3323: Architectural and Structural Metals Manufacturing 0.77
3231: Printing and Related Support Activities 0.74
8121: Personal Care Services 0.74
6222: Psychiatric and Substance Abuse Hospitals 0.71
4422: Home Furnishings Stores 0.70
3322: Cutlery and Handtool Manufacturing 0.69
5611: Office Administrative Services 0.69
3352: Household Appliance Manufacturing 0.66
3272: Glass and Glass Product Manufacturing 0.66
3362: Motor Vehicle Body and Trailer Manufacturing 0.66
5615: Travel Arrangement and Reservation Services 0.61
7113: Promoters of Performing Arts, Sports, and Similar Events 0.61
6244: Child Day Care Services 0.61
3332: Industrial Machinery Manufacturing 0.61
3251: Basic Chemical Manufacturing 0.55
3279: Other Nonmetallic Mineral Product Manufacturing 0.53
3371: Household and Institutional Furniture and Kitchen Cabinet Manufacturing 0.51
3169: Other Leather and Allied Product Manufacturing 0.47
3369: Other Transportation Equipment Manufacturing 0.45
5614: Business Support Services 0.43
3325: Hardware Manufacturing 0.41
3372: Office Furniture (including Fixtures) Manufacturing 0.35
4483: Jewelry, Luggage, and Leather Goods Stores 0.33
3379: Other Furniture Related Product Manufacturing 0.33
2383: Building Finishing Contractors 0.33
47
Table A.2: Covid-19 Factor, portfolio weights (cont)
Portfolio
Industry (NAICS) Weight (%)
4237: Hardware, and Plumbing and Heating Equipment and Supplies Merchant Wholesalers 0.32
7111: Performing Arts Companies 0.31
3351: Electric Lighting Equipment Manufacturing 0.28
8129: Other Personal Services 0.25
5619: Other Support Services 0.23
3212: Veneer, Plywood, and Engineered Wood Product Manufacturing 0.22
4238: Machinery, Equipment, and Supplies Merchant Wholesalers 0.19
3321: Forging and Stamping 0.18
3273: Cement and Concrete Product Manufacturing 0.18
6116: Other Schools and Instruction 0.15
3326: Spring and Wire Product Manufacturing 0.15
3132: Fabric Mills 0.15
4412: Other Motor Vehicle Dealers 0.14
4512: Book Stores and News Dealers 0.14
4832: Inland Water Transportation 0.13
3259: Other Chemical Product and Preparation Manufacturing 0.09
4883: Support Activities for Water Transportation 0.09
4421: Furniture Stores 0.08
5612: Facilities Support Services 0.08
3274: Lime and Gypsum Product Manufacturing 0.07
4881: Support Activities for Air Transportation 0.06
7224: Drinking Places (Alcoholic Beverages) 0.05
4869: Other Pipeline Transportation 0.04
4543: Direct Selling Establishment 0.04
2372: Land Subdivision 0.03
3211: Sawmills and Wood Preservation 0.03
6213: Offices of Other Health Practitioners 0.03
3335: Metalworking Machinery Manufacturing 0.02
4247: Petroleum and Petroleum Products Merchant Wholesalers 0.01
4889: Other Support Activities for Transportation 0.01
4243: Apparel, Piece Goods, and Notions Merchant Wholesalers 0.01
3327: Machine Shops; Turned Product; and Screw, Nut, and Bolt Manufacturing 0.00
4442: Lawn and Garden Equipment and Supplies Stores 0.00
7114: Agents and Managers for Artists, Athletes, Entertainers, and Other Public Figures 0.00
8111: Automotive Repair and Maintenance 0.00
7112: Spectator Sports 0.00
4239: Miscellaneous Durable Goods Merchant Wholesalers -0.01
5174: Satellite Telecommunications -0.05
48
Table A.2: Covid-19 Factor, portfolio weights (cont)
Portfolio
Industry (NAICS) Weight (%)
8122: Death Care Services -0.09
4251: Wholesale Electronic Markets and Agents and Brokers -0.09
4241: Paper and Paper Product Merchant Wholesalers -0.09
5414: Specialized Design Services -0.10
2111: Oil and Gas Extraction -0.15
4246: Chemical and Allied Products Merchant Wholesalers -0.24
3343: Audio and Video Equipment Manufacturing -0.29
4233: Lumber and Other Construction Materials Merchant Wholesalers -0.40
5419: Other Professional, Scientific, and Technical Services -0.50
3333: Commercial and Service Industry Machinery Manufacturing -0.52
4249: Miscellaneous Nondurable Goods Merchant Wholesalers -0.57
5111: Newspaper, Periodical, Book, and Directory Publishers -0.60
5418: Advertising, Public Relations, and Related Services -1.05
3241: Petroleum and Coal Products Manufacturing -1.08
3336: Engine, Turbine, and Power Transmission Equipment Manufacturing -1.32
5413: Architectural, Engineering, and Related Services -1.93
5416: Management, Scientific, and Technical Consulting Services -2.33
3255: Paint, Coating, and Adhesive Manufacturing -2.70
5417: Scientific Research and Development Services -2.91
4234: Professional and Commercial Equipment and Supplies Merchant Wholesalers -3.14
3365: Railroad Rolling Stock Manufacturing -3.47
3341: Computer and Peripheral Equipment Manufacturing -5.69
3342: Communications Equipment Manufacturing -10.47
3345: Navigational, Measuring, Electromedical, and Control Instruments Manufacturing -10.51
3364: Aerospace Product and Parts Manufacturing -14.72
5112: Software Publishers -16.97
5415: Computer Systems Design and Related Services -31.57
5182: Data Processing, Hosting, and Related Services -35.46
5191: Other Information Services -41.46
49
File and source
- File
- working-remotely-and-the-supply-side-impact-of-covid-19.pdf
- Size
- 729,682 bytes
- SHA-256
- 73afa10ec85d33de080d40605364cf8c7041789d05e2d700782e5287f694973a
- Original
- doi.org