causal impact of masks, policies, behavior on early covid ...€¦ · 27/05/2020 · causal impact...

CAUSAL IMPACT OF MASKS POLICIES BEHAVIOR ON EARLY

COVID-19 PANDEMIC IN THE US

VICTOR CHERNOZHUKOV HIROYUKI KASAHARA AND PAUL SCHRIMPF

Abstract This paper evaluates the dynamic impact of various policies such as schoolbusiness and restaurant closures adopted by the US states on the growth rates of con-firmed Covid-19 cases and social distancing behavior measured by Google Mobility Re-ports where we take into consideration of peoplersquos voluntarily behavioral response to newinformation of transmission risks Using the US state-level data our analysis finds thatboth policies and information on transmission risks are important determinants of peoplersquossocial distancing behavior and shows that a change in policies explains a large fraction ofobserved changes in social distancing behavior Our counterfactual experiments indicatethat removing all policies would have lead to 30 to 200 times more additional cases by lateMay Removing only the non-essential businesses closures (while maintaining restrictionson movie theaters and restaurants) would have increased the weekly growth rate of casesbetween -002 and 006 and would have lead to -10 to 40 more cases by late MayFinally nationally mandating face masks for employees on April 1st would have reducedthe case growth rate by 01-025 This leads to 30 to 57 fewer reported cases by lateMay which translates into roughly 30-57 thousand saved lives

1 Introduction

Accumulating evidence suggests that various policies in the US have reduced social inter-actions and have slowed down the growth of Covid-19 infections1 An important outstand-ing issue is however how much of the observed slow down in the spread is attributable tothe effect of policies per se relative to a voluntarily change in peoplersquos behavior out of fearof being infected This question is critical to evaluate the effectiveness of restrictive policiesin the US relative to an alternative policy of just providing recommendations and informa-tion such as the one adopted by Sweden More generally understanding peoplersquos dynamicbehavioral response to policies and information is indispensable for properly evaluating theeffect of policies on the spread of Covid-19

This paper quantitatively assesses the impact of various policies adopted by the USstates on the spread of Covid-19 such as non-essential businesses closure and mandatoryface masks paying a particular attention to how people adjust their behavior in responseto policies as well as new information on cases

Date May 29 2020Key words and phrases Covid-19 causal impact masks non-essential business policies behavior1See Courtemanche et al (2020) Hsiang et al (2020) Pei Kandula and Shaman (2020) and Abouk and

Heydari (2020)

1

All rights reserved No reuse allowed without permission (which was not certified by peer review) is the authorfunder who has granted medRxiv a license to display the preprint in perpetuity

The copyright holder for this preprintthis version posted May 30 2020 httpsdoiorg1011012020052720115139doi medRxiv preprint

NOTE This preprint reports new research that has not been certified by peer review and should not be used to guide clinical practice

2 VICTOR CHERNOZHUKOV HIROYUKI KASAHARA AND PAUL SCHRIMPF

We present a conceptual framework that spells out the causal structure on how theCovid-19 spread is dynamically determined by policies and human behavior Our approachexplicitly recognizes that policies not only directly affect the spread of Covid-19 (eg maskrequirement) but also indirectly affect its spread by changing peoplersquos behavior (eg stayat home order) It also recognizes that people react to new information on Covid-19 casesand voluntarily adjust their behavior (eg voluntary social distancing and hand washing)even without any policy in place Our casual model provides a framework to quantita-tively decompose the growth of Covid-19 cases into three components (1) direct policyeffect (2) policy effect through behavior and (3) direct behavior effect in response to newinformation2

Guided by the causal model our empirical analysis examines how the weekly growth ratesof confirmed Covid-19 cases are determined by policies and behavior using the US state-level data To examine how policies and information affect peoplersquos behavior we also regresssocial distancing measures on policy and information variables Our regression specificationfor case growth is explicitly driven from a SIR model and the estimated regression coefficientscan be interpreted from the viewpoint of the effective reproduction number although ourcausal approach does not hinge on validity of a SIR model

As policy variables we consider state of emergency mandatory face masks for employ-ees in public business stay at home order closure of school closure of restaurants excepttake out closure of movie theaters and closure of non-essential business Our behaviorvariables are four mobility measures that capture the intensity of visits to ldquotransitrdquo ldquogro-ceryrdquo ldquoretailrdquo and ldquoworkplacesrdquo from Google Mobility Reports We take time trend thelagged growth rate of cases and the log of lagged cases as our measures of information oninfection risks that affects peoplersquos behavior We also consider the growth rate of tests thelog number of tests and the state-level characteristics (eg population size and total area)as the confounders that has to be controlled for in order to identify the causal relationshipbetween policybehavior and the growth rate of cases

Our key findings from regression analysis are as follows We find that both policies andinformation on past cases are important determinants of peoplersquos social distancing behaviorwhere policy effects explain at around 80 or more of observed decline in four behaviorvariables We also find that people dynamically adjust their behavior in response to policiesand information with time lag

There are both large policy effects and large behavior effects on case growth The esti-mates from our lag cases information model imply that all policies combined would reducethe weekly growth rate of cases by around 1 Taking into account of a change in thenumber of tests the weekly growth rates of infections arguably declined from around 2 in

2The causal model is framed using the language of structural equations models and causal diagrams ofeconometrics (Wright (1928) Haavelmo (1944) Heckman and Vytlacil (2007) see Greenland Pearl andRobins (1999) for developments in computer science) with natural unfolding potential outcomes represen-tation (Rubin (1974) Tinbergen (1930) Neyman (1925) Imbens and Rubin (2015)) As such it naturallyconverses in all languages for causal inference



CAUSAL IMPACT OF MASKS POLICIES BEHAVIOR 3

early March to near 0 in April Our result suggests that total policy effects explain aroundone-half of this decline

Using the estimated model we evaluate the dynamic impact of the following counterfac-tual policies on Covid-19 cases removing all policies allowing non-essential businesses toopen and mandating face masks The results of counterfactual experiments show a largeimpact of some of those policies on the number of cases They also highlight the importanceof voluntary behavioral response to infection risks for evaluating the dynamic policy effects

Figure 1 shows how removing all policies would have changed the average growth rate ofcases as well as the number of cases across states The effect of removing policies on casegrowth peaks at 06 after two to three weeks and then gradually declines This declinehappens because people voluntarily adjust their behavior in response to information on ahigher number of cases The estimate implies that there would have been 30-200 timesmore cases (10 to 100 million additional cases) by late May if all policies had never beenimplemented

Figure 2 illustrates how allowing non-essential businesses open would have affected thecase growth and cases Keeping non-essential businesses open (other than movie theatersgyms and keeping restaurants in the ldquotake-outrdquo mode) would have increased the casesby minus10 to 40 These estimates contribute to the ongoing efforts of evaluating variousreopening approaches (Stock 2020a)

Figure 3 show that implementing mandatory face masks on April 1st would have reducedthe case growth rate by 01-025 leading to reductions of minus30 to minus57 in reportedcases in late May This roughly implies that 30-57 thousand lives would have been saved3

This finding is significant given this potentially large benefit for reducing the spread ofCovid-19 mask mandates is an attractive policy instrument especially because it involvesrelatively little economic disruption Again these estimates contribute to the ongoing effortsof designing approaches to minimize risks from reopening

A growing number of other papers have examined the link between non pharmaceuticalinterventions and Covid-19 cases4 Hsiang et al (2020) estimate the effect of policies onCovid-19 case growth rate using data from the United States China Iran Italy Franceand South Korea In the United States they find that the combined effect on the growthrate of all policies they consider is minus0347 (0061) Courtemanche et al (2020) use UScounty level data to analyze the effect of interventions on case growth rates They find thatthe combination of policies they study reduced growth rates by 91 percentage points 16-20days after implementation out of which 59 percentage points is attributed to shelter inplace orders Both Hsiang et al (2020) and Courtemanche et al (2020) adopted ldquoreduced-formrdquo approach to estimate the total policy effect on case growth without using any social

3As of May 27 2020 the US Centers for Disease Control and Prevention reports 99031 deaths in theUS

4We refer the reader to Avery et al (2020) for a comprehensive review of a larger body of work researchingCovid-19 here we focus on few quintessential comparisons on our work with other works that we are awareof




00

02

04

06

08

Mar Apr Maydate

Cha

nge

in ∆

log∆

C

Effect of removing policies on case growth

0

50

100

150

200

Mar Apr Maydate

rela

tive

incr

ease

in c

ases

Relative effect of removing policies on cases

Figure 1 Average change in case growth and relative change in cases ofremoving all policies

minus002

000

002

004

006

Mar Apr Maydate

Cha

nge

in ∆

log∆

C

Effect of never shutting down non essential businesses on case growth

minus01

00

01

02

03

04

05

Mar Apr Maydate

rela

tive

incr

ease

in c

ases

Relative effect of never shutting down non essential businesses on cases

Figure 2 Average change in case growth and and relative change in casesfrom leaving non-essential businesses open

distancing behavior measures In contrast our study highlights the role of behavioralresponse to policies and information

Existing evidence for the impact of social distancing policies on behavior in the US ismixed Abouk and Heydari (2020) employ a difference-in-differences methodology to findthat statewide stay-at-home orders has strong causal impact on reducing social interactions




minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C

Effect of mandating masks on April 1st on case growth

minus06

minus04

minus02

00

Mar Apr Maydate

rela

tive

incr

ease

in c

ases

Relative effect of mandating masks on April 1st on cases

Figure 3 Average change in case growth and and relative change in casesfrom mandating masks for employees on April 1st

In contrast using the data from Google Mobility Reports Maloney and Taskin (2020) findthat the increase in social distancing is largely voluntary and driven by information5 An-other study by Gupta et al (2020) also found little evidence that stay-at-home mandatesinduced distancing using mobility measures from PlaceIQ and SafeGraph Using data fromSafeGraph Andersen (2020) show that there has been substantial voluntary social distanc-ing but also provide evidence that mandatory measures such as stay at home order havebeen effective at reducing the frequency of visits outside of onersquos home

Pei Kandula and Shaman (2020) use county-level observations of reported infectionsand deaths in conjunction with mobility data from SafeGraph to estimate how effectivereproductive numbers in major metropolitan areas change over time They conduct simu-lation of implementing all policies 1-2 weeks earlier and found that it would have resultedin reducing the number of cases and deaths by more than half Without using any policyvariables however their study does not explicitly analyze how policies are related to theeffective reproduction numbers

Epidemiologists use model simulations to predict how cases and deaths evolve for thepurpose of policy recommendation As reviewed by Avery et al (2020) there exist sub-stantial uncertainty in parameters associated with the determinants of infection rates andasymptomatic rates (Stock 2020b) Simulations are often done under strong assumptionsabout the impact of social distancing policies without connecting to the relevant data (egFerguson et al 2020) Furthermore simulated models do not take into account that peoplemay limit their contact with other people in response to higher transmission risks When

5Specifically they find that of the 60 percentage point drop in workplace intensity 40 percentage pointscan be explained by changes in information as proxied by case numbers while roughly 8 percentage pointscan be explained by policy changes




such a voluntary behavioral response is ignored simulations would produce exponentialspread of disease and would over-predict cases and deaths Our counterfactual experimentsillustrate the importance of this voluntary behavioral change

Whether wearing masks in public place should be mandatory or not has been one of themost contested policy issues where health authorities of different countries provide con-tradicting recommendations Reviewing evidence Greenhalgh et al (2020) recognizes thatthere is no randomized controlled trial evidence for effectiveness of face masks but theystate ldquoindirect evidence exists to support the argument for the public wearing masks inthe Covid-19 pandemicrdquo6 Howard et al (2020) also review available medical evidence andconclude that ldquomask wearing reduces the transmissibility per contact by reducing trans-mission of infected droplets in both laboratory and clinical contextsrdquo Given the lack ofexperimental evidence conducting observational studies is useful and important To thebest of our knowledge our paper is the first empirical study that shows the effectiveness ofmask mandates on reducing the spread of Covid-19 by analyzing the US state-level dataThis finding corroborates and is complementary to the medical observational evidence inHoward et al (2020)

2 The Causal Model for the Effect of Policies Behavior and Informationon Growth of Infection

21 The Causal Model and Its Structural Equation Form We introduce our ap-proach through the Wright-style causal diagram shown in Figure 4 The diagram describeshow policies behavior and information interact together

bull The forward health outcome Yit is determined last after all other variables havebeen determinedbull The adopted policies Pit affect Yit either directly or indirectly by altering human

behavior Bitbull Information variables Iit such as lagged valued of outcomes can affect both human

behavior adopted policies and future outcomebull The confounding factors Wit which could vary across states and time affect all

other variables

The index i describes observational unit the state and t describes the time

We begin to introduce more context by noting that our main outcome of interest isthe forward growth rate in the reported new Covid-19 cases behavioral variables includeproportion of time spent in transit or shopping and others policy variables include stay-at-home order and school and business closure variables the information variables includelagged values of outcome We provide detailed description and time of these variables inthe next section

6The virus remains viable in the air for several hours for which surgical masks may be effective Also asubstantial fraction of individual who are infected become infectious before showing symptom onset




Pit

Iit Yit

Bit

Iit

Wit

Figure 4 P Wright type causal path diagram for our model

The causal structure allows for the effect of the policy to be either direct or indirect ndashthrough-behavior or through dynamics and all of these effects are not mutually exclusiveThe structure allows for changes in behavior to be brought by change in policies and infor-mation These are all realistic properties that we expect from the contextual knowledge ofthe problem Policy such as closures of schools non-essential business restaurants alterand constrain behavior in strong ways In contrast policies such as mandating employeesto wear masks can potentially affect the Covid-19 transmission directly The informationvariables such as recent growth in the number of cases can cause people to spend moretime at home regardless of adopted state policies these changes in behavior in turn affectthe transmission of Covid-19

The causal ordering induced by this directed acyclical graph is determined by the follow-ing timing sequence within a period

(1) confounders and information get determined(2) policies are set in place given information and confounders(3) behavior is realized given policies information and confounders and(4) outcomes get realized given policies behavior and confounders

The model also allows for direct dynamic effect of the information variables on the out-come through autoregressive structures that capture persistence in the growth patternsAs further highlighted below realized outcomes may become new information for futureperiods inducing dynamics over multiple periods

Our quantitative model for causal structure in Figure 4 is given by the following econo-metric structural equation model

Yit(b p) =αprimeb+ πprimep+ microprimeι+ δprimeYWit + εyit εyit perp IitWit

Bit(p ι) =βprimep+ γprimeι+ δprimeBWit + εbit εbit perp IitWit

(SEM)

which is a collection of functional relations with stochastic shocks decomposed into observ-able part δW and unobservable part ε The terms εyit and εbit are the centered stochasticshocks that obey the orthogonality restrictions posed in these equations (We say thatV perp U if EV U = 0)




The policies can be modeled via a linear form as well

Pit(ι) = ηprimeι+ δprimePWit + εpit εpit perp IitWit (P)

although our identification and inference strategies do not rely on the linearity7

The observed variables are generated by setting ι = Iit and propagating the system fromthe last equation to the first

Yit =Yit(Bit Pit Iit)

Bit =Bit(Pit Iit)

Pit =Pit(Iit)

(O)

The system above implies the following collection of stochastic equations for realizedvariables

Yit = αprimeBit + πprimePit + microprimeIit + δprimeYWit + εyit εyit perp Bit Pit IitWit (BPIrarrY)

Bit = βprimePit + γprimeIit + δprimeBWit + εbit εbit perp Pit IitWit (PIrarrB)

Pit = ηprimeIit + δprimePWit + εpit εpit perp IitWit (IrarrP)

which in turn imply

Yit = (αprimeβprime + πprime)Pit + (αprimeγprime + microprime)Iit + δprimeWit + εit εit perp Pit IitWit (PIrarrY)

for δprime = αprimeδprimeB + δprimeY and εit = αprimeεbit + εyit These equations form the basis of our empiricalanalysis

Identification and Parameter Estimation The orthogonality equations imply thatthese are all projection equations and the parameters of SEM are identified by the pa-rameters of these regression equation provided the latter are identified by sufficient jointvariation of these variables across states and time

The last point can be stated formally as follows Consider the previous system of equa-tions after partialling out the confounders

Yit =αprimeBit + πprimePit + microprimeIit + εyit εyit perp Bit Pit Iit

Bit =βprimePit + γprimeIit + εbit εbit perp Pit Iit

Pit =ηprimeIit + εpit εpit perp Iit

(1)

where Vit = VitminusW primeitE[WitWprimeit]minusE[WitVit] denotes the residual after removing the orthogonal

projection of Vit on Wit The residualization is a linear operator implying that (1) followsimmediately from above The parameters of (1) are identified as projection coefficients inthese equations provided that residualized vectors appearing in each of the equations havenon-singular variance that is

Var(P primeit Bprimeit I

primeit) gt 0 Var(P primeit I

primeit) gt 0 and Var(I primeit) gt 0 (2)

7This can be replaced by an arbitrary non-additive function Pit(ι) = p(ιWit εpit) where (possibly infinite-

dimensional) εpit is independent of Iit and Wit




Our main estimation method is the standard correlated random effects estimator wherethe random effects are parameterized as functions of observable characteristic The state-level random effects are modeled as a function of state level characteristics and the timerandom effects are modeled by including a smooth trend and its interaction with state levelcharacteristics The stochastic shocks εitTt=1 are treated as independent across states iand can be arbitrarily dependent across time t within a state

A secondary estimation method is the fixed effects estimator where Wit includes latent(unobserved) state level effects Wi and and time level effects Wt which must be estimatedfrom the data This approach is much more demanding on the data and relies on longcross-sectional and time histories When histories are relatively short large biases emergeand they need to be removed using debiasing methods In our context debiasing materiallychange the estimates often changing the sign8 However we find the debiased fixed effectestimates are qualitatively and quantitatively similar to the correlated random effects esti-mates Given this finding we chose to focus on the latter as a more standard and familiarmethod and report the former estimates in the supplementary materials for this paper9

22 Information Structures and Induced Dynamics We consider three examples ofinformation structures

(I1) Information variable is time trend

Iit = log(t)

(I2) Information variable is lagged value of outcome

Iit = Yitminus`

(I3) Information variables include trend lagged and integrated values of outcome

Iit =

(log(t) Yitminus`

infinsumm=1

Yitminus`m

)prime

with the convention that Yit = 0 for t le 0

The first information structure captures the basic idea that as individuals discover moreinformation about covid over time they adapt safer modes of behavior (stay at homewear masks wash hands) Under this structure information is common across states andexogenously evolves over time independent of the number of cases The second structurearises for considering autoregressive components and captures peoplersquos behavioral responseto information on cases in the state they reside Specifically we model persistence in growthYit rate through AR(1) model which leads to Iit = Yitminus7 This provides useful local state-specific information about the forward growth rate and people may adjust their behaviorto safer modes when they see a high value We model this adjustment via the term γprimeIt in

8This is cautionary message for other researchers intending to use fixed effects estimator in the contextof Covid-19 analysis Only debiased fixed effects estimators must be used

9The similarity of the debiased fixed effects and correlated random effects served as a useful specificationcheck Moreover using the fixed effects estimators only yielded minor gains in predictive performances asjudging by the adjusted R2rsquos providing another useful specification check




the behavior equation The third information structure is the combination of the first twostructures plus an additional term representing the (log of) total number of new cases inthe state We use this information structure in our empirical specification In this structurepeople respond to both global information captured by trend and local information sourcescaptured by the local growth rate and the total number of cases The last element of theinformation set can be thought of as local stochastic trend in cases

All of these examples fold into a specification of the form

Iit = Iit(Ii(tminus`) Yi(tminus`) t) t = 1 T (I)

with the initialization Ii0 = 0 and Yi0 = 010

With any structure of this form realized outcomes may become new information forfuture periods inducing a dynamical system over multiple periods We show the resultingdynamical system in a causal diagram of Figure 5 Specification of this system is usefulfor studying delayed effects of policies and behaviors and in considering the counterfactualpolicy analysis

Ii(tminus7)

Yi(tminus7)

Iit

Yit

Ii(t+7)

Yi(t+7)

SE

M(t-7

)

SE

M(t)

SE

M(t+

7)

Figure 5 Dynamic System Induced by Information Structure and SEM

23 Outcome and Key Confounders via SIR Model Letting Cit denotes cumulativenumber of confirmed cases in state i at time t our outcome

Yit = ∆ log(∆Cit) = log(∆Cit)minus log(∆Citminus7) (3)

approximates the weekly growth rate in new cases from t minus 7 to t11 Here ∆ denotes thedifferencing operator over 7 days from t to tminus 7 so that ∆Cit = CitminusCitminus7 is the numberof new confirmed cases in the past 7 days

We chose this metric as this is the key metric for policy makers deciding when to relaxCovid mitigation policies The US governmentrsquos guidelines for state reopening recommendthat states display a ldquodownward trajectory of documented cases within a 14-day periodrdquo

10The initialization is appropriate in our context for analyzing pandemics from the very beginning butother initializations could be appropriate in other contexts

11We may show that log(∆Cit) minus log(∆Citminus7) approximates the average growth rate of cases from tminus 7to t




(White House 2020) A negative value of Yit is an indication of meeting this criteria forreopening By focusing on weekly cases rather than daily cases we smooth idiosyncraticdaily fluctuations as well as periodic fluctuations associated with days of the week

Our measurement equation for estimating equations (BPIrarrY) and (PIrarrY) will take theform

Yit = θprimeXit + δT∆ log(T )it + δDTit∆Dit

∆Cit+ εit (M)

where i is state t is day Cit is cumulative confirmed cases Dit is deaths Tit is the number oftests over 7 days ∆ is a 7-days differencing operator εit is an unobserved error term whereXit collects other behavioral policy and confounding variables depending on whether weestimate (BPIrarrY) or (PIrarrY) Here

∆ log(T )it and δDTit∆Dit

∆Cit

are the key confounding variables derived from considering the SIR model below Wedescribe other confounders in the empirical section

Our main estimating equation (M) is motivated by a variant of a SIR model where weincorporate a delay between infection and death instead of a constant death rate and weadd confirmed cases to the model Let S I R and D denote the number of susceptibleinfected recovered and dead individuals in a given state or province Each of these variablesare a function of time We model them as evolving as

S(t) = minusS(t)

Nβ(t)I(t) (4)

I(t) =S(t)

Nβ(t)I(t)minus S(tminus `)

Nβ(tminus `)I(tminus `) (5)

R(t) = (1minus π)S(tminus `)N

β(tminus `)I(tminus `) (6)

D(t) = πS(tminus `)N


where N is the population β(t) is the rate of infection spread ` is the duration betweeninfection and death and π is the probability of death conditional on infection12

Confirmed cases C(t) evolve as

C(t) = τ(t)I(t) (8)

where τ(t) is the rate that infections are detected

Our goal is to examine how the rate of infection β(t) varies with observed policies andmeasures of social distancing behavior A key challenge is that we only observed C(t) and

12The model can easily be extended to allow a non-deterministic disease duration This leaves our mainestimating equation (9) unchanged as long as the distribution of duration between infection and death isequal to the distribution of duration between infection and recovery




D(t) but not I(t) The unobserved I(t) can be eliminated by differentiating (8) and using(5) and (7) as

C(t)

C(t)=S(t)

Nβ(t) +

τ(t)

τ(t)minus 1

π

τ(t)D(t)

C(t) (9)

We consider a discrete-time analogue of equation (9) to motivate our empirical specification

by relating the detection rate τ(t) to the number of tests Tit while specifying S(t)N β(t) as a

linear function of variables Xit This results in

YitC(t)

C(t)

= X primeitθ + εitS(t)N

β(t)

+ δT∆ log(T )itτ(t)τ(t)

+ δDTit∆Dit

∆Cit

1πτ(t)D(t)

C(t)

which is equation (M) where Xit captures a vector of variables related to β(t)

Structural Interpretation In the early pandemics when the num-ber of susceptibles is approximately the same as the entire population ieSitNit asymp 1 the component X primeitθ is the projection of infection rate βi(t)on Xit (policy behavioral information and confounders other than testingrate) provided the stochastic component εit is orthogonal to Xit and thetesting variables

βi(t)SitNit = X primeitθ + εit εit perp Xit

3 Decomposition and Counterfactual Policy Analysis

31 Total Change Decomposition Given the SEM formulation above we can carryout the following decomposition analysis after removing the effects of confounders Forexample we can decompose the total change EYitminusEYio in the expected outcome measuredat two different time points t and o = tminus ` into sum of three components

EYit minus EYioTotal Change

= αprimeβprime(

EPit minus EPio

)Policy Effect via Behavior

+ πprime(

EPit minus EPio

)Direct Policy Effect

+ αprimeγprime(

EIit minus EIio

)+ microprime

(EIit minus EIio

)Dynamic Effect

= PEBt + PEDt + DynEt

(10)

where the first two components are immediate effect and the third is delayed or dynamiceffect




In the three examples of information structure given earlier we have the following formfor the dynamic effect

(I1) DynEt = γα log `+ micro log `

(I2) DynEt =sumt`

m=1 (γα+ micro)m (PEBtminusm` + PEDtminusm`)

(I3) DynEt =sumt`

m=0 (((γα)2 + micro2 + (γα)3 + micro3)m ((γα)1 log `+ micro1 log `)

+sumt`

m=1 (((γα)2 + micro2 + (γα)3 + micro3)m (PEBtminusm` + DPEtminusm`)

+sumt`minus1

m=1 ((γα)3 + micro3)m(PEBtminus(m+1)` + DPEtminus(m+1)`

)

The effects can be decomposed into (a) delayed policy effects via behavior by summing termscontaining PEB (b) delayed policy effects via direct impact by summing terms containingDPE and (c) pure behavior effects and (d) pure dynamic feedback effects

32 Counterfactuals We also consider simple counterfactual exercises where we examinethe effects of setting a sequence of counterfactual policies for each state

P itTt=1 i = 1 N

We assume that the SEM remains invariant except of course for the policy equation13

Given the policies we propagate the dynamic equations

Y it =Yit(B

it P

it I

it)

Bit =Bit(P

it I

it)

Iit =Iit(Ii(tminus1) Y

lowasti(tminus1) t)

(CEF-SEM)

with the initialization Ii0 = 0 Y i0 = 0 B

i0 = 0 P i0 = 0 In stating this counterfactualsystem of equations we make the following invariance assumption

Invariance Assumption The equations of (CF-SEM) remain exactly ofthe same form as in the (SEM) and (I) That is under the policy interventionP it that is parameters and stochastic shocks in (SEM) and (I) remainthe same as under the original policy intervention Pit

Let PY it and PYit denote the predicted values produced by working with the counter-

factual system (CEF-SEM) and the factual system (SEM)

PY it = (αprimeβprime + πprime)P it + (αprimeγprime + microprime)Iit + δprimeWit

PYit = (αprimeβprime + πprime)Pit + (αprimeγprime + microprime)Iit + δprimeWit

In generating these predictions we make the assumption of invariance stated above

13It is possible to consider counterfactual exercises in which policy responds to information through thepolicy equation if we are interested in endogenous policy responses to information Although this is beyondthe scope of the current paper counterfactual experiments with endogenous government policy would beimportant for example to understand the issues related to the lagged response of government policies tohigher infection rates due to incomplete information




Then we can write the difference into the sum of three components

PY it minus PYit

Predicted CF Change

= αprimeβprime(P it minus Pit)CF Policy Effect via Behavior

+ πprime (P it minus Pit)CF Direct Effect

+ αprimeγprime (Iit minus Iit) + microprime (Iit minus Iit)CF Dynamic Effect

= PEBit + PED

it + DynEit

(11)

Similarly to what we had before the counterfactual dynamic effects take the form

(I1) DynEit = γα log `+ micro log `

(I2) DynEit =sumt`

m=1 (γα+ micro)m (PEBi(tminusm`) + PED

i(tminusm`))

(I3) DynEit =sumt`

m=0 ((γα)2 + micro2 + (γα)3 + micro3)m ((γα)1 log `+ micro1 log `)

+sumt`

m=1 ((γα)2 + micro2 + (γα)3 + micro3)m(

PEBi(tminusm`) + DPEi(tminusm)`

)+sumt`minus1

m=1 ((γα)3 + micro3)m(

PEBi(tminus(m+1)`) + DPEi(tminus(m+1)`)

)


4 Empirical Analysis

41 Data Our baseline measures for daily Covid-19 cases and deaths are from The NewYork Times (NYT) When there are missing values in NYT we use reported cases anddeaths from JHU CSSE and then the Covid Tracking Project The number of tests foreach state is from Covid Tracking Project As shown in Figure 23 in the appendix therewas a rapid increase in testing in the second half of March and then the number of testsincreased very slowly in each state in April

We use the database on US state policies created by Raifman et al (2020) In ouranalysis we focus on 7 policies state of emergency stay at home closed nonessentialbusinesses closed K-12 schools closed restaurants except takeout close movie theatersand mandate face mask by employees in public facing businesses We believe that the firstfive of these policies are the most widespread and important Closed movie theaters isincluded because it captures common bans on gatherings of more than a handful of peopleWe also include mandatory face mask use by employees because its effectiveness on slowingdown Covid-19 spread is a controversial policy issue (Howard et al 2020 Greenhalgh et al2020) Table 1 provides summary statistics where N is the number of states that have everimplemented the policy We also obtain information on state-level covariates from Raifman




et al (2020) which include population size total area unemployment rate poverty rateand a percentage of people who are subject to illness

N Min Median MaxState of emergency 51 2020-02-29 2020-03-11 2020-03-16

Date closed K 12 schools 51 2020-03-13 2020-03-17 2020-04-03Stay at home shelter in place 42 2020-03-19 2020-03-28 2020-04-07

Closed movie theaters 49 2020-03-16 2020-03-21 2020-04-06Closed restaurants except take out 48 2020-03-15 2020-03-17 2020-04-03

Closed non essential businesses 43 2020-03-19 2020-03-25 2020-04-06Mandate face mask use by employees 36 2020-04-03 2020-05-01 2020-05-11

Table 1 State Policies

We obtain social distancing behavior measures fromldquoGoogle COVID-19 Community Mo-bility Reportsrdquo (LLC 2020) The dataset provides six measures of ldquomobility trendsrdquo thatreport a percentage change in visits and length of stay at different places relative to abaseline computed by their median values of the same day of the week from January 3to February 6 2020 Our analysis focuses on the following four measures ldquoGrocery amppharmacyrdquo ldquoTransit stationsrdquo ldquoRetail amp recreationrdquo and ldquoWorkplacesrdquo14

In our empirical analysis we use weekly measures for cases deaths and tests by summingup their daily measures from day t to t minus 6 We focus on weekly cases because daily newcases are affected by the timing of reporting and testing and are quite volatile as shown inFigure 20 in the appendix Similarly reported daily deaths and tests are subject to highvolatility Aggregating to weekly new casesdeathstests smooths out idiosyncratic dailynoises as well as periodic fluctuations associated with days of the week We also constructweekly policy and behavior variables by taking 7 day moving averages from day t minus 14 totminus 21 where the delay reflects the time lag between infection and case confirmation Thefour weekly behavior variables are referred as ldquoTransit Intensityrdquo ldquoWorkplace IntensityrdquoldquoRetail Intensityrdquo and ldquoGrocery Intensityrdquo Consequently our empirical analysis uses 7days moving averages of all variables recorded at daily frequencies

Table 2 reports that weekly policy and behavior variables are highly correlated with eachother except for theldquomasks for employeesrdquo policy High correlations may cause multicol-inearity problem and potentially limits our ability to separately identify the effect of eachpolicy or behavior variable on case growth but this does not prevent us from identifying theaggregate effect of all policies and behavior variables on case growth

42 The Effect of Policies and Information on Behavior We first examine how poli-cies and information affect social distancing behaviors by estimating a version of (PIrarrB)

Bjit = (βj)primePit + (γj)primeIit + (δjB)primeWit + εbjit

14The other two measures are ldquoResidentialrdquo and ldquoParksrdquo We drop ldquoResidentialrdquo because it is highlycorrelated with ldquoWorkplacesrdquo and ldquoRetail amp recreationrdquo at correlation coefficients of -099 and -098 as shownin Table 2 We also drop ldquoParksrdquo because it does not have clear implication on the spread of Covid-19




Table 2 Correlations among Policies and Behavior

resi

den

tial

tran

sit

wor

kp

lace

s

reta

il

groce

ry

clos

edK

-12

sch

ool

s

stay

ath

ome

clos

edm

ovie

thea

ters

clos

edre

stau

rants

clos

edb

usi

nes

ses

mas

ks

for

emp

loye

es

stat

eof

emer

gen

cy

residential 100transit -095 100workplaces -099 093 100retail -098 093 096 100grocery -081 083 078 082 100

closed K-12 schools 090 -080 -093 -087 -065 100stay at home 073 -075 -074 -070 -072 070 100closed movie theaters 083 -076 -085 -081 -069 088 078 100closed restaurants 085 -079 -085 -084 -070 084 074 088 100closed businesses 073 -070 -073 -071 -070 066 080 075 075 100

masks for employees 029 -032 -031 -024 -022 041 036 039 032 021 100state of emergency 084 -075 -085 -082 -045 092 063 081 079 060 037 100

Each off-diagonal entry reports a correlation coefficient of a pair of policy and behavior variables

where Bjit represents behavior variable j in state i at tminus 14 Pit collects the Covid related

policies in state i at t minus 14 Confounders Wit include state-level covariates the growthrate of tests the past number of tests and one week lagged dependent variable which maycapture unobserved differences in policies across states

Iit is a set of information variables that affect peoplersquos behaviors at t minus 14 Here andthroughout our empirical analysis we focus on two specifications for information In thefirst we assume that information is captured by a trend and a trend interacted with state-level covariates which is a version of information structure (I1) We think of this trend assummarizing national aggregate information on Covid-19 The second information specifi-cation is based on information structure (I3) where we additionally allow information tovary with each statersquos lagged growth of new cases lag(∆ log ∆C 7) and lagged log newcases lag(log ∆C 14)

Table 3 reports the estimates Across different specifications our results imply that poli-cies have large effects on behavior School closures appear to have the largest effect followedby restaurant closures Somewhat surprisingly stay at home and closure of non essentialbusinesses appear to have modest effects on behavior The effect of masks for employees isalso small However these results should be interpreted with caution Differences betweenpolicy effects are generally statistically insignificant except for masks for employees thepolicies are highly correlated and it is difficult to separate their effects



CA

US

AL

IMP

AC

TO

FM

AS

KS

P

OL

ICIE

S

BE

HA

VIO

R17

Table 3 The Effect of Policies and Information on Behavior (PIrarrB)

Dependent variable

Trend Information Lag Cases Informationworkplaces retail grocery transit workplaces retail grocery transit

(1) (2) (3) (4) (5) (6) (7) (8)

masks for employees 1149 minus0613 minus0037 0920 0334 minus1038 minus0953 1243(1100) (1904) (1442) (2695) (1053) (1786) (1357) (2539)

closed K-12 schools minus14443lowastlowastlowast minus16428lowastlowastlowast minus14201lowastlowastlowast minus16937lowastlowastlowast minus11595lowastlowastlowast minus13650lowastlowastlowast minus12389lowastlowastlowast minus15622lowastlowastlowast

(3748) (5323) (3314) (5843) (3241) (4742) (3190) (6046)stay at home minus3989lowastlowastlowast minus5894lowastlowastlowast minus7558lowastlowastlowast minus10513lowastlowastlowast minus3725lowastlowastlowast minus5542lowastlowastlowast minus7120lowastlowastlowast minus9931lowastlowastlowast

(1282) (1448) (1710) (2957) (1302) (1399) (1680) (2970)closed movie theaters minus2607lowast minus6240lowastlowastlowast minus3260lowastlowast 0600 minus1648 minus5224lowastlowastlowast minus2721lowast 1208

(1460) (1581) (1515) (2906) (1369) (1466) (1436) (2845)closed restaurants minus2621lowast minus4746lowastlowastlowast minus2487lowastlowast minus6277lowast minus2206lowast minus4325lowastlowastlowast minus2168lowastlowast minus5996lowast

(1389) (1612) (1090) (3278) (1171) (1385) (0957) (3171)closed businesses minus4469lowastlowastlowast minus4299lowastlowastlowast minus4358lowastlowastlowast minus3593 minus2512lowast minus2217 minus2570lowast minus1804

(1284) (1428) (1352) (2298) (1297) (1577) (1415) (2558)∆ log Tst 1435lowastlowastlowast 1420lowastlowastlowast 1443lowastlowastlowast 1547lowastlowastlowast 1060lowastlowastlowast 1193lowastlowastlowast 0817lowastlowast 1451lowastlowastlowast

(0230) (0340) (0399) (0422) (0203) (0322) (0370) (0434)lag(tests per 1000 7) 0088 0042 0025 minus0703lowast 0497lowastlowastlowast 0563lowast 0242 minus0217

(0178) (0367) (0202) (0371) (0176) (0292) (0198) (0333)log(days since 2020-01-15) minus7589 38828 2777 27182 minus14697 26449 6197 15213

(25758) (32010) (27302) (26552) (20427) (30218) (25691) (27375)lag(∆ log ∆Cst 7) minus1949lowastlowastlowast minus2024lowast minus0315 minus0526

(0527) (1035) (0691) (1481)lag(log ∆Cst 14) minus2398lowastlowastlowast minus2390lowastlowastlowast minus1898lowastlowastlowast minus1532

(0416) (0786) (0596) (1127)∆D∆C Tst minus1362 minus4006 1046 minus6424lowast

(2074) (3061) (1685) (3793)

log t times state variables Yes Yes Yes Yes Yes Yes Yes Yesstate variables Yes Yes Yes Yes Yes Yes Yes Yessum

j Policyj -2698 -3822 -3190 -3580 -2135 -3200 -2792 -3090

(391) (558) (404) (592) (360) (542) (437) (677)Observations 3009 3009 3009 3009 3009 3009 3009 3009R2 0750 0678 0676 0717 0798 0710 0704 0728Adjusted R2 0749 0676 0674 0715 0796 0708 0702 0726

Yeste lowastplt01 lowastlowastplt005 lowastlowastlowastplt001

Dependent variables are ldquoTransit Intensityrdquo ldquoWorkplace Intensityrdquo ldquoRetail Intensityrdquo and ldquoGrocery Intensityrdquo defined as 7 days moving averages of corresponding dailymeasures obtained from Google Mobility Reports Columns (1)-(4) use specifications with the information structure (I1) while columns (5)-(8) use those with theinformation structure (I3) All specifications include state-level characteristics (population area unemployment rate poverty rate and a percentage of people subject toillness) as well as their interactions with the log of days since Jan 15 2020 The row ldquo

sumj Policyjrdquo reports the sum of six policy coefficients

All rights reserved N

o reuse allowed w

ithout permission

(which w

as not certified by peer review) is the authorfunder w

ho has granted medR

xiv a license to display the preprint in perpetuity T

he copyright holder for this preprintthis version posted M

ay 30 2020

httpsdoiorg1011012020052720115139doi

medR

xiv preprint


In columns (1)-(4) the estimated sum effect of all policies are substantial and can ac-count for a large fraction of observed declines in behavior variables For example retailintensity was approximately -40 at its lowest point in mid April and the estimated policycoefficients imply that imposing all six policies would lead to 38240 asymp 95 of the observeddecline

Columns (5)-(8) reports the estimated coefficients for the model with lagged dependentvariables Except for transit the estimated coefficients of lagged growth of new cases andlagged new cases are negative and often significant at conventional levels This suggests thatboth the growth and the number of past cases reduce social interactions perhaps becausepeople are increasingly aware of prevalence of Covid-19 (Maloney and Taskin 2020)

The estimated aggregate effects of policies in columns (5)-(8) are largely consistent withthose reported in columns (1)-(4) The policy coefficients in columns (5)-(8) are generally15-20 smaller in magnitude than in columns (1)-(4) It may be that the simple trend incolumns (1)-(4) does not capture information as well as the lagged growth and cases usedas information variables in columns (5)-(8)

43 The Direct Effect of Policies and Behavior on Case Growth We now ana-lyze how behavior and policies together influence case growth rates by estimating equation(BPIrarrY)

Yit = αprimeBit + πprimePit + microprimeIit + δprimeYWit

δprimeXWXit+δT∆ log(T )it+δDTit∆Dit

∆Cit

+εyit (12)

where Bit = (B1it B

4it)prime is a vector of four behavior variables in state i at t minus 14 where

a lag length of 14 days is chosen to reflect the delay between infection and confirmation15

Pit includes the Covid related policies in state i at tminus 14 that directly affect the spread ofCovid-19 after controlling for behavior variables (eg masks for employees) We includeinformation variables Iit as covariates because some information variables such as the pastconfirmed cases may be correlated with (latent) government policies or peoplersquos behaviorsthat are not fully captured by our observed policy and behavior variables WXit consistsof state-level covariates As discussed in section 23 equation (12) corresponds to (M)

derived from the SIR model and the terms δT∆ log(T )it and δDTit∆Dit

∆Citcapture the effect

of changing test rates on confirmed cases

Table 4 shows the results of estimating (12) In columns (1) and (3) we use ldquostate ofemergencyrdquo policy as a proxy for all other policies because a ldquostate of emergencyrdquo ordercan be viewed as preparation for implementing a variety of policies subsequently The re-sults indicate that mandatory face masks reduce infections while holding behavior constantsuggesting that requiring masks may be an effective preventive measure Except for maskrequirements policies appear to have little direct effect on confirmed case growth whenbehavior is held constant This supports a directed causal model where causality flows frompolicies to behavior to infections

15As we review in the Appendix A6 a lag length of 14 days is broadly consistent with currently availableevidence




Table 4 The Direct Effect of Behavior and Policies on Case Growth (PBIrarrY )

Dependent variable

Trend Information Lag Cases Information

(1) (2) (3) (4)

lag(state of emergency 14) minus0502lowastlowastlowast minus0342lowastlowast

(0141) (0137)

lag(masks for employees 14) minus0072 minus0039 minus0104lowastlowast minus0092lowast

(0050) (0055) (0045) (0048)

lag(closed K-12 schools 14) minus0132 minus0130

(0116) (0115)lag(stay at home 14) minus0028 minus0054

(0068) (0077)

lag(closed movie theaters 14) 0144lowastlowast 0132lowastlowast

(0056) (0061)

lag(closed restaurants 14) minus0035 minus0077

(0060) (0079)lag(closed businesses 14) minus0022 0008

(0049) (0057)

lag(workplaces 14) 0011lowast 0010 0004 0001(0006) (0007) (0008) (0009)

lag(retail 14) 0006 0014lowastlowastlowast 0002 0006

(0004) (0005) (0005) (0006)lag(grocery 14) 0002 minus0007lowastlowast 0007lowastlowast 0002

(0003) (0003) (0003) (0003)lag(transit 14) 0002 0002 0004 0005

(0003) (0003) (0004) (0004)

∆ log Tst 0106lowastlowastlowast 0115lowastlowastlowast 0120lowastlowastlowast 0121lowastlowastlowast

(0020) (0022) (0019) (0019)

lag(tests per 1000 7) minus0007 minus0004 0025lowastlowastlowast 0029lowastlowastlowast

(0006) (0006) (0009) (0009)log(days since 2020-01-15) 4691lowastlowastlowast 3779lowastlowast 3170lowastlowast 2519lowast

(1562) (1504) (1268) (1379)

lag(∆ log ∆Cst 7) minus0139lowastlowastlowast minus0140lowastlowastlowast

(0049) (0049)

lag(log ∆Cst 14) minus0132lowastlowastlowast minus0149lowastlowastlowast

(0043) (0042)∆D∆C

Tst minus0267lowastlowastlowast minus0247lowastlowastlowast

(0103) (0094)

log t times state variables Yes Yes Yes Yesstate variables Yes Yes Yes Yessum

policies except masks -007 -012

(017) (018)sumwkbehaviork -105 -054

(015) (019)Observations 3403 3403 3403 3403R2 0771 0765 0793 0791Adjusted R2 0769 0763 0791 0790


Dependent variable is the weekly growth rate of confirmed cases as defined in equation (3) The covariates include

14 days lagged policy and behavior variables which are constructed as 7 days moving averages between tminus 14 totminus 21 of corresponding daily measures Columns (1)-(2) use specifications with the information structure (I1) while

columns (3)-(4) use those with the information structure (I3) All specifications include state-level characteristics

(population area unemployment rate poverty rate and a percentage of people subject to illness) and theirinteractions with the log of days since Jan 15 2020 The row ldquo

sumpolicies except for masksrdquo reports the sum of five

policy coefficients except for ldquomasks for employeesrdquo while the row ldquosumk wkbehaviorkrdquo reports the sum of four

coefficients of behavior variables weighted by the average of each behavioral variable from April 1st-10th




Table 5 Immediate Effects


masks -004 -009(006) (005)sum

policies except masks -007 -012(017) (018)sum

wkbehaviork -105 -054

(015) (019)

The reported numbers are from columns (2) and (4) of Table 4 The row ldquomasksrdquo reports the estimated coefficientsof ldquomasks for employeesrdquo on weekly case growth The row ldquo




A useful practical implication of these results are that Google Mobility Reports andsimilar data might be useful as a leading indicator of potential case growth This shouldbe done cautiously however because other changes in the environment might alter therelationship between behavior and infections Preventative measures including mandatoryface masks and changes in habit that are not captured in our data might alter the futurerelationship between Google Mobility Reports and case growth

Table 5 reports the sum of short run policy and behavior effects Policy effects are simplythe sum of coefficients The behavior effect is the sum of coefficients weighted by the averageof the behavioral variables from April 1st-10th

In the trend information model the short run and long run effects coincide abstractingfrom any feedback mechanism On the other hand if policies are enacted and behaviorchanges then future cases and information will change The lag cases information modelbased on information structure (I3) can capture such feedback effects However since thelag cases information model includes lags of both case growth and log cases computinga long-run effect is not completely straightforward We investigate dynamic effects thatincorporate feedback mechanism through a change in information in section 5

In this section we focus our analysis of policy effects when we hold information constantThe estimates of the effect of policy on behavior in table 4 and of the effect of policies andbehavior and case growth in table 3 can be combined to calculate the total effect of policyas well as its decomposition into the direct and the indirect effects

Table 7 show the direct (holding behavior constant) and indirect (through behaviorchanges) effects of policy on case growth for the lag cases information model where thestandard errors are computed by bootstrap and clustered on state The estimates implythat all policies combined would reduce the growth rate of cases by 063 out of which 021 isattributable to direct effect while 042 is attributable to indirect effect through their impacton behavior although we need to interpret these numbers carefully given that informationis held constant The estimate also indicates that the effect of ldquomasks for employeesrdquo policyis mostly direct




Table 6 The Total Effect of Policies on Case Growth (PIrarrY )

Dependent variable


(1) (2) (3) (4)

lag(state of emergency 14) minus0998lowastlowastlowast minus0395lowastlowastlowast

(0145) (0150)

lag(masks for employees 14) minus0076 minus0118lowast minus0065 minus0174lowastlowast

(0095) (0067) (0063) (0071)

lag(closed K-12 schools 14) minus0842lowastlowastlowast minus0516lowastlowastlowast

(0141) (0123)lag(stay at home 14) minus0164lowastlowast minus0196lowastlowast

(0067) (0076)

lag(closed movie theaters 14) 0042 0060(0074) (0071)

lag(closed restaurants 14) minus0190lowastlowast minus0155lowastlowast

(0082) (0077)lag(closed businesses 14) minus0090lowast minus0028

(0049) (0058)


(0029) (0024) (0023) (0022)


(0011) (0007) (0009) (0008)log(days since 2020-01-15) 3975lowast 3277 1973 2020

(2283) (2149) (1594) (1652)lag(∆ log ∆Cst 7) minus0012 minus0085lowast

(0061) (0048)


(0042) (0036)∆D∆C


(0112) (0087)

log t times state variables Yes Yes Yes Yesstate variables Yes Yes Yes Yessumj Policyj -136 -101

(018) (020)

Observations 3433 3433 3427 3427R2 0666 0707 0730 0757

Adjusted R2 0665 0705 0729 0755


Dependent variable is the weekly growth rate of confirmed cases as defined in equation (3) The covariates include14 days lagged policy variables which are constructed as 7 days moving averages between tminus 14 to tminus 21 ofcorresponding daily measures Columns (1)-(2) use specifications with the information structure (I1) while columns(3)-(4) use those with the information structure (I3) All specifications include state-level characteristics

(population area unemployment rate poverty rate and a percentage of people subject to illness) and theirinteractions with the log of days The row ldquo


44 The Total Effect of Policies on Case Growth We can also examine the totaleffect of policies and information on case growth by estimating (PIrarrY) The coefficientson policy in this regression combine both the direct and indirect effects As show in thefinal two columns of table 7 the policy effects from estimating (PIrarrY) are about twice aslarge as the policy effect from combining (BPIrarrY) and (PIrarrB)




Table 7 Direct and Indirect Policy Effects Lag Cases Information Model

Direct Via-behavior Total PIrarrY Coefficient Difference

masks for employees -009 -000 -009 -017 008(006) (003) (006) (009) (007)

closed K-12 schools -013 -019 -032 -052 019(013) (009) (014) (014) (007)

stay at home -005 -010 -015 -020 004

(009) (005) (009) (009) (005)closed movie theaters 013 -003 010 006 004

(007) (004) (008) (008) (004)closed restaurants -008 -006 -014 -015 002

(010) (004) (009) (009) (004)

closed businesses 001 -003 -002 -003 001(007) (003) (007) (007) (002)

all -021 -042 -063 -101 038(022) (018) (020) (023) (018)

Direct effects capture the effect of policy on case growth holding behavior information and confounders constant

Direct effects are given by π in equation (BPIrarrY) Indirect effects capture how policy changes behavior and

behavior shift case growth They are given by α from (BPIrarrY) times β from (PIrarrB) The total effect is π + βαColumn ldquoPIrarrY Coefficientsrdquo shows the coefficient estimates from PIrarrY Column ldquoDifferencerdquo are the differences

between the estimates from (PIrarrY) and the combination of (BPIrarrY) and (PIrarrB) Standard errors are computed

by bootstrap and clustered on state

Table 6 shows the full set of coefficient estimates for (PIrarrY) The results are broadlyconsistent with what we found above As in tables 3 and 7 school closures appear to havethe largest effect on growth rates followed by restaurant closures However as mentionedpreviously the policies are highly correlated so it is difficult to separately identify theirimpacts except mask mandates The coefficient of mandatory face masks is estimated at-0174 and significant at 5 percent level

As reported in column ldquoPIrarrY Coefficientrdquo of Table 7 the estimates imply that all policiescombined would reduce ∆ log ∆C by 101 for the lag information model For comparisonthe daily average of ∆ log ∆C reached its peak in mid-March of about three Since then ithas declined to near 0 This peak of 3 is arguably somewhat overstated because there wasalso a rapid increase in testing at that time If we adjust ∆ log ∆C for increasing in testingby subtracting 0166∆ log T its peak is around 2 Thus the estimate of -101 roughly implythat policy changes can account for one half of the observed decrease in case growth Theremainder of the decline is likely due to changes in behavior from information




5 Empirical Evaluation of Counterfactual Policies

We now turn our focus to the specification with information as lagged dependent infor-mation This allows dynamic feedback effects Policy and behavior changes that reducecase growth today can lead to a more optimistic outlook in the future attenuating longerrun effects To simplify and focus on dynamics in this section we work with the estimatesof the total policy effects (equation PIrarrY and table 6) rather than decomposing into directand via-behavior effects

51 Removing Policies We first consider the impact of changing from the observedpolicies to none For illustrative purpose we begin by focusing on Washington Figure6 shows the observed estimated average and counterfactual without policies average of∆ log ∆C in Washington To compute the estimated and counterfactual paths we use theestimates from column 2 of table 6 We set initial ∆ log ∆C and log ∆C to their valuesfirst observed in the state we are simulating We hold all other regressors at their observedvalues Error terms are drawn with replacement from the residuals We do this many timesand report the average over draws of the residuals To calculate confidence intervals werepeat the above with coefficients drawn randomly from their asymptotic distribution Theshaded region is a point-wise 90 confidence interval We can see that fit of the estimatedand observed ∆ log ∆C is quite good

Figure 6 shows the change in case growth from removing policies Since there is a two weeklag between policy implementation and their effect on recorded cases the counterfactualwithout policy average begins differing from the average with policies two weeks after eachpolicy was implemented The effect of removing policies on case growth peaks at about 075and then slowly declines This decline is due to a feedback effect of information Withoutpolicies cases grow more rapidly With more dire case numbers people adjust their behaviorto reduce infection This somewhat offsets the increase in infection from removing policiesGiven the delay between infection and reported case numbers this offsetting effect takessome time to appear

Even small differences in case growth can lead to very large differences in case num-bers Removing policies in Washington increases case growth from around 0 to above 05throughout most of April Figure 7 shows that this increase in growth implies a tremendousincrease in cases

The effect of removing cases in other states is broadly similar to that in Washington Fig-ures 8 and 9 show the changes in growth rate and cases from removing policy interventionsin Illinois and Massachusetts

Figure 10 shows the average across states of the change in case growth Similar towhat we saw in Washington removing policies leads to an increase of about 06 in casegrowth throughout April This increase in growth implies a very large increase in cases by




Figure 6 Case growth with and without policies in Washington

0

2

4

Mar Apr Maydate

∆log

∆C

Estimated

Observed

Without Policies

∆log∆C

Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C

Effect of removing policies on ∆log∆C

Figure 7 Change in cases from removing policies in WashingtonM

anda

te fa

ce m

ask

use

by e

mpl

oyee

s in

pub

lic fa

cing

bus

ines

ses

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

0

500000

1000000

1500000

2000000

2500000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek

Effect of removing policies on cases

late May16 Figure 11 displays the national increase in aggregate cases without any policyintervention The estimates imply 10 to 100 million additional cases by late-May or a30-200 fold increase

16This estimate should be interpreted somewhat cautiously because it involves some extrapolation Theinformation variables in particular logC soon reach levels not seen in the data




Figure 8 Effect of removing policies in Illinois

Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

0e+00

2e+06

4e+06

6e+06

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek


Figure 9 Effect of removing policies in Massachusetts

Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

0e+00

1e+06

2e+06

3e+06

4e+06

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek





Figure 10 Average change in case growth from removing policies

00

02

04

06

08

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Figure 11 National change in cases from removing policies

00e+00

25e+07

50e+07

75e+07

10e+08

Mar Apr Maydate

case

s in

the

past

wee

k


0

50

100

150

200

Mar Apr Maydate

rela

tive

incr

ease

in c

ases





52 Non essential Business Closures A particularly controversial policy is the closureof non essential businesses We now examine a counterfactual where non essential businessesare never closed Figure 12 shows the effect of leaving non essential businesses open inWashington Non essential businesses have a modest impact on cases with an averageincrease of 1000 by late May with a 90 confidence interval from roughly -1000 to 4000

Figure 12 Effect of leaving non essential businesses open in Washington

Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus2000

0

2000

4000

6000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek


Figure 13 Effect of leaving businesses open in Illinois

Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C

Effect of removing policies on ∆log∆CM

anda

te fa

ce m

ask

use

by e

mpl

oyee

s in

pub

lic fa

cing

bus

ines

ses

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus5000

0

5000

10000

15000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek


Figure 15 shows the national aggregate effect of leaving non essential businesses openThe estimates imply that with non essential businesses open cases would be about -10 to50 higher in late May




Figure 14 Effect of leaving businesses open in Massachusetts

Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

0

4000

8000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek


Figure 15 National change in cases from leaving non essential businesses open

minus002

000

002

004

006

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


minus01

00

01

02

03

04

05

Mar Apr Maydate

rela

tive

incr

ease

in c

ases


53 Mask Mandate As discussed earlier we find that mask mandates reduce case growtheven holding behavior constant In other words mask mandates may reduce infections withrelatively little economic disruption This makes mask mandates a particularly attractivepolicy instrument In this section we examine what would have happened to cases if allstates had imposed a mask mandate on April 1st Figure 16 shows the changes in casegrowth and cases in Washington In Figures 17 and 18 implementing mandatory maskpolicy on April 1st would have reduced the total number of confirmed cases by 2500-12500in Illinois and 2000-11000 in Massachusetts by late-May Figure 19 shows the national




Figure 16 Effect of mandating masks on April 1st in Washington

Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus03

minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus5000

minus4000

minus3000

minus2000

minus1000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek


Figure 17 Effect of mandating masks on April 1st in Illinois

Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus03

minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus20000

minus15000

minus10000

minus5000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek





Figure 18 Effect of mandating masks on April 1st in Massachusetts

Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus02

minus01

00

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus10000

minus5000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek


Figure 19 National change in cases from mandating masks on April 1st

minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


minus06

minus04

minus02

00

Mar Apr Maydate

rela

tive

incr

ease

in c

ases





change in case growth and cases Mandating masks on April 1st leads to 30 to 57 fewercases nationally in late May

6 Conclusion

This paper assesses the effects of policies on the spread of Covid-19 in the US using theUS state-level data on cases tests policies and social distancing behavior measures fromGoogle Mobility Reports Our findings are summarized as follows

First our empirical analysis indicates that mandating face masks have reduced the spreadof Covid-19 without affecting peoplersquos social distancing behavior measured by Google Mo-bility Reports Our counterfactual experiment based on the estimated model indicates thatif all states had have adopted mandatory face mask policy on April 1st of 2020 then thenumber of cases in late-May would have been smaller by 30 to 57 percent potentially savingtens of thousands of lives

Second we find that the impact of all policies combined on case growth is quite largereducing the growth rate of cases by 60 percent Except for mandatory face mask policythe effect of all other policies on case growth is realized through their impact on socialdistancing behavior Our counterfactual experiment shows that had all policies have beenremoved from all states there would have been 10 to 100 million additional cases by late-May suggesting that policy makers should be careful about removing all policies together

Third we find that people voluntarily reduce their visits to workplace transit groceryand transits when they receive information on a higher number of new cases This suggeststhat individuals make decisions to voluntarily limit their contact with others in responseto greater transmission risks leading to an important feedback mechanism to future casesAs our counterfactual experiments illustrate model simulations that ignore this peoplersquosvoluntary behavioral response to information on transmission risks would over-predict thefuture number of cases

Beyond these findings our paper presents a useful conceptual framework to investigatethe relative roles of policies and information on determining the spread of Covid-19 throughtheir impact on peoplersquos behavior Our causal model allows us to explicitly define coun-terfactual scenarios to properly evaluate the effect of alternative policies on the spread ofCovid-19 and can be used to quantitatively examine various Covid-19 related issues forfuture research

References

Abouk Rahi and Babak Heydari 2020 ldquoThe Immediate Effect of COVID-19 Policies on Social DistancingBehavior in the United Statesrdquo medRxiv URL httpswwwmedrxivorgcontentearly20200428

2020040720057356Andersen Martin 2020 ldquoEarly Evidence on Social Distancing in Response to COVID-19 in the United

Statesrdquo Tech rep UNC Greensboro




Avery Christopher William Bossert Adam Clark Glenn Ellison and Sara Fisher Ellison 2020 ldquoPolicyImplications of Models of the Spread of Coronavirus Perspectives and Opportunities for EconomistsrdquoNBER Working Papers 27007 National Bureau of Economic Research Inc URL httpsideasrepec

orgpnbrnberwo27007htmlCereda D M Tirani F Rovida V Demicheli M Ajelli P Poletti F Trentini G Guzzetta V Marziano

A Barone M Magoni S Deandrea G Diurno M Lombardo M Faccini A Pan R Bruno E ParianiG Grasselli A Piatti M Gramegna F Baldanti A Melegaro and S Merler 2020 ldquoThe early phase ofthe COVID-19 outbreak in Lombardy Italyrdquo

Courtemanche Charles Joseph Garuccio Anh Le Joshua Pinkston and Aaron Yelowitz 2020 ldquoStrongSocial Distancing Measures In The United States Reduced The COVID-19 Growth Raterdquo Health Affairs101377hlthaff202000608URL httpsdoiorg101377hlthaff202000608

Ferguson Neil Daniel Laydon Gemma Nedjati-Gilani Natsuko Imai Kylie Ainslie Marc BaguelinSangeeta Bhatia Adhiratha Boonyasiri Zulma Cucunuba Gina Cuomo-Dannenburg Amy Dighe IlariaDorigatti Han Fu Katy Gaythorpe Will Green Arran Hamlet Wes Hinsley Lucy C Okell Sabine vanElsland Hayley Thompson Robert Verity Erik Volz Haowei Wang Yuanrong Wang Patrick GT WalkerCaroline Walters Peter Winskill Charles Whittaker Christl A Donnelly Steven Riley and Azra C Ghani2020 ldquoReport 9 Impact of non-pharmaceutical interventions (NPIs) to reduce COVID-19 mortality andhealthcare demandrdquo Tech rep Imperial College London

Greenhalgh Trisha Manuel B Schmid Thomas Czypionka Dirk Bassler and Laurence Gruer 2020 ldquoFacemasks for the public during the covid-19 crisisrdquo BMJ 369 URL httpswwwbmjcomcontent369

bmjm1435Greenland Sander Judea Pearl and James M Robins 1999 ldquoCausal diagrams for epidemiologic researchrdquo

Epidemiology 37ndash48Gupta Sumedha Thuy D Nguyen Felipe Lozano Rojas Shyam Raman Byungkyu Lee Ana Bento Kosali I

Simon and Coady Wing 2020 ldquoTracking Public and Private Responses to the COVID-19 EpidemicEvidence from State and Local Government Actionsrdquo Working Paper 27027 National Bureau of EconomicResearch URL httpwwwnberorgpapersw27027

Haavelmo Trygve 1944 ldquoThe probability approach in econometricsrdquo Econometrica Journal of the Econo-metric Society iiindash115

Heckman James J and Edward J Vytlacil 2007 ldquoEconometric Evaluation of Social Programs Part ICausal Models Structural Models and Econometric Policy Evaluationrdquo In Handbook of EconometricsHandbook of Econometrics vol 6 edited by JJ Heckman and EE Leamer chap 70 Elsevier URLhttpsideasrepecorgheeeecochp6b-70html

Howard Jeremy Austin Huang Zhiyuan Li Zeynep Tufekci Vladimir Zdimal Helene-Mari van derWesthuizen Arne von Delft Amy Price Lex Fridman Lei-Han Tang Viola Tang Gregory Wat-son Christina Bax Reshama Shaikh Frederik Questier Danny Hernandez Larry Chu ChristinaRamirez and Anne Rimoin 2020 ldquoFace Masks Against COVID-19 An Evidence Reviewrdquo URLhttpsdoiorg1020944preprints2020040203v1

Hsiang Solomon Daniel Allen Sebastien Annan-Phan Kendon Bell Ian Bolliger Trinetta Chong HannahDruckenmiller Andrew Hultgren Luna Yue Huang Emma Krasovich Peiley Lau Jaecheol Lee EstherRolf Jeanette Tseng and Tiffany Wu 2020 ldquoThe Effect of Large-Scale Anti-Contagion Policies on theCoronavirus (COVID-19) Pandemicrdquo medRxiv URL httpswwwmedrxivorgcontentearly2020

04292020032220040642Imbens Guido W and Donald B Rubin 2015 Causal Inference for Statistics Social and Biomedical

Sciences An Introduction Cambridge University PressLi Qun Xuhua Guan Peng Wu Xiaoye Wang Lei Zhou Yeqing Tong Ruiqi Ren Kathy SM Leung

Eric HY Lau Jessica Y Wong Xuesen Xing Nijuan Xiang Yang Wu Chao Li Qi Chen Dan Li TianLiu Jing Zhao Man Liu Wenxiao Tu Chuding Chen Lianmei Jin Rui Yang Qi Wang Suhua ZhouRui Wang Hui Liu Yinbo Luo Yuan Liu Ge Shao Huan Li Zhongfa Tao Yang Yang Zhiqiang DengBoxi Liu Zhitao Ma Yanping Zhang Guoqing Shi Tommy TY Lam Joseph T Wu George F GaoBenjamin J Cowling Bo Yang Gabriel M Leung and Zijian Feng 2020 ldquoEarly Transmission Dynam-ics in Wuhan China of Novel CoronavirusndashInfected Pneumoniardquo New England Journal of Medicine382 (13)1199ndash1207 URL httpsdoiorg101056NEJMoa2001316 PMID 31995857




LLC Google 2020 ldquordquoGoogle COVID-19 Community Mobility Reportsrdquordquo URL httpswwwgooglecom

covid19mobilityMaloney William F and Temel Taskin 2020 ldquoDeterminants of Social Distanc-

ing and Economic Activity during COVID-19 A Global Viewrdquo Tech Rep Pol-icy Research working paper no WPS 9242 COVID-19 (Coronavirus) World BankGroup URL httpdocumentsworldbankorgcurateden325021589288466494

Determinants-of-Social-Distancing-and-Economic-Activity-during-COVID-19-A-Global-ViewMIDAS 2020 ldquoMIDAS 2019 Novel Coronavirus Repository Parameter Estimatesrdquo URL httpsgithub

commidas-networkCOVID-19treemasterparameter_estimates2019_novel_coronavirusNeyman 1925 ldquoOn the application of probability theory to agricultural experiments Essay on principles

Section 9rdquo Statistical Science (1990) 465ndash472Pei Sen Sasikiran Kandula and Jeffrey Shaman 2020 ldquoDifferential Effects of Intervention Timing on

COVID-19 Spread in the United Statesrdquo medRxiv URL httpswwwmedrxivorgcontentearly

202005202020051520103655Raifman Julia Kristen Nocka David Jones Jacob Bor Sarah Ketchen Lipson Jonathan Jay Philip Chan

Megan Cole Brahim Carolyn Hoffman Claire Corkish Elizabeth Ferrara Elizabeth Long Emily BaroniFaith Contador Hannah Simon Morgan Simko Rachel Scheckman Sarah Brewer Sue Kulkarni FeliciaHeykoop Manish Patel Aishwarya Vidyasagaran Andrew Chiao Cara Safon and Samantha Burkhart2020 ldquoCOVID-19 US state policy databaserdquo URL httpstinyurlcomstatepolicies

Rubin Donald B 1974 ldquoEstimating causal effects of treatments in randomized and nonrandomized studiesrdquoJournal of educational Psychology 66 (5)688

Siordia Juan A 2020 ldquoEpidemiology and clinical features of COVID-19 A review of current literaturerdquoJournal of Clinical Virology 127104357 URL httpwwwsciencedirectcomsciencearticlepii

S1386653220300998Stock James 2020a ldquoReopening the Coronavirus-Closed Economyrdquo Tech rep Hutchins Center Working

Paper 60Stock James H 2020b ldquoData Gaps and the Policy Response to the Novel Coronavirusrdquo Working Paper

26902 National Bureau of Economic Research URL httpwwwnberorgpapersw26902Tinbergen J 1930 ldquoDetermination and interpretation of supply curves an examplerdquo Tech rep Zeitschrift

fur NationalokonomieWhite House The 2020 ldquoGuidelines for Opening Up America Againrdquo URL httpswwwwhitehouse

govopeningamericaWright Philip G 1928 Tariff on animal and vegetable oils Macmillan Company New YorkZhang Juanjuan Maria Litvinova Wei Wang Yan Wang Xiaowei Deng Xinghui Chen Mei Li Wen

Zheng Lan Yi Xinhua Chen Qianhui Wu Yuxia Liang Xiling Wang Juan Yang Kaiyuan Sun Ira MLongini M Elizabeth Halloran Peng Wu Benjamin J Cowling Stefano Merler Cecile Viboud AlessandroVespignani Marco Ajelli and Hongjie Yu 2020 ldquoEvolving epidemiology and transmission dynamics ofcoronavirus disease 2019 outside Hubei province China a descriptive and modelling studyrdquo The LancetInfectious Diseases URL httpwwwsciencedirectcomsciencearticlepiiS1473309920302309




Appendix A Data Construction

A1 Measuring ∆C and ∆ log ∆C We have three data sets with information on dailycumulative confirmed cases in each state As shown in table 8 these cumulative casenumbers are very highly correlated However table 9 shows that the numbers are differentmore often than not

NYT JHU CTPNYT 100000 099995 099990JHU 099995 100000 099986CTP 099990 099986 100000

Table 8 Correlation of cumulative cases

1 2 3NYT 100 029 038JHU 029 100 033CTP 038 033 100

Table 9 Portion of cumulative cases that are equal between data sets

Figure 20 shows the evolution of new cases in each of these three datasets In all casesdaily changes in cumulative cases displays some excessive volatility This is likely due todelays and bunching in testing and reporting of results Table 10 shows the variance of lognew cases in each data set as well as their correlations As shown the correlations areapproximately 09 Also of note log of The New York Times new case numbers has thelowest variance17 In our subsequent results we will primarily use the case numbers fromThe New York Times


Variance 584 717 678Table 10 Correlation and variance of log daily new cases

For most of our results we focus on new cases in a week instead of in a day We do thisfor two reasons as discussed in the main text First a decline of new cases over two weekshas become a key metric for decision makers Secondly aggregating to weekly new casessmooths out the noise associated with the timing of reporting and testing

Table 11 reports the correlation and variance of weekly log new cases across the threedata sets Figure 21 shows the evolution of weekly new cases in each state over time

17This comparison is somewhat sensitive to how you handle negative and zero cases when taking logsHere we replaced log(0) with log(01) In our main results we work with weekly new cases which are veryrarely zero




Figure 20 Daily cases

1

10

100

1000

10000

Apr May Jundate

New

cas

es

NYT

1

10

100

1000

10000

Apr May Jundate

New

cas

es

JHU

1

10

100

1000

10000

Apr May Jundate

New

cas

es

CTP

Each line shows daily new cases in a state


Variance 426 446 431Table 11 Correlation and variance of log weekly new cases

A2 Deaths Table 12 reports the correlation and variance of weekly deaths in the threedata sets Figure 22 shows the evolution of weekly deaths in each state As with cases weuse death data from The New York Times in our main results

NYT JHU CTPNYT 100JHU 100CTP 100

Variance 33561976 33031042 23230330Table 12 Correlation and variance of weekly deaths

A3 Tests Our test data comes from The Covid Tracking Project Figure 23 show theevolution of tests over time

A4 Social Distancing Measures In measuring social distancing we focus on GoogleMobility Reports This data has international coverage and is publicly available We also




Figure 21 Weekly Cases

1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

NYT

1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

JHU

1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

CTP

Each line shows weekly new cases in a state

Figure 22 Weekly Deaths

1

10

100

1000

10000

Apr May Jundate

Dea

ths

NYT

1

10

100

1000

10000

Apr May Jundate

Dea

ths

JHU

1

10

100

1000

Apr May Jundate

Dea

ths

CTP

Each line shows weekly deaths in a state

compare Google Mobility Reports to other social distancing measures from SafeGraph Un-acast and PlaceIQ Each of these other data sources are limited to the United StatesSafeGraph and Unacast data is not publicly available (both companies make the data avail-able to researchers at no cost)




Figure 23 Number of Tests

1e+01

1e+03

1e+05

Apr May Jundate

Test

s

Total Cumulative Tests

1e+01

1e+03

1e+05

Apr May Jundate

Test

s

New Tests in the Past Week

These figures use the ldquototal test resultsrdquo reported by The Covid Tracking Project This is meant to reflectthe number of people tested (as opposed to the number of specimens tested)

A5 Policy Variables We use the database on US state policies created by Raifmanet al (2020) As discussed in the main text our analysis focuses on seven policies Forstay at home closed nonessential businesses closed K-12 schools closed restaurants excepttakeout and close movie theaters we double-checked any state for which Raifman et al(2020) does not record a date We filled in a few missing dates Our modified data isavailable here Our modifications fill in 1 value for school closures 2 for stay at home 3for movie theater closure and 4 for non-essential business closures Table 13 displays all 25dated policy variables in Raifman et al (2020)rsquos database with our modifications describedabove

A6 Timing There is a delay between infection and when a person is tested and appearsin our case data MIDAS (2020) maintain a list of estimates of the duration of variousstages of Covid-19 infections The incubation period the time from infection to symptomonset is widely believed to be 5 days For example using data from Wuhan Li et al (2020)estimate a mean incubation period of 52 days Siordia (2020) reviews the literature andconcludes the mean incubation period is 3-9 days

Estimates of the time between symptom onset and case reporting are less common UsingItalian data Cereda et al (2020) estimate an average of 73 days between symptom onsetand reporting Zhang et al (2020) find an average of 74 days using Chinese data fromDecember to early February but they find this period declined from 89 days in January to54 days in the first week of February Both of these papers on time from symptom onsetto reporting have large confidence intervals covering approximately 1 to 20 days




Based on the above we expect a delay of roughly two weeks between changes in behavioror policies and changes in reported cases




Date closed K 12 schools 51 2020-03-13 2020-03-17 2020-04-03Closed day cares 15 2020-03-16 2020-03-23 2020-04-06

Date banned visitors to nursing homes 30 2020-03-09 2020-03-16 2020-04-06Stay at home shelter in place 42 2020-03-19 2020-03-28 2020-04-07

End relax stay at home shelter in place 22 2020-04-26 2020-05-06 2020-05-22Closed non essential businesses 43 2020-03-19 2020-03-25 2020-04-06

Began to reopen businesses statewide 43 2020-04-20 2020-05-04 2020-05-20Mandate face mask use by all individuals in public spaces 16 2020-04-08 2020-04-19 2020-05-16

Mandate face mask use by employees in public facing businesses 36 2020-04-03 2020-05-01 2020-05-11Closed restaurants except take out 48 2020-03-15 2020-03-17 2020-04-03

Reopen restaurants 30 2020-04-24 2020-05-10 2020-05-22Closed gyms 46 2020-03-16 2020-03-20 2020-04-03

Reopened gyms 21 2020-04-24 2020-05-13 2020-05-20Closed movie theaters 49 2020-03-16 2020-03-21 2020-04-06

Reopened movie theaters 11 2020-04-27 2020-05-08 2020-05-16Stop Initiation of Evictions overall or due to COVID related issues 24 2020-03-16 2020-03-24 2020-04-20

Stop enforcement of evictions overall or due to COVID related issues 26 2020-03-15 2020-03-23 2020-04-20Renter grace period or use of security deposit to pay rent 2 2020-04-10 2020-04-17 2020-04-24

Order freezing utility shut offs 34 2020-03-12 2020-03-19 2020-04-13Froze mortgage payments 1 2020-03-21 2020-03-21 2020-03-21

Modify Medicaid requirements with 1135 waivers date of CMS approval 51 2020-03-16 2020-03-26 2020-04-22Suspended elective medical dental procedures 33 2020-03-16 2020-03-21 2020-04-16

Resumed elective medical procedures 34 2020-04-20 2020-04-29 2020-05-29Waived one week waiting period for unemployment insurance 36 2020-03-08 2020-03-18 2020-04-06



o reuse allowed w

ithout permission

(which w


ho has granted medR



ay 30 2020

httpsdoiorg1011012020052720115139doi

medR

xiv preprint


Department of Economics and Center for Statistics and Data Science MIT MA 02139

Email address vchernmitedu

Vancouver School of Economics UBC 6000 Iona Drive Vancouver BC

Email address hkasaharmailubcca


Email address schrimpfmailubcca



1 Introduction

2 The Causal Model for the Effect of Policies Behavior and Information on Growth of Infection

21 The Causal Model and Its Structural Equation Form

Identification and Parameter Estimation

22 Information Structures and Induced Dynamics

23 Outcome and Key Confounders via SIR Model


31 Total Change Decomposition

32 Counterfactuals


41 Data

42 The Effect of Policies and Information on Behavior

43 The Direct Effect of Policies and Behavior on Case Growth

44 The Total Effect of Policies on Case Growth


51 Removing Policies

52 Non essential Business Closures

53 Mask Mandate

6 Conclusion

References


A1 Measuring C and logC

A2 Deaths

A3 Tests

A4 Social Distancing Measures

A5 Policy Variables

A6 Timing


We present a conceptual framework that spells out the causal structure on how theCovid-19 spread is dynamically determined by policies and human behavior Our approachexplicitly recognizes that policies not only directly affect the spread of Covid-19 (eg maskrequirement) but also indirectly affect its spread by changing peoplersquos behavior (eg stayat home order) It also recognizes that people react to new information on Covid-19 casesand voluntarily adjust their behavior (eg voluntary social distancing and hand washing)even without any policy in place Our casual model provides a framework to quantita-tively decompose the growth of Covid-19 cases into three components (1) direct policyeffect (2) policy effect through behavior and (3) direct behavior effect in response to newinformation2

Guided by the causal model our empirical analysis examines how the weekly growth ratesof confirmed Covid-19 cases are determined by policies and behavior using the US state-level data To examine how policies and information affect peoplersquos behavior we also regresssocial distancing measures on policy and information variables Our regression specificationfor case growth is explicitly driven from a SIR model and the estimated regression coefficientscan be interpreted from the viewpoint of the effective reproduction number although ourcausal approach does not hinge on validity of a SIR model

As policy variables we consider state of emergency mandatory face masks for employ-ees in public business stay at home order closure of school closure of restaurants excepttake out closure of movie theaters and closure of non-essential business Our behaviorvariables are four mobility measures that capture the intensity of visits to ldquotransitrdquo ldquogro-ceryrdquo ldquoretailrdquo and ldquoworkplacesrdquo from Google Mobility Reports We take time trend thelagged growth rate of cases and the log of lagged cases as our measures of information oninfection risks that affects peoplersquos behavior We also consider the growth rate of tests thelog number of tests and the state-level characteristics (eg population size and total area)as the confounders that has to be controlled for in order to identify the causal relationshipbetween policybehavior and the growth rate of cases

Our key findings from regression analysis are as follows We find that both policies andinformation on past cases are important determinants of peoplersquos social distancing behaviorwhere policy effects explain at around 80 or more of observed decline in four behaviorvariables We also find that people dynamically adjust their behavior in response to policiesand information with time lag

There are both large policy effects and large behavior effects on case growth The esti-mates from our lag cases information model imply that all policies combined would reducethe weekly growth rate of cases by around 1 Taking into account of a change in thenumber of tests the weekly growth rates of infections arguably declined from around 2 in

2The causal model is framed using the language of structural equations models and causal diagrams ofeconometrics (Wright (1928) Haavelmo (1944) Heckman and Vytlacil (2007) see Greenland Pearl andRobins (1999) for developments in computer science) with natural unfolding potential outcomes represen-tation (Rubin (1974) Tinbergen (1930) Neyman (1925) Imbens and Rubin (2015)) As such it naturallyconverses in all languages for causal inference
















00

02

04

06

08

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


0

50

100

150

200

Mar Apr Maydate

rela

tive

incr

ease

in c

ases



minus002

000

002

004

006

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


minus01

00

01

02

03

04

05

Mar Apr Maydate

rela

tive

incr

ease

in c

ases








minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


minus06

minus04

minus02

00

Mar Apr Maydate

rela

tive

incr

ease

in c

ases

















other variables







Pit

Iit Yit

Bit

Iit

Wit









(SEM)










Bit =Bit(Pit Iit)

Pit =Pit(Iit)

(O)





which in turn imply








(1)















Iit = log(t)


Iit = Yitminus`


Iit =

(log(t) Yitminus`

infinsumm=1

Yitminus`m

)prime













Ii(tminus7)

Yi(tminus7)

Iit

Yit

Ii(t+7)

Yi(t+7)

SE

M(t-7

)

SE

M(t)

SE

M(t+

7)














∆Cit+ εit (M)



∆Cit



S(t) = minusS(t)

Nβ(t)I(t) (4)

I(t) =S(t)









C(t) = τ(t)I(t) (8)








C(t)

C(t)=S(t)

Nβ(t) +

τ(t)

τ(t)minus 1

π

τ(t)D(t)

C(t) (9)




YitC(t)

C(t)


β(t)


+ δDTit∆Dit

∆Cit

1πτ(t)D(t)

C(t)







= αprimeβprime(

EPit minus EPio


+ πprime(

EPit minus EPio


+ αprimeγprime(

EIit minus EIio

)+ microprime

(EIit minus EIio

)Dynamic Effect


(10)







(I2) DynEt =sumt`


(I3) DynEt =sumt`


+sumt`


+sumt`minus1


)



P itTt=1 i = 1 N



Y it =Yit(B

it P

it I

it)

Bit =Bit(P

it I

it)


lowasti(tminus1) t)

(CEF-SEM)














PY it minus PYit

Predicted CF Change




= PEBit + PED

it + DynEit

(11)



(I2) DynEit =sumt`


i(tminusm`))

(I3) DynEit =sumt`


+sumt`



)+sumt`minus1



)
























resi

den

tial

tran

sit

wor

kp

lace

s

reta

il

groce

ry

clos

edK

-12

sch

ool

s

stay

ath

ome

clos

edm

ovie

thea

ters

clos

edre

stau

rants

clos

edb

usi

nes

ses

mas

ks

for

emp

loye

es

stat

eof

emer

gen

cy











CA

US

AL

IMP

AC

TO

FM

AS

KS

P

OL

ICIE

S

BE

HA

VIO

R17


Dependent variable


(1) (2) (3) (4) (5) (6) (7) (8)













(2074) (3061) (1685) (3793)


j Policyj -2698 -3822 -3190 -3580 -2135 -3200 -2792 -3090






o reuse allowed w

ithout permission

(which w


ho has granted medR



ay 30 2020

httpsdoiorg1011012020052720115139doi

medR

xiv preprint








∆Cit

+εyit (12)

where Bit = (B1it B













Dependent variable


(1) (2) (3) (4)


(0141) (0137)


(0050) (0055) (0045) (0048)



(0068) (0077)


(0056) (0061)



(0049) (0057)




(0003) (0003) (0003) (0003)lag(transit 14) 0002 0002 0004 0005

(0003) (0003) (0004) (0004)


(0020) (0022) (0019) (0019)



(1562) (1504) (1268) (1379)


(0049) (0049)


(0043) (0042)∆D∆C


(0103) (0094)


















masks -004 -009(006) (005)sum



(015) (019)














Dependent variable


(1) (2) (3) (4)


(0145) (0150)


(0095) (0067) (0063) (0071)



(0067) (0076)




(0049) (0058)


(0029) (0024) (0023) (0022)




(0061) (0048)


(0042) (0036)∆D∆C


(0112) (0087)


(018) (020)

Observations 3433 3433 3427 3427R2 0666 0707 0730 0757

Adjusted R2 0665 0705 0729 0755












closed K-12 schools -013 -019 -032 -052 019(013) (009) (014) (014) (007)

stay at home -005 -010 -015 -020 004



(010) (004) (009) (009) (004)


all -021 -042 -063 -101 038(022) (018) (020) (023) (018)






















0

2

4

Mar Apr Maydate

∆log

∆C

Estimated

Observed

Without Policies

∆log∆C

Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C



anda

te fa

ce m

ask

use

by e

mpl

oyee

s in

pub

lic fa

cing

bus

ines

ses

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

0

500000

1000000

1500000

2000000

2500000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek








Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

0e+00

2e+06

4e+06

6e+06

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

0e+00

1e+06

2e+06

3e+06

4e+06

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek






00

02

04

06

08

Mar Apr Maydate

Cha

nge

in ∆

log∆

C



00e+00

25e+07

50e+07

75e+07

10e+08

Mar Apr Maydate

case

s in

the

past

wee

k


0

50

100

150

200

Mar Apr Maydate

rela

tive

incr

ease

in c

ases







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus2000

0

2000

4000

6000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


anda

te fa

ce m

ask

use

by e

mpl

oyee

s in

pub

lic fa

cing

bus

ines

ses

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus5000

0

5000

10000

15000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

0

4000

8000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



minus002

000

002

004

006

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


minus01

00

01

02

03

04

05

Mar Apr Maydate

rela

tive

incr

ease

in c

ases







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus03

minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus5000

minus4000

minus3000

minus2000

minus1000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus03

minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus20000

minus15000

minus10000

minus5000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek






Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus02

minus01

00

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus10000

minus5000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


minus06

minus04

minus02

00

Mar Apr Maydate

rela

tive

incr

ease

in c

ases






6 Conclusion






References


















































1 2 3NYT 100 029 038JHU 029 100 033CTP 038 033 100












1

10

100

1000

10000

Apr May Jundate

New

cas

es

NYT

1

10

100

1000

10000

Apr May Jundate

New

cas

es

JHU

1

10

100

1000

10000

Apr May Jundate

New

cas

es

CTP













1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

NYT

1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

JHU

1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

CTP



1

10

100

1000

10000

Apr May Jundate

Dea

ths

NYT

1

10

100

1000

10000

Apr May Jundate

Dea

ths

JHU

1

10

100

1000

Apr May Jundate

Dea

ths

CTP







1e+01

1e+03

1e+05

Apr May Jundate

Test

s


1e+01

1e+03

1e+05

Apr May Jundate

Test

s



























o reuse allowed w

ithout permission

(which w


ho has granted medR



ay 30 2020

httpsdoiorg1011012020052720115139doi

medR

xiv preprint










1 Introduction








32 Counterfactuals


41 Data







53 Mask Mandate

6 Conclusion

References



A2 Deaths

A3 Tests


A5 Policy Variables

A6 Timing














00

02

04

06

08

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


0

50

100

150

200

Mar Apr Maydate

rela

tive

incr

ease

in c

ases



minus002

000

002

004

006

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


minus01

00

01

02

03

04

05

Mar Apr Maydate

rela

tive

incr

ease

in c

ases








minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


minus06

minus04

minus02

00

Mar Apr Maydate

rela

tive

incr

ease

in c

ases

















other variables







Pit

Iit Yit

Bit

Iit

Wit









(SEM)










Bit =Bit(Pit Iit)

Pit =Pit(Iit)

(O)





which in turn imply








(1)















Iit = log(t)


Iit = Yitminus`


Iit =

(log(t) Yitminus`

infinsumm=1

Yitminus`m

)prime













Ii(tminus7)

Yi(tminus7)

Iit

Yit

Ii(t+7)

Yi(t+7)

SE

M(t-7

)

SE

M(t)

SE

M(t+

7)














∆Cit+ εit (M)



∆Cit



S(t) = minusS(t)

Nβ(t)I(t) (4)

I(t) =S(t)









C(t) = τ(t)I(t) (8)








C(t)

C(t)=S(t)

Nβ(t) +

τ(t)

τ(t)minus 1

π

τ(t)D(t)

C(t) (9)




YitC(t)

C(t)


β(t)


+ δDTit∆Dit

∆Cit

1πτ(t)D(t)

C(t)







= αprimeβprime(

EPit minus EPio


+ πprime(

EPit minus EPio


+ αprimeγprime(

EIit minus EIio

)+ microprime

(EIit minus EIio

)Dynamic Effect


(10)







(I2) DynEt =sumt`


(I3) DynEt =sumt`


+sumt`


+sumt`minus1


)



P itTt=1 i = 1 N



Y it =Yit(B

it P

it I

it)

Bit =Bit(P

it I

it)


lowasti(tminus1) t)

(CEF-SEM)














PY it minus PYit

Predicted CF Change




= PEBit + PED

it + DynEit

(11)



(I2) DynEit =sumt`


i(tminusm`))

(I3) DynEit =sumt`


+sumt`



)+sumt`minus1



)
























resi

den

tial

tran

sit

wor

kp

lace

s

reta

il

groce

ry

clos

edK

-12

sch

ool

s

stay

ath

ome

clos

edm

ovie

thea

ters

clos

edre

stau

rants

clos

edb

usi

nes

ses

mas

ks

for

emp

loye

es

stat

eof

emer

gen

cy











CA

US

AL

IMP

AC

TO

FM

AS

KS

P

OL

ICIE

S

BE

HA

VIO

R17


Dependent variable


(1) (2) (3) (4) (5) (6) (7) (8)













(2074) (3061) (1685) (3793)


j Policyj -2698 -3822 -3190 -3580 -2135 -3200 -2792 -3090






o reuse allowed w

ithout permission

(which w


ho has granted medR



ay 30 2020

httpsdoiorg1011012020052720115139doi

medR

xiv preprint








∆Cit

+εyit (12)

where Bit = (B1it B













Dependent variable


(1) (2) (3) (4)


(0141) (0137)


(0050) (0055) (0045) (0048)



(0068) (0077)


(0056) (0061)



(0049) (0057)




(0003) (0003) (0003) (0003)lag(transit 14) 0002 0002 0004 0005

(0003) (0003) (0004) (0004)


(0020) (0022) (0019) (0019)



(1562) (1504) (1268) (1379)


(0049) (0049)


(0043) (0042)∆D∆C


(0103) (0094)


















masks -004 -009(006) (005)sum



(015) (019)














Dependent variable


(1) (2) (3) (4)


(0145) (0150)


(0095) (0067) (0063) (0071)



(0067) (0076)




(0049) (0058)


(0029) (0024) (0023) (0022)




(0061) (0048)


(0042) (0036)∆D∆C


(0112) (0087)


(018) (020)

Observations 3433 3433 3427 3427R2 0666 0707 0730 0757

Adjusted R2 0665 0705 0729 0755












closed K-12 schools -013 -019 -032 -052 019(013) (009) (014) (014) (007)

stay at home -005 -010 -015 -020 004



(010) (004) (009) (009) (004)


all -021 -042 -063 -101 038(022) (018) (020) (023) (018)






















0

2

4

Mar Apr Maydate

∆log

∆C

Estimated

Observed

Without Policies

∆log∆C

Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C



anda

te fa

ce m

ask

use

by e

mpl

oyee

s in

pub

lic fa

cing

bus

ines

ses

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

0

500000

1000000

1500000

2000000

2500000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek








Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

0e+00

2e+06

4e+06

6e+06

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

0e+00

1e+06

2e+06

3e+06

4e+06

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek






00

02

04

06

08

Mar Apr Maydate

Cha

nge

in ∆

log∆

C



00e+00

25e+07

50e+07

75e+07

10e+08

Mar Apr Maydate

case

s in

the

past

wee

k


0

50

100

150

200

Mar Apr Maydate

rela

tive

incr

ease

in c

ases







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus2000

0

2000

4000

6000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


anda

te fa

ce m

ask

use

by e

mpl

oyee

s in

pub

lic fa

cing

bus

ines

ses

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus5000

0

5000

10000

15000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

0

4000

8000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



minus002

000

002

004

006

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


minus01

00

01

02

03

04

05

Mar Apr Maydate

rela

tive

incr

ease

in c

ases







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus03

minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus5000

minus4000

minus3000

minus2000

minus1000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus03

minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus20000

minus15000

minus10000

minus5000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek






Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus02

minus01

00

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus10000

minus5000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


minus06

minus04

minus02

00

Mar Apr Maydate

rela

tive

incr

ease

in c

ases






6 Conclusion






References


















































1 2 3NYT 100 029 038JHU 029 100 033CTP 038 033 100












1

10

100

1000

10000

Apr May Jundate

New

cas

es

NYT

1

10

100

1000

10000

Apr May Jundate

New

cas

es

JHU

1

10

100

1000

10000

Apr May Jundate

New

cas

es

CTP













1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

NYT

1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

JHU

1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

CTP



1

10

100

1000

10000

Apr May Jundate

Dea

ths

NYT

1

10

100

1000

10000

Apr May Jundate

Dea

ths

JHU

1

10

100

1000

Apr May Jundate

Dea

ths

CTP







1e+01

1e+03

1e+05

Apr May Jundate

Test

s


1e+01

1e+03

1e+05

Apr May Jundate

Test

s



























o reuse allowed w

ithout permission

(which w


ho has granted medR



ay 30 2020

httpsdoiorg1011012020052720115139doi

medR

xiv preprint










1 Introduction








32 Counterfactuals


41 Data







53 Mask Mandate

6 Conclusion

References



A2 Deaths

A3 Tests


A5 Policy Variables

A6 Timing


minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


minus06

minus04

minus02

00

Mar Apr Maydate

rela

tive

incr

ease

in c

ases

















other variables







Pit

Iit Yit

Bit

Iit

Wit









(SEM)










Bit =Bit(Pit Iit)

Pit =Pit(Iit)

(O)





which in turn imply








(1)















Iit = log(t)


Iit = Yitminus`


Iit =

(log(t) Yitminus`

infinsumm=1

Yitminus`m

)prime













Ii(tminus7)

Yi(tminus7)

Iit

Yit

Ii(t+7)

Yi(t+7)

SE

M(t-7

)

SE

M(t)

SE

M(t+

7)














∆Cit+ εit (M)



∆Cit



S(t) = minusS(t)

Nβ(t)I(t) (4)

I(t) =S(t)









C(t) = τ(t)I(t) (8)








C(t)

C(t)=S(t)

Nβ(t) +

τ(t)

τ(t)minus 1

π

τ(t)D(t)

C(t) (9)




YitC(t)

C(t)


β(t)


+ δDTit∆Dit

∆Cit

1πτ(t)D(t)

C(t)







= αprimeβprime(

EPit minus EPio


+ πprime(

EPit minus EPio


+ αprimeγprime(

EIit minus EIio

)+ microprime

(EIit minus EIio

)Dynamic Effect


(10)







(I2) DynEt =sumt`


(I3) DynEt =sumt`


+sumt`


+sumt`minus1


)



P itTt=1 i = 1 N



Y it =Yit(B

it P

it I

it)

Bit =Bit(P

it I

it)


lowasti(tminus1) t)

(CEF-SEM)














PY it minus PYit

Predicted CF Change




= PEBit + PED

it + DynEit

(11)



(I2) DynEit =sumt`


i(tminusm`))

(I3) DynEit =sumt`


+sumt`



)+sumt`minus1



)
























resi

den

tial

tran

sit

wor

kp

lace

s

reta

il

groce

ry

clos

edK

-12

sch

ool

s

stay

ath

ome

clos

edm

ovie

thea

ters

clos

edre

stau

rants

clos

edb

usi

nes

ses

mas

ks

for

emp

loye

es

stat

eof

emer

gen

cy











CA

US

AL

IMP

AC

TO

FM

AS

KS

P

OL

ICIE

S

BE

HA

VIO

R17


Dependent variable


(1) (2) (3) (4) (5) (6) (7) (8)













(2074) (3061) (1685) (3793)


j Policyj -2698 -3822 -3190 -3580 -2135 -3200 -2792 -3090






o reuse allowed w

ithout permission

(which w


ho has granted medR



ay 30 2020

httpsdoiorg1011012020052720115139doi

medR

xiv preprint








∆Cit

+εyit (12)

where Bit = (B1it B













Dependent variable


(1) (2) (3) (4)


(0141) (0137)


(0050) (0055) (0045) (0048)



(0068) (0077)


(0056) (0061)



(0049) (0057)




(0003) (0003) (0003) (0003)lag(transit 14) 0002 0002 0004 0005

(0003) (0003) (0004) (0004)


(0020) (0022) (0019) (0019)



(1562) (1504) (1268) (1379)


(0049) (0049)


(0043) (0042)∆D∆C


(0103) (0094)


















masks -004 -009(006) (005)sum



(015) (019)














Dependent variable


(1) (2) (3) (4)


(0145) (0150)


(0095) (0067) (0063) (0071)



(0067) (0076)




(0049) (0058)


(0029) (0024) (0023) (0022)




(0061) (0048)


(0042) (0036)∆D∆C


(0112) (0087)


(018) (020)

Observations 3433 3433 3427 3427R2 0666 0707 0730 0757

Adjusted R2 0665 0705 0729 0755












closed K-12 schools -013 -019 -032 -052 019(013) (009) (014) (014) (007)

stay at home -005 -010 -015 -020 004



(010) (004) (009) (009) (004)


all -021 -042 -063 -101 038(022) (018) (020) (023) (018)






















0

2

4

Mar Apr Maydate

∆log

∆C

Estimated

Observed

Without Policies

∆log∆C

Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C



anda

te fa

ce m

ask

use

by e

mpl

oyee

s in

pub

lic fa

cing

bus

ines

ses

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

0

500000

1000000

1500000

2000000

2500000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek








Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

0e+00

2e+06

4e+06

6e+06

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

0e+00

1e+06

2e+06

3e+06

4e+06

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek






00

02

04

06

08

Mar Apr Maydate

Cha

nge

in ∆

log∆

C



00e+00

25e+07

50e+07

75e+07

10e+08

Mar Apr Maydate

case

s in

the

past

wee

k


0

50

100

150

200

Mar Apr Maydate

rela

tive

incr

ease

in c

ases







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus2000

0

2000

4000

6000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


anda

te fa

ce m

ask

use

by e

mpl

oyee

s in

pub

lic fa

cing

bus

ines

ses

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus5000

0

5000

10000

15000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

0

4000

8000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



minus002

000

002

004

006

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


minus01

00

01

02

03

04

05

Mar Apr Maydate

rela

tive

incr

ease

in c

ases







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus03

minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus5000

minus4000

minus3000

minus2000

minus1000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus03

minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus20000

minus15000

minus10000

minus5000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek






Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus02

minus01

00

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus10000

minus5000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


minus06

minus04

minus02

00

Mar Apr Maydate

rela

tive

incr

ease

in c

ases






6 Conclusion






References


















































1 2 3NYT 100 029 038JHU 029 100 033CTP 038 033 100












1

10

100

1000

10000

Apr May Jundate

New

cas

es

NYT

1

10

100

1000

10000

Apr May Jundate

New

cas

es

JHU

1

10

100

1000

10000

Apr May Jundate

New

cas

es

CTP













1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

NYT

1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

JHU

1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

CTP



1

10

100

1000

10000

Apr May Jundate

Dea

ths

NYT

1

10

100

1000

10000

Apr May Jundate

Dea

ths

JHU

1

10

100

1000

Apr May Jundate

Dea

ths

CTP







1e+01

1e+03

1e+05

Apr May Jundate

Test

s


1e+01

1e+03

1e+05

Apr May Jundate

Test

s



























o reuse allowed w

ithout permission

(which w


ho has granted medR



ay 30 2020

httpsdoiorg1011012020052720115139doi

medR

xiv preprint










1 Introduction








32 Counterfactuals


41 Data







53 Mask Mandate

6 Conclusion

References



A2 Deaths

A3 Tests


A5 Policy Variables

A6 Timing









other variables







Pit

Iit Yit

Bit

Iit

Wit









(SEM)










Bit =Bit(Pit Iit)

Pit =Pit(Iit)

(O)





which in turn imply








(1)















Iit = log(t)


Iit = Yitminus`


Iit =

(log(t) Yitminus`

infinsumm=1

Yitminus`m

)prime













Ii(tminus7)

Yi(tminus7)

Iit

Yit

Ii(t+7)

Yi(t+7)

SE

M(t-7

)

SE

M(t)

SE

M(t+

7)














∆Cit+ εit (M)



∆Cit



S(t) = minusS(t)

Nβ(t)I(t) (4)

I(t) =S(t)









C(t) = τ(t)I(t) (8)








C(t)

C(t)=S(t)

Nβ(t) +

τ(t)

τ(t)minus 1

π

τ(t)D(t)

C(t) (9)




YitC(t)

C(t)


β(t)


+ δDTit∆Dit

∆Cit

1πτ(t)D(t)

C(t)







= αprimeβprime(

EPit minus EPio


+ πprime(

EPit minus EPio


+ αprimeγprime(

EIit minus EIio

)+ microprime

(EIit minus EIio

)Dynamic Effect


(10)







(I2) DynEt =sumt`


(I3) DynEt =sumt`


+sumt`


+sumt`minus1


)



P itTt=1 i = 1 N



Y it =Yit(B

it P

it I

it)

Bit =Bit(P

it I

it)


lowasti(tminus1) t)

(CEF-SEM)














PY it minus PYit

Predicted CF Change




= PEBit + PED

it + DynEit

(11)



(I2) DynEit =sumt`


i(tminusm`))

(I3) DynEit =sumt`


+sumt`



)+sumt`minus1



)
























resi

den

tial

tran

sit

wor

kp

lace

s

reta

il

groce

ry

clos

edK

-12

sch

ool

s

stay

ath

ome

clos

edm

ovie

thea

ters

clos

edre

stau

rants

clos

edb

usi

nes

ses

mas

ks

for

emp

loye

es

stat

eof

emer

gen

cy











CA

US

AL

IMP

AC

TO

FM

AS

KS

P

OL

ICIE

S

BE

HA

VIO

R17


Dependent variable


(1) (2) (3) (4) (5) (6) (7) (8)













(2074) (3061) (1685) (3793)


j Policyj -2698 -3822 -3190 -3580 -2135 -3200 -2792 -3090






o reuse allowed w

ithout permission

(which w


ho has granted medR



ay 30 2020

httpsdoiorg1011012020052720115139doi

medR

xiv preprint








∆Cit

+εyit (12)

where Bit = (B1it B













Dependent variable


(1) (2) (3) (4)


(0141) (0137)


(0050) (0055) (0045) (0048)



(0068) (0077)


(0056) (0061)



(0049) (0057)




(0003) (0003) (0003) (0003)lag(transit 14) 0002 0002 0004 0005

(0003) (0003) (0004) (0004)


(0020) (0022) (0019) (0019)



(1562) (1504) (1268) (1379)


(0049) (0049)


(0043) (0042)∆D∆C


(0103) (0094)


















masks -004 -009(006) (005)sum



(015) (019)














Dependent variable


(1) (2) (3) (4)


(0145) (0150)


(0095) (0067) (0063) (0071)



(0067) (0076)




(0049) (0058)


(0029) (0024) (0023) (0022)




(0061) (0048)


(0042) (0036)∆D∆C


(0112) (0087)


(018) (020)

Observations 3433 3433 3427 3427R2 0666 0707 0730 0757

Adjusted R2 0665 0705 0729 0755












closed K-12 schools -013 -019 -032 -052 019(013) (009) (014) (014) (007)

stay at home -005 -010 -015 -020 004



(010) (004) (009) (009) (004)


all -021 -042 -063 -101 038(022) (018) (020) (023) (018)






















0

2

4

Mar Apr Maydate

∆log

∆C

Estimated

Observed

Without Policies

∆log∆C

Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C



anda

te fa

ce m

ask

use

by e

mpl

oyee

s in

pub

lic fa

cing

bus

ines

ses

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

0

500000

1000000

1500000

2000000

2500000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek








Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

0e+00

2e+06

4e+06

6e+06

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

0e+00

1e+06

2e+06

3e+06

4e+06

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek






00

02

04

06

08

Mar Apr Maydate

Cha

nge

in ∆

log∆

C



00e+00

25e+07

50e+07

75e+07

10e+08

Mar Apr Maydate

case

s in

the

past

wee

k


0

50

100

150

200

Mar Apr Maydate

rela

tive

incr

ease

in c

ases







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus2000

0

2000

4000

6000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


anda

te fa

ce m

ask

use

by e

mpl

oyee

s in

pub

lic fa

cing

bus

ines

ses

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus5000

0

5000

10000

15000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

0

4000

8000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



minus002

000

002

004

006

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


minus01

00

01

02

03

04

05

Mar Apr Maydate

rela

tive

incr

ease

in c

ases







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus03

minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus5000

minus4000

minus3000

minus2000

minus1000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus03

minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus20000

minus15000

minus10000

minus5000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek






Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus02

minus01

00

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus10000

minus5000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


minus06

minus04

minus02

00

Mar Apr Maydate

rela

tive

incr

ease

in c

ases






6 Conclusion






References


















































1 2 3NYT 100 029 038JHU 029 100 033CTP 038 033 100












1

10

100

1000

10000

Apr May Jundate

New

cas

es

NYT

1

10

100

1000

10000

Apr May Jundate

New

cas

es

JHU

1

10

100

1000

10000

Apr May Jundate

New

cas

es

CTP













1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

NYT

1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

JHU

1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

CTP



1

10

100

1000

10000

Apr May Jundate

Dea

ths

NYT

1

10

100

1000

10000

Apr May Jundate

Dea

ths

JHU

1

10

100

1000

Apr May Jundate

Dea

ths

CTP







1e+01

1e+03

1e+05

Apr May Jundate

Test

s


1e+01

1e+03

1e+05

Apr May Jundate

Test

s



























o reuse allowed w

ithout permission

(which w


ho has granted medR



ay 30 2020

httpsdoiorg1011012020052720115139doi

medR

xiv preprint










1 Introduction








32 Counterfactuals


41 Data







53 Mask Mandate

6 Conclusion

References



A2 Deaths

A3 Tests


A5 Policy Variables

A6 Timing


Pit

Iit Yit

Bit

Iit

Wit









(SEM)










Bit =Bit(Pit Iit)

Pit =Pit(Iit)

(O)





which in turn imply








(1)















Iit = log(t)


Iit = Yitminus`


Iit =

(log(t) Yitminus`

infinsumm=1

Yitminus`m

)prime













Ii(tminus7)

Yi(tminus7)

Iit

Yit

Ii(t+7)

Yi(t+7)

SE

M(t-7

)

SE

M(t)

SE

M(t+

7)














∆Cit+ εit (M)



∆Cit



S(t) = minusS(t)

Nβ(t)I(t) (4)

I(t) =S(t)









C(t) = τ(t)I(t) (8)








C(t)

C(t)=S(t)

Nβ(t) +

τ(t)

τ(t)minus 1

π

τ(t)D(t)

C(t) (9)




YitC(t)

C(t)


β(t)


+ δDTit∆Dit

∆Cit

1πτ(t)D(t)

C(t)







= αprimeβprime(

EPit minus EPio


+ πprime(

EPit minus EPio


+ αprimeγprime(

EIit minus EIio

)+ microprime

(EIit minus EIio

)Dynamic Effect


(10)







(I2) DynEt =sumt`


(I3) DynEt =sumt`


+sumt`


+sumt`minus1


)



P itTt=1 i = 1 N



Y it =Yit(B

it P

it I

it)

Bit =Bit(P

it I

it)


lowasti(tminus1) t)

(CEF-SEM)














PY it minus PYit

Predicted CF Change




= PEBit + PED

it + DynEit

(11)



(I2) DynEit =sumt`


i(tminusm`))

(I3) DynEit =sumt`


+sumt`



)+sumt`minus1



)
























resi

den

tial

tran

sit

wor

kp

lace

s

reta

il

groce

ry

clos

edK

-12

sch

ool

s

stay

ath

ome

clos

edm

ovie

thea

ters

clos

edre

stau

rants

clos

edb

usi

nes

ses

mas

ks

for

emp

loye

es

stat

eof

emer

gen

cy











CA

US

AL

IMP

AC

TO

FM

AS

KS

P

OL

ICIE

S

BE

HA

VIO

R17


Dependent variable


(1) (2) (3) (4) (5) (6) (7) (8)













(2074) (3061) (1685) (3793)


j Policyj -2698 -3822 -3190 -3580 -2135 -3200 -2792 -3090






o reuse allowed w

ithout permission

(which w


ho has granted medR



ay 30 2020

httpsdoiorg1011012020052720115139doi

medR

xiv preprint








∆Cit

+εyit (12)

where Bit = (B1it B













Dependent variable


(1) (2) (3) (4)


(0141) (0137)


(0050) (0055) (0045) (0048)



(0068) (0077)


(0056) (0061)



(0049) (0057)




(0003) (0003) (0003) (0003)lag(transit 14) 0002 0002 0004 0005

(0003) (0003) (0004) (0004)


(0020) (0022) (0019) (0019)



(1562) (1504) (1268) (1379)


(0049) (0049)


(0043) (0042)∆D∆C


(0103) (0094)


















masks -004 -009(006) (005)sum



(015) (019)














Dependent variable


(1) (2) (3) (4)


(0145) (0150)


(0095) (0067) (0063) (0071)



(0067) (0076)




(0049) (0058)


(0029) (0024) (0023) (0022)




(0061) (0048)


(0042) (0036)∆D∆C


(0112) (0087)


(018) (020)

Observations 3433 3433 3427 3427R2 0666 0707 0730 0757

Adjusted R2 0665 0705 0729 0755












closed K-12 schools -013 -019 -032 -052 019(013) (009) (014) (014) (007)

stay at home -005 -010 -015 -020 004



(010) (004) (009) (009) (004)


all -021 -042 -063 -101 038(022) (018) (020) (023) (018)






















0

2

4

Mar Apr Maydate

∆log

∆C

Estimated

Observed

Without Policies

∆log∆C

Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C



anda

te fa

ce m

ask

use

by e

mpl

oyee

s in

pub

lic fa

cing

bus

ines

ses

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

0

500000

1000000

1500000

2000000

2500000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek








Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

0e+00

2e+06

4e+06

6e+06

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

0e+00

1e+06

2e+06

3e+06

4e+06

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek






00

02

04

06

08

Mar Apr Maydate

Cha

nge

in ∆

log∆

C



00e+00

25e+07

50e+07

75e+07

10e+08

Mar Apr Maydate

case

s in

the

past

wee

k


0

50

100

150

200

Mar Apr Maydate

rela

tive

incr

ease

in c

ases







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus2000

0

2000

4000

6000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


anda

te fa

ce m

ask

use

by e

mpl

oyee

s in

pub

lic fa

cing

bus

ines

ses

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus5000

0

5000

10000

15000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

0

4000

8000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



minus002

000

002

004

006

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


minus01

00

01

02

03

04

05

Mar Apr Maydate

rela

tive

incr

ease

in c

ases







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus03

minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus5000

minus4000

minus3000

minus2000

minus1000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus03

minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus20000

minus15000

minus10000

minus5000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek






Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus02

minus01

00

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus10000

minus5000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


minus06

minus04

minus02

00

Mar Apr Maydate

rela

tive

incr

ease

in c

ases






6 Conclusion






References


















































1 2 3NYT 100 029 038JHU 029 100 033CTP 038 033 100












1

10

100

1000

10000

Apr May Jundate

New

cas

es

NYT

1

10

100

1000

10000

Apr May Jundate

New

cas

es

JHU

1

10

100

1000

10000

Apr May Jundate

New

cas

es

CTP













1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

NYT

1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

JHU

1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

CTP



1

10

100

1000

10000

Apr May Jundate

Dea

ths

NYT

1

10

100

1000

10000

Apr May Jundate

Dea

ths

JHU

1

10

100

1000

Apr May Jundate

Dea

ths

CTP







1e+01

1e+03

1e+05

Apr May Jundate

Test

s


1e+01

1e+03

1e+05

Apr May Jundate

Test

s



























o reuse allowed w

ithout permission

(which w


ho has granted medR



ay 30 2020

httpsdoiorg1011012020052720115139doi

medR

xiv preprint










1 Introduction








32 Counterfactuals


41 Data







53 Mask Mandate

6 Conclusion

References



A2 Deaths

A3 Tests


A5 Policy Variables

A6 Timing







Bit =Bit(Pit Iit)

Pit =Pit(Iit)

(O)





which in turn imply








(1)















Iit = log(t)


Iit = Yitminus`


Iit =

(log(t) Yitminus`

infinsumm=1

Yitminus`m

)prime













Ii(tminus7)

Yi(tminus7)

Iit

Yit

Ii(t+7)

Yi(t+7)

SE

M(t-7

)

SE

M(t)

SE

M(t+

7)














∆Cit+ εit (M)



∆Cit



S(t) = minusS(t)

Nβ(t)I(t) (4)

I(t) =S(t)









C(t) = τ(t)I(t) (8)








C(t)

C(t)=S(t)

Nβ(t) +

τ(t)

τ(t)minus 1

π

τ(t)D(t)

C(t) (9)




YitC(t)

C(t)


β(t)


+ δDTit∆Dit

∆Cit

1πτ(t)D(t)

C(t)







= αprimeβprime(

EPit minus EPio


+ πprime(

EPit minus EPio


+ αprimeγprime(

EIit minus EIio

)+ microprime

(EIit minus EIio

)Dynamic Effect


(10)







(I2) DynEt =sumt`


(I3) DynEt =sumt`


+sumt`


+sumt`minus1


)



P itTt=1 i = 1 N



Y it =Yit(B

it P

it I

it)

Bit =Bit(P

it I

it)


lowasti(tminus1) t)

(CEF-SEM)














PY it minus PYit

Predicted CF Change




= PEBit + PED

it + DynEit

(11)



(I2) DynEit =sumt`


i(tminusm`))

(I3) DynEit =sumt`


+sumt`



)+sumt`minus1



)
























resi

den

tial

tran

sit

wor

kp

lace

s

reta

il

groce

ry

clos

edK

-12

sch

ool

s

stay

ath

ome

clos

edm

ovie

thea

ters

clos

edre

stau

rants

clos

edb

usi

nes

ses

mas

ks

for

emp

loye

es

stat

eof

emer

gen

cy











CA

US

AL

IMP

AC

TO

FM

AS

KS

P

OL

ICIE

S

BE

HA

VIO

R17


Dependent variable


(1) (2) (3) (4) (5) (6) (7) (8)













(2074) (3061) (1685) (3793)


j Policyj -2698 -3822 -3190 -3580 -2135 -3200 -2792 -3090






o reuse allowed w

ithout permission

(which w


ho has granted medR



ay 30 2020

httpsdoiorg1011012020052720115139doi

medR

xiv preprint








∆Cit

+εyit (12)

where Bit = (B1it B













Dependent variable


(1) (2) (3) (4)


(0141) (0137)


(0050) (0055) (0045) (0048)



(0068) (0077)


(0056) (0061)



(0049) (0057)




(0003) (0003) (0003) (0003)lag(transit 14) 0002 0002 0004 0005

(0003) (0003) (0004) (0004)


(0020) (0022) (0019) (0019)



(1562) (1504) (1268) (1379)


(0049) (0049)


(0043) (0042)∆D∆C


(0103) (0094)


















masks -004 -009(006) (005)sum



(015) (019)














Dependent variable


(1) (2) (3) (4)


(0145) (0150)


(0095) (0067) (0063) (0071)



(0067) (0076)




(0049) (0058)


(0029) (0024) (0023) (0022)




(0061) (0048)


(0042) (0036)∆D∆C


(0112) (0087)


(018) (020)

Observations 3433 3433 3427 3427R2 0666 0707 0730 0757

Adjusted R2 0665 0705 0729 0755












closed K-12 schools -013 -019 -032 -052 019(013) (009) (014) (014) (007)

stay at home -005 -010 -015 -020 004



(010) (004) (009) (009) (004)


all -021 -042 -063 -101 038(022) (018) (020) (023) (018)






















0

2

4

Mar Apr Maydate

∆log

∆C

Estimated

Observed

Without Policies

∆log∆C

Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C



anda

te fa

ce m

ask

use

by e

mpl

oyee

s in

pub

lic fa

cing

bus

ines

ses

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

0

500000

1000000

1500000

2000000

2500000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek








Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

0e+00

2e+06

4e+06

6e+06

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

0e+00

1e+06

2e+06

3e+06

4e+06

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek






00

02

04

06

08

Mar Apr Maydate

Cha

nge

in ∆

log∆

C



00e+00

25e+07

50e+07

75e+07

10e+08

Mar Apr Maydate

case

s in

the

past

wee

k


0

50

100

150

200

Mar Apr Maydate

rela

tive

incr

ease

in c

ases







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus2000

0

2000

4000

6000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


anda

te fa

ce m

ask

use

by e

mpl

oyee

s in

pub

lic fa

cing

bus

ines

ses

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus5000

0

5000

10000

15000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

0

4000

8000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



minus002

000

002

004

006

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


minus01

00

01

02

03

04

05

Mar Apr Maydate

rela

tive

incr

ease

in c

ases







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus03

minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus5000

minus4000

minus3000

minus2000

minus1000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus03

minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus20000

minus15000

minus10000

minus5000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek






Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus02

minus01

00

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus10000

minus5000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


minus06

minus04

minus02

00

Mar Apr Maydate

rela

tive

incr

ease

in c

ases






6 Conclusion






References


















































1 2 3NYT 100 029 038JHU 029 100 033CTP 038 033 100












1

10

100

1000

10000

Apr May Jundate

New

cas

es

NYT

1

10

100

1000

10000

Apr May Jundate

New

cas

es

JHU

1

10

100

1000

10000

Apr May Jundate

New

cas

es

CTP













1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

NYT

1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

JHU

1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

CTP



1

10

100

1000

10000

Apr May Jundate

Dea

ths

NYT

1

10

100

1000

10000

Apr May Jundate

Dea

ths

JHU

1

10

100

1000

Apr May Jundate

Dea

ths

CTP







1e+01

1e+03

1e+05

Apr May Jundate

Test

s


1e+01

1e+03

1e+05

Apr May Jundate

Test

s



























o reuse allowed w

ithout permission

(which w


ho has granted medR



ay 30 2020

httpsdoiorg1011012020052720115139doi

medR

xiv preprint










1 Introduction








32 Counterfactuals


41 Data







53 Mask Mandate

6 Conclusion

References



A2 Deaths

A3 Tests


A5 Policy Variables

A6 Timing






Iit = log(t)


Iit = Yitminus`


Iit =

(log(t) Yitminus`

infinsumm=1

Yitminus`m

)prime













Ii(tminus7)

Yi(tminus7)

Iit

Yit

Ii(t+7)

Yi(t+7)

SE

M(t-7

)

SE

M(t)

SE

M(t+

7)














∆Cit+ εit (M)



∆Cit



S(t) = minusS(t)

Nβ(t)I(t) (4)

I(t) =S(t)









C(t) = τ(t)I(t) (8)








C(t)

C(t)=S(t)

Nβ(t) +

τ(t)

τ(t)minus 1

π

τ(t)D(t)

C(t) (9)




YitC(t)

C(t)


β(t)


+ δDTit∆Dit

∆Cit

1πτ(t)D(t)

C(t)







= αprimeβprime(

EPit minus EPio


+ πprime(

EPit minus EPio


+ αprimeγprime(

EIit minus EIio

)+ microprime

(EIit minus EIio

)Dynamic Effect


(10)







(I2) DynEt =sumt`


(I3) DynEt =sumt`


+sumt`


+sumt`minus1


)



P itTt=1 i = 1 N



Y it =Yit(B

it P

it I

it)

Bit =Bit(P

it I

it)


lowasti(tminus1) t)

(CEF-SEM)














PY it minus PYit

Predicted CF Change




= PEBit + PED

it + DynEit

(11)



(I2) DynEit =sumt`


i(tminusm`))

(I3) DynEit =sumt`


+sumt`



)+sumt`minus1



)
























resi

den

tial

tran

sit

wor

kp

lace

s

reta

il

groce

ry

clos

edK

-12

sch

ool

s

stay

ath

ome

clos

edm

ovie

thea

ters

clos

edre

stau

rants

clos

edb

usi

nes

ses

mas

ks

for

emp

loye

es

stat

eof

emer

gen

cy











CA

US

AL

IMP

AC

TO

FM

AS

KS

P

OL

ICIE

S

BE

HA

VIO

R17


Dependent variable


(1) (2) (3) (4) (5) (6) (7) (8)













(2074) (3061) (1685) (3793)


j Policyj -2698 -3822 -3190 -3580 -2135 -3200 -2792 -3090






o reuse allowed w

ithout permission

(which w


ho has granted medR



ay 30 2020

httpsdoiorg1011012020052720115139doi

medR

xiv preprint








∆Cit

+εyit (12)

where Bit = (B1it B













Dependent variable


(1) (2) (3) (4)


(0141) (0137)


(0050) (0055) (0045) (0048)



(0068) (0077)


(0056) (0061)



(0049) (0057)




(0003) (0003) (0003) (0003)lag(transit 14) 0002 0002 0004 0005

(0003) (0003) (0004) (0004)


(0020) (0022) (0019) (0019)



(1562) (1504) (1268) (1379)


(0049) (0049)


(0043) (0042)∆D∆C


(0103) (0094)


















masks -004 -009(006) (005)sum



(015) (019)














Dependent variable


(1) (2) (3) (4)


(0145) (0150)


(0095) (0067) (0063) (0071)



(0067) (0076)




(0049) (0058)


(0029) (0024) (0023) (0022)




(0061) (0048)


(0042) (0036)∆D∆C


(0112) (0087)


(018) (020)

Observations 3433 3433 3427 3427R2 0666 0707 0730 0757

Adjusted R2 0665 0705 0729 0755












closed K-12 schools -013 -019 -032 -052 019(013) (009) (014) (014) (007)

stay at home -005 -010 -015 -020 004



(010) (004) (009) (009) (004)


all -021 -042 -063 -101 038(022) (018) (020) (023) (018)






















0

2

4

Mar Apr Maydate

∆log

∆C

Estimated

Observed

Without Policies

∆log∆C

Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C



anda

te fa

ce m

ask

use

by e

mpl

oyee

s in

pub

lic fa

cing

bus

ines

ses

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

0

500000

1000000

1500000

2000000

2500000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek








Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

0e+00

2e+06

4e+06

6e+06

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

0e+00

1e+06

2e+06

3e+06

4e+06

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek






00

02

04

06

08

Mar Apr Maydate

Cha

nge

in ∆

log∆

C



00e+00

25e+07

50e+07

75e+07

10e+08

Mar Apr Maydate

case

s in

the

past

wee

k


0

50

100

150

200

Mar Apr Maydate

rela

tive

incr

ease

in c

ases







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus2000

0

2000

4000

6000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


anda

te fa

ce m

ask

use

by e

mpl

oyee

s in

pub

lic fa

cing

bus

ines

ses

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus5000

0

5000

10000

15000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

0

4000

8000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



minus002

000

002

004

006

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


minus01

00

01

02

03

04

05

Mar Apr Maydate

rela

tive

incr

ease

in c

ases







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus03

minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus5000

minus4000

minus3000

minus2000

minus1000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus03

minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus20000

minus15000

minus10000

minus5000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek






Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus02

minus01

00

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus10000

minus5000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


minus06

minus04

minus02

00

Mar Apr Maydate

rela

tive

incr

ease

in c

ases






6 Conclusion






References


















































1 2 3NYT 100 029 038JHU 029 100 033CTP 038 033 100












1

10

100

1000

10000

Apr May Jundate

New

cas

es

NYT

1

10

100

1000

10000

Apr May Jundate

New

cas

es

JHU

1

10

100

1000

10000

Apr May Jundate

New

cas

es

CTP













1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

NYT

1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

JHU

1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

CTP



1

10

100

1000

10000

Apr May Jundate

Dea

ths

NYT

1

10

100

1000

10000

Apr May Jundate

Dea

ths

JHU

1

10

100

1000

Apr May Jundate

Dea

ths

CTP







1e+01

1e+03

1e+05

Apr May Jundate

Test

s


1e+01

1e+03

1e+05

Apr May Jundate

Test

s



























o reuse allowed w

ithout permission

(which w


ho has granted medR



ay 30 2020

httpsdoiorg1011012020052720115139doi

medR

xiv preprint










1 Introduction








32 Counterfactuals


41 Data







53 Mask Mandate

6 Conclusion

References



A2 Deaths

A3 Tests


A5 Policy Variables

A6 Timing







Ii(tminus7)

Yi(tminus7)

Iit

Yit

Ii(t+7)

Yi(t+7)

SE

M(t-7

)

SE

M(t)

SE

M(t+

7)














∆Cit+ εit (M)



∆Cit



S(t) = minusS(t)

Nβ(t)I(t) (4)

I(t) =S(t)









C(t) = τ(t)I(t) (8)








C(t)

C(t)=S(t)

Nβ(t) +

τ(t)

τ(t)minus 1

π

τ(t)D(t)

C(t) (9)




YitC(t)

C(t)


β(t)


+ δDTit∆Dit

∆Cit

1πτ(t)D(t)

C(t)







= αprimeβprime(

EPit minus EPio


+ πprime(

EPit minus EPio


+ αprimeγprime(

EIit minus EIio

)+ microprime

(EIit minus EIio

)Dynamic Effect


(10)







(I2) DynEt =sumt`


(I3) DynEt =sumt`


+sumt`


+sumt`minus1


)



P itTt=1 i = 1 N



Y it =Yit(B

it P

it I

it)

Bit =Bit(P

it I

it)


lowasti(tminus1) t)

(CEF-SEM)














PY it minus PYit

Predicted CF Change




= PEBit + PED

it + DynEit

(11)



(I2) DynEit =sumt`


i(tminusm`))

(I3) DynEit =sumt`


+sumt`



)+sumt`minus1



)
























resi

den

tial

tran

sit

wor

kp

lace

s

reta

il

groce

ry

clos

edK

-12

sch

ool

s

stay

ath

ome

clos

edm

ovie

thea

ters

clos

edre

stau

rants

clos

edb

usi

nes

ses

mas

ks

for

emp

loye

es

stat

eof

emer

gen

cy











CA

US

AL

IMP

AC

TO

FM

AS

KS

P

OL

ICIE

S

BE

HA

VIO

R17


Dependent variable


(1) (2) (3) (4) (5) (6) (7) (8)













(2074) (3061) (1685) (3793)


j Policyj -2698 -3822 -3190 -3580 -2135 -3200 -2792 -3090






o reuse allowed w

ithout permission

(which w


ho has granted medR



ay 30 2020

httpsdoiorg1011012020052720115139doi

medR

xiv preprint








∆Cit

+εyit (12)

where Bit = (B1it B













Dependent variable


(1) (2) (3) (4)


(0141) (0137)


(0050) (0055) (0045) (0048)



(0068) (0077)


(0056) (0061)



(0049) (0057)




(0003) (0003) (0003) (0003)lag(transit 14) 0002 0002 0004 0005

(0003) (0003) (0004) (0004)


(0020) (0022) (0019) (0019)



(1562) (1504) (1268) (1379)


(0049) (0049)


(0043) (0042)∆D∆C


(0103) (0094)


















masks -004 -009(006) (005)sum



(015) (019)














Dependent variable


(1) (2) (3) (4)


(0145) (0150)


(0095) (0067) (0063) (0071)



(0067) (0076)




(0049) (0058)


(0029) (0024) (0023) (0022)




(0061) (0048)


(0042) (0036)∆D∆C


(0112) (0087)


(018) (020)

Observations 3433 3433 3427 3427R2 0666 0707 0730 0757

Adjusted R2 0665 0705 0729 0755












closed K-12 schools -013 -019 -032 -052 019(013) (009) (014) (014) (007)

stay at home -005 -010 -015 -020 004



(010) (004) (009) (009) (004)


all -021 -042 -063 -101 038(022) (018) (020) (023) (018)






















0

2

4

Mar Apr Maydate

∆log

∆C

Estimated

Observed

Without Policies

∆log∆C

Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C



anda

te fa

ce m

ask

use

by e

mpl

oyee

s in

pub

lic fa

cing

bus

ines

ses

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

0

500000

1000000

1500000

2000000

2500000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek








Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

0e+00

2e+06

4e+06

6e+06

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

0e+00

1e+06

2e+06

3e+06

4e+06

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek






00

02

04

06

08

Mar Apr Maydate

Cha

nge

in ∆

log∆

C



00e+00

25e+07

50e+07

75e+07

10e+08

Mar Apr Maydate

case

s in

the

past

wee

k


0

50

100

150

200

Mar Apr Maydate

rela

tive

incr

ease

in c

ases







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus2000

0

2000

4000

6000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


anda

te fa

ce m

ask

use

by e

mpl

oyee

s in

pub

lic fa

cing

bus

ines

ses

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus5000

0

5000

10000

15000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

0

4000

8000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



minus002

000

002

004

006

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


minus01

00

01

02

03

04

05

Mar Apr Maydate

rela

tive

incr

ease

in c

ases







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus03

minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus5000

minus4000

minus3000

minus2000

minus1000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus03

minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus20000

minus15000

minus10000

minus5000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek






Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus02

minus01

00

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus10000

minus5000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


minus06

minus04

minus02

00

Mar Apr Maydate

rela

tive

incr

ease

in c

ases






6 Conclusion






References


















































1 2 3NYT 100 029 038JHU 029 100 033CTP 038 033 100












1

10

100

1000

10000

Apr May Jundate

New

cas

es

NYT

1

10

100

1000

10000

Apr May Jundate

New

cas

es

JHU

1

10

100

1000

10000

Apr May Jundate

New

cas

es

CTP













1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

NYT

1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

JHU

1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

CTP



1

10

100

1000

10000

Apr May Jundate

Dea

ths

NYT

1

10

100

1000

10000

Apr May Jundate

Dea

ths

JHU

1

10

100

1000

Apr May Jundate

Dea

ths

CTP







1e+01

1e+03

1e+05

Apr May Jundate

Test

s


1e+01

1e+03

1e+05

Apr May Jundate

Test

s



























o reuse allowed w

ithout permission

(which w


ho has granted medR



ay 30 2020

httpsdoiorg1011012020052720115139doi

medR

xiv preprint










1 Introduction








32 Counterfactuals


41 Data







53 Mask Mandate

6 Conclusion

References



A2 Deaths

A3 Tests


A5 Policy Variables

A6 Timing





∆Cit+ εit (M)



∆Cit



S(t) = minusS(t)

Nβ(t)I(t) (4)

I(t) =S(t)









C(t) = τ(t)I(t) (8)








C(t)

C(t)=S(t)

Nβ(t) +

τ(t)

τ(t)minus 1

π

τ(t)D(t)

C(t) (9)




YitC(t)

C(t)


β(t)


+ δDTit∆Dit

∆Cit

1πτ(t)D(t)

C(t)







= αprimeβprime(

EPit minus EPio


+ πprime(

EPit minus EPio


+ αprimeγprime(

EIit minus EIio

)+ microprime

(EIit minus EIio

)Dynamic Effect


(10)







(I2) DynEt =sumt`


(I3) DynEt =sumt`


+sumt`


+sumt`minus1


)



P itTt=1 i = 1 N



Y it =Yit(B

it P

it I

it)

Bit =Bit(P

it I

it)


lowasti(tminus1) t)

(CEF-SEM)














PY it minus PYit

Predicted CF Change




= PEBit + PED

it + DynEit

(11)



(I2) DynEit =sumt`


i(tminusm`))

(I3) DynEit =sumt`


+sumt`



)+sumt`minus1



)
























resi

den

tial

tran

sit

wor

kp

lace

s

reta

il

groce

ry

clos

edK

-12

sch

ool

s

stay

ath

ome

clos

edm

ovie

thea

ters

clos

edre

stau

rants

clos

edb

usi

nes

ses

mas

ks

for

emp

loye

es

stat

eof

emer

gen

cy











CA

US

AL

IMP

AC

TO

FM

AS

KS

P

OL

ICIE

S

BE

HA

VIO

R17


Dependent variable


(1) (2) (3) (4) (5) (6) (7) (8)













(2074) (3061) (1685) (3793)


j Policyj -2698 -3822 -3190 -3580 -2135 -3200 -2792 -3090






o reuse allowed w

ithout permission

(which w


ho has granted medR



ay 30 2020

httpsdoiorg1011012020052720115139doi

medR

xiv preprint








∆Cit

+εyit (12)

where Bit = (B1it B













Dependent variable


(1) (2) (3) (4)


(0141) (0137)


(0050) (0055) (0045) (0048)



(0068) (0077)


(0056) (0061)



(0049) (0057)




(0003) (0003) (0003) (0003)lag(transit 14) 0002 0002 0004 0005

(0003) (0003) (0004) (0004)


(0020) (0022) (0019) (0019)



(1562) (1504) (1268) (1379)


(0049) (0049)


(0043) (0042)∆D∆C


(0103) (0094)


















masks -004 -009(006) (005)sum



(015) (019)














Dependent variable


(1) (2) (3) (4)


(0145) (0150)


(0095) (0067) (0063) (0071)



(0067) (0076)




(0049) (0058)


(0029) (0024) (0023) (0022)




(0061) (0048)


(0042) (0036)∆D∆C


(0112) (0087)


(018) (020)

Observations 3433 3433 3427 3427R2 0666 0707 0730 0757

Adjusted R2 0665 0705 0729 0755












closed K-12 schools -013 -019 -032 -052 019(013) (009) (014) (014) (007)

stay at home -005 -010 -015 -020 004



(010) (004) (009) (009) (004)


all -021 -042 -063 -101 038(022) (018) (020) (023) (018)






















0

2

4

Mar Apr Maydate

∆log

∆C

Estimated

Observed

Without Policies

∆log∆C

Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C



anda

te fa

ce m

ask

use

by e

mpl

oyee

s in

pub

lic fa

cing

bus

ines

ses

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

0

500000

1000000

1500000

2000000

2500000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek








Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

0e+00

2e+06

4e+06

6e+06

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

0e+00

1e+06

2e+06

3e+06

4e+06

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek






00

02

04

06

08

Mar Apr Maydate

Cha

nge

in ∆

log∆

C



00e+00

25e+07

50e+07

75e+07

10e+08

Mar Apr Maydate

case

s in

the

past

wee

k


0

50

100

150

200

Mar Apr Maydate

rela

tive

incr

ease

in c

ases







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus2000

0

2000

4000

6000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


anda

te fa

ce m

ask

use

by e

mpl

oyee

s in

pub

lic fa

cing

bus

ines

ses

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus5000

0

5000

10000

15000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

0

4000

8000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



minus002

000

002

004

006

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


minus01

00

01

02

03

04

05

Mar Apr Maydate

rela

tive

incr

ease

in c

ases







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus03

minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus5000

minus4000

minus3000

minus2000

minus1000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus03

minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus20000

minus15000

minus10000

minus5000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek






Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus02

minus01

00

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus10000

minus5000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


minus06

minus04

minus02

00

Mar Apr Maydate

rela

tive

incr

ease

in c

ases






6 Conclusion






References


















































1 2 3NYT 100 029 038JHU 029 100 033CTP 038 033 100












1

10

100

1000

10000

Apr May Jundate

New

cas

es

NYT

1

10

100

1000

10000

Apr May Jundate

New

cas

es

JHU

1

10

100

1000

10000

Apr May Jundate

New

cas

es

CTP













1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

NYT

1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

JHU

1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

CTP



1

10

100

1000

10000

Apr May Jundate

Dea

ths

NYT

1

10

100

1000

10000

Apr May Jundate

Dea

ths

JHU

1

10

100

1000

Apr May Jundate

Dea

ths

CTP







1e+01

1e+03

1e+05

Apr May Jundate

Test

s


1e+01

1e+03

1e+05

Apr May Jundate

Test

s



























o reuse allowed w

ithout permission

(which w


ho has granted medR



ay 30 2020

httpsdoiorg1011012020052720115139doi

medR

xiv preprint










1 Introduction








32 Counterfactuals


41 Data







53 Mask Mandate

6 Conclusion

References



A2 Deaths

A3 Tests


A5 Policy Variables

A6 Timing



C(t)

C(t)=S(t)

Nβ(t) +

τ(t)

τ(t)minus 1

π

τ(t)D(t)

C(t) (9)




YitC(t)

C(t)


β(t)


+ δDTit∆Dit

∆Cit

1πτ(t)D(t)

C(t)







= αprimeβprime(

EPit minus EPio


+ πprime(

EPit minus EPio


+ αprimeγprime(

EIit minus EIio

)+ microprime

(EIit minus EIio

)Dynamic Effect


(10)







(I2) DynEt =sumt`


(I3) DynEt =sumt`


+sumt`


+sumt`minus1


)



P itTt=1 i = 1 N



Y it =Yit(B

it P

it I

it)

Bit =Bit(P

it I

it)


lowasti(tminus1) t)

(CEF-SEM)














PY it minus PYit

Predicted CF Change




= PEBit + PED

it + DynEit

(11)



(I2) DynEit =sumt`


i(tminusm`))

(I3) DynEit =sumt`


+sumt`



)+sumt`minus1



)
























resi

den

tial

tran

sit

wor

kp

lace

s

reta

il

groce

ry

clos

edK

-12

sch

ool

s

stay

ath

ome

clos

edm

ovie

thea

ters

clos

edre

stau

rants

clos

edb

usi

nes

ses

mas

ks

for

emp

loye

es

stat

eof

emer

gen

cy











CA

US

AL

IMP

AC

TO

FM

AS

KS

P

OL

ICIE

S

BE

HA

VIO

R17


Dependent variable


(1) (2) (3) (4) (5) (6) (7) (8)













(2074) (3061) (1685) (3793)


j Policyj -2698 -3822 -3190 -3580 -2135 -3200 -2792 -3090






o reuse allowed w

ithout permission

(which w


ho has granted medR



ay 30 2020

httpsdoiorg1011012020052720115139doi

medR

xiv preprint








∆Cit

+εyit (12)

where Bit = (B1it B













Dependent variable


(1) (2) (3) (4)


(0141) (0137)


(0050) (0055) (0045) (0048)



(0068) (0077)


(0056) (0061)



(0049) (0057)




(0003) (0003) (0003) (0003)lag(transit 14) 0002 0002 0004 0005

(0003) (0003) (0004) (0004)


(0020) (0022) (0019) (0019)



(1562) (1504) (1268) (1379)


(0049) (0049)


(0043) (0042)∆D∆C


(0103) (0094)


















masks -004 -009(006) (005)sum



(015) (019)














Dependent variable


(1) (2) (3) (4)


(0145) (0150)


(0095) (0067) (0063) (0071)



(0067) (0076)




(0049) (0058)


(0029) (0024) (0023) (0022)




(0061) (0048)


(0042) (0036)∆D∆C


(0112) (0087)


(018) (020)

Observations 3433 3433 3427 3427R2 0666 0707 0730 0757

Adjusted R2 0665 0705 0729 0755












closed K-12 schools -013 -019 -032 -052 019(013) (009) (014) (014) (007)

stay at home -005 -010 -015 -020 004



(010) (004) (009) (009) (004)


all -021 -042 -063 -101 038(022) (018) (020) (023) (018)






















0

2

4

Mar Apr Maydate

∆log

∆C

Estimated

Observed

Without Policies

∆log∆C

Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C



anda

te fa

ce m

ask

use

by e

mpl

oyee

s in

pub

lic fa

cing

bus

ines

ses

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

0

500000

1000000

1500000

2000000

2500000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek








Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

0e+00

2e+06

4e+06

6e+06

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

0e+00

1e+06

2e+06

3e+06

4e+06

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek






00

02

04

06

08

Mar Apr Maydate

Cha

nge

in ∆

log∆

C



00e+00

25e+07

50e+07

75e+07

10e+08

Mar Apr Maydate

case

s in

the

past

wee

k


0

50

100

150

200

Mar Apr Maydate

rela

tive

incr

ease

in c

ases







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus2000

0

2000

4000

6000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


anda

te fa

ce m

ask

use

by e

mpl

oyee

s in

pub

lic fa

cing

bus

ines

ses

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus5000

0

5000

10000

15000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

0

4000

8000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



minus002

000

002

004

006

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


minus01

00

01

02

03

04

05

Mar Apr Maydate

rela

tive

incr

ease

in c

ases







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus03

minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus5000

minus4000

minus3000

minus2000

minus1000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus03

minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus20000

minus15000

minus10000

minus5000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek






Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus02

minus01

00

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus10000

minus5000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


minus06

minus04

minus02

00

Mar Apr Maydate

rela

tive

incr

ease

in c

ases






6 Conclusion






References


















































1 2 3NYT 100 029 038JHU 029 100 033CTP 038 033 100












1

10

100

1000

10000

Apr May Jundate

New

cas

es

NYT

1

10

100

1000

10000

Apr May Jundate

New

cas

es

JHU

1

10

100

1000

10000

Apr May Jundate

New

cas

es

CTP













1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

NYT

1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

JHU

1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

CTP



1

10

100

1000

10000

Apr May Jundate

Dea

ths

NYT

1

10

100

1000

10000

Apr May Jundate

Dea

ths

JHU

1

10

100

1000

Apr May Jundate

Dea

ths

CTP







1e+01

1e+03

1e+05

Apr May Jundate

Test

s


1e+01

1e+03

1e+05

Apr May Jundate

Test

s



























o reuse allowed w

ithout permission

(which w


ho has granted medR



ay 30 2020

httpsdoiorg1011012020052720115139doi

medR

xiv preprint










1 Introduction








32 Counterfactuals


41 Data







53 Mask Mandate

6 Conclusion

References



A2 Deaths

A3 Tests


A5 Policy Variables

A6 Timing




(I2) DynEt =sumt`


(I3) DynEt =sumt`


+sumt`


+sumt`minus1


)



P itTt=1 i = 1 N



Y it =Yit(B

it P

it I

it)

Bit =Bit(P

it I

it)


lowasti(tminus1) t)

(CEF-SEM)














PY it minus PYit

Predicted CF Change




= PEBit + PED

it + DynEit

(11)



(I2) DynEit =sumt`


i(tminusm`))

(I3) DynEit =sumt`


+sumt`



)+sumt`minus1



)
























resi

den

tial

tran

sit

wor

kp

lace

s

reta

il

groce

ry

clos

edK

-12

sch

ool

s

stay

ath

ome

clos

edm

ovie

thea

ters

clos

edre

stau

rants

clos

edb

usi

nes

ses

mas

ks

for

emp

loye

es

stat

eof

emer

gen

cy











CA

US

AL

IMP

AC

TO

FM

AS

KS

P

OL

ICIE

S

BE

HA

VIO

R17


Dependent variable


(1) (2) (3) (4) (5) (6) (7) (8)













(2074) (3061) (1685) (3793)


j Policyj -2698 -3822 -3190 -3580 -2135 -3200 -2792 -3090






o reuse allowed w

ithout permission

(which w


ho has granted medR



ay 30 2020

httpsdoiorg1011012020052720115139doi

medR

xiv preprint








∆Cit

+εyit (12)

where Bit = (B1it B













Dependent variable


(1) (2) (3) (4)


(0141) (0137)


(0050) (0055) (0045) (0048)



(0068) (0077)


(0056) (0061)



(0049) (0057)




(0003) (0003) (0003) (0003)lag(transit 14) 0002 0002 0004 0005

(0003) (0003) (0004) (0004)


(0020) (0022) (0019) (0019)



(1562) (1504) (1268) (1379)


(0049) (0049)


(0043) (0042)∆D∆C


(0103) (0094)


















masks -004 -009(006) (005)sum



(015) (019)














Dependent variable


(1) (2) (3) (4)


(0145) (0150)


(0095) (0067) (0063) (0071)



(0067) (0076)




(0049) (0058)


(0029) (0024) (0023) (0022)




(0061) (0048)


(0042) (0036)∆D∆C


(0112) (0087)


(018) (020)

Observations 3433 3433 3427 3427R2 0666 0707 0730 0757

Adjusted R2 0665 0705 0729 0755












closed K-12 schools -013 -019 -032 -052 019(013) (009) (014) (014) (007)

stay at home -005 -010 -015 -020 004



(010) (004) (009) (009) (004)


all -021 -042 -063 -101 038(022) (018) (020) (023) (018)






















0

2

4

Mar Apr Maydate

∆log

∆C

Estimated

Observed

Without Policies

∆log∆C

Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C



anda

te fa

ce m

ask

use

by e

mpl

oyee

s in

pub

lic fa

cing

bus

ines

ses

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

0

500000

1000000

1500000

2000000

2500000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek








Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

0e+00

2e+06

4e+06

6e+06

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

0e+00

1e+06

2e+06

3e+06

4e+06

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek






00

02

04

06

08

Mar Apr Maydate

Cha

nge

in ∆

log∆

C



00e+00

25e+07

50e+07

75e+07

10e+08

Mar Apr Maydate

case

s in

the

past

wee

k


0

50

100

150

200

Mar Apr Maydate

rela

tive

incr

ease

in c

ases







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus2000

0

2000

4000

6000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


anda

te fa

ce m

ask

use

by e

mpl

oyee

s in

pub

lic fa

cing

bus

ines

ses

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus5000

0

5000

10000

15000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

0

4000

8000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



minus002

000

002

004

006

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


minus01

00

01

02

03

04

05

Mar Apr Maydate

rela

tive

incr

ease

in c

ases







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus03

minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus5000

minus4000

minus3000

minus2000

minus1000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus03

minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus20000

minus15000

minus10000

minus5000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek






Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus02

minus01

00

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus10000

minus5000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


minus06

minus04

minus02

00

Mar Apr Maydate

rela

tive

incr

ease

in c

ases






6 Conclusion






References


















































1 2 3NYT 100 029 038JHU 029 100 033CTP 038 033 100












1

10

100

1000

10000

Apr May Jundate

New

cas

es

NYT

1

10

100

1000

10000

Apr May Jundate

New

cas

es

JHU

1

10

100

1000

10000

Apr May Jundate

New

cas

es

CTP













1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

NYT

1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

JHU

1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

CTP



1

10

100

1000

10000

Apr May Jundate

Dea

ths

NYT

1

10

100

1000

10000

Apr May Jundate

Dea

ths

JHU

1

10

100

1000

Apr May Jundate

Dea

ths

CTP







1e+01

1e+03

1e+05

Apr May Jundate

Test

s


1e+01

1e+03

1e+05

Apr May Jundate

Test

s



























o reuse allowed w

ithout permission

(which w


ho has granted medR



ay 30 2020

httpsdoiorg1011012020052720115139doi

medR

xiv preprint










1 Introduction








32 Counterfactuals


41 Data







53 Mask Mandate

6 Conclusion

References



A2 Deaths

A3 Tests


A5 Policy Variables

A6 Timing



PY it minus PYit

Predicted CF Change




= PEBit + PED

it + DynEit

(11)



(I2) DynEit =sumt`


i(tminusm`))

(I3) DynEit =sumt`


+sumt`



)+sumt`minus1



)
























resi

den

tial

tran

sit

wor

kp

lace

s

reta

il

groce

ry

clos

edK

-12

sch

ool

s

stay

ath

ome

clos

edm

ovie

thea

ters

clos

edre

stau

rants

clos

edb

usi

nes

ses

mas

ks

for

emp

loye

es

stat

eof

emer

gen

cy











CA

US

AL

IMP

AC

TO

FM

AS

KS

P

OL

ICIE

S

BE

HA

VIO

R17


Dependent variable


(1) (2) (3) (4) (5) (6) (7) (8)













(2074) (3061) (1685) (3793)


j Policyj -2698 -3822 -3190 -3580 -2135 -3200 -2792 -3090






o reuse allowed w

ithout permission

(which w


ho has granted medR



ay 30 2020

httpsdoiorg1011012020052720115139doi

medR

xiv preprint








∆Cit

+εyit (12)

where Bit = (B1it B













Dependent variable


(1) (2) (3) (4)


(0141) (0137)


(0050) (0055) (0045) (0048)



(0068) (0077)


(0056) (0061)



(0049) (0057)




(0003) (0003) (0003) (0003)lag(transit 14) 0002 0002 0004 0005

(0003) (0003) (0004) (0004)


(0020) (0022) (0019) (0019)



(1562) (1504) (1268) (1379)


(0049) (0049)


(0043) (0042)∆D∆C


(0103) (0094)


















masks -004 -009(006) (005)sum



(015) (019)














Dependent variable


(1) (2) (3) (4)


(0145) (0150)


(0095) (0067) (0063) (0071)



(0067) (0076)




(0049) (0058)


(0029) (0024) (0023) (0022)




(0061) (0048)


(0042) (0036)∆D∆C


(0112) (0087)


(018) (020)

Observations 3433 3433 3427 3427R2 0666 0707 0730 0757

Adjusted R2 0665 0705 0729 0755












closed K-12 schools -013 -019 -032 -052 019(013) (009) (014) (014) (007)

stay at home -005 -010 -015 -020 004



(010) (004) (009) (009) (004)


all -021 -042 -063 -101 038(022) (018) (020) (023) (018)






















0

2

4

Mar Apr Maydate

∆log

∆C

Estimated

Observed

Without Policies

∆log∆C

Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C



anda

te fa

ce m

ask

use

by e

mpl

oyee

s in

pub

lic fa

cing

bus

ines

ses

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

0

500000

1000000

1500000

2000000

2500000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek








Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

0e+00

2e+06

4e+06

6e+06

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

0e+00

1e+06

2e+06

3e+06

4e+06

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek






00

02

04

06

08

Mar Apr Maydate

Cha

nge

in ∆

log∆

C



00e+00

25e+07

50e+07

75e+07

10e+08

Mar Apr Maydate

case

s in

the

past

wee

k


0

50

100

150

200

Mar Apr Maydate

rela

tive

incr

ease

in c

ases







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus2000

0

2000

4000

6000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


anda

te fa

ce m

ask

use

by e

mpl

oyee

s in

pub

lic fa

cing

bus

ines

ses

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus5000

0

5000

10000

15000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

0

4000

8000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



minus002

000

002

004

006

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


minus01

00

01

02

03

04

05

Mar Apr Maydate

rela

tive

incr

ease

in c

ases







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus03

minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus5000

minus4000

minus3000

minus2000

minus1000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus03

minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus20000

minus15000

minus10000

minus5000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek






Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus02

minus01

00

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus10000

minus5000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


minus06

minus04

minus02

00

Mar Apr Maydate

rela

tive

incr

ease

in c

ases






6 Conclusion






References


















































1 2 3NYT 100 029 038JHU 029 100 033CTP 038 033 100












1

10

100

1000

10000

Apr May Jundate

New

cas

es

NYT

1

10

100

1000

10000

Apr May Jundate

New

cas

es

JHU

1

10

100

1000

10000

Apr May Jundate

New

cas

es

CTP













1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

NYT

1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

JHU

1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

CTP



1

10

100

1000

10000

Apr May Jundate

Dea

ths

NYT

1

10

100

1000

10000

Apr May Jundate

Dea

ths

JHU

1

10

100

1000

Apr May Jundate

Dea

ths

CTP







1e+01

1e+03

1e+05

Apr May Jundate

Test

s


1e+01

1e+03

1e+05

Apr May Jundate

Test

s



























o reuse allowed w

ithout permission

(which w


ho has granted medR



ay 30 2020

httpsdoiorg1011012020052720115139doi

medR

xiv preprint










1 Introduction








32 Counterfactuals


41 Data







53 Mask Mandate

6 Conclusion

References



A2 Deaths

A3 Tests


A5 Policy Variables

A6 Timing


















resi

den

tial

tran

sit

wor

kp

lace

s

reta

il

groce

ry

clos

edK

-12

sch

ool

s

stay

ath

ome

clos

edm

ovie

thea

ters

clos

edre

stau

rants

clos

edb

usi

nes

ses

mas

ks

for

emp

loye

es

stat

eof

emer

gen

cy











CA

US

AL

IMP

AC

TO

FM

AS

KS

P

OL

ICIE

S

BE

HA

VIO

R17


Dependent variable


(1) (2) (3) (4) (5) (6) (7) (8)













(2074) (3061) (1685) (3793)


j Policyj -2698 -3822 -3190 -3580 -2135 -3200 -2792 -3090






o reuse allowed w

ithout permission

(which w


ho has granted medR



ay 30 2020

httpsdoiorg1011012020052720115139doi

medR

xiv preprint








∆Cit

+εyit (12)

where Bit = (B1it B













Dependent variable


(1) (2) (3) (4)


(0141) (0137)


(0050) (0055) (0045) (0048)



(0068) (0077)


(0056) (0061)



(0049) (0057)




(0003) (0003) (0003) (0003)lag(transit 14) 0002 0002 0004 0005

(0003) (0003) (0004) (0004)


(0020) (0022) (0019) (0019)



(1562) (1504) (1268) (1379)


(0049) (0049)


(0043) (0042)∆D∆C


(0103) (0094)


















masks -004 -009(006) (005)sum



(015) (019)














Dependent variable


(1) (2) (3) (4)


(0145) (0150)


(0095) (0067) (0063) (0071)



(0067) (0076)




(0049) (0058)


(0029) (0024) (0023) (0022)




(0061) (0048)


(0042) (0036)∆D∆C


(0112) (0087)


(018) (020)

Observations 3433 3433 3427 3427R2 0666 0707 0730 0757

Adjusted R2 0665 0705 0729 0755












closed K-12 schools -013 -019 -032 -052 019(013) (009) (014) (014) (007)

stay at home -005 -010 -015 -020 004



(010) (004) (009) (009) (004)


all -021 -042 -063 -101 038(022) (018) (020) (023) (018)






















0

2

4

Mar Apr Maydate

∆log

∆C

Estimated

Observed

Without Policies

∆log∆C

Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C



anda

te fa

ce m

ask

use

by e

mpl

oyee

s in

pub

lic fa

cing

bus

ines

ses

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

0

500000

1000000

1500000

2000000

2500000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek








Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

0e+00

2e+06

4e+06

6e+06

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

0e+00

1e+06

2e+06

3e+06

4e+06

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek






00

02

04

06

08

Mar Apr Maydate

Cha

nge

in ∆

log∆

C



00e+00

25e+07

50e+07

75e+07

10e+08

Mar Apr Maydate

case

s in

the

past

wee

k


0

50

100

150

200

Mar Apr Maydate

rela

tive

incr

ease

in c

ases







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus2000

0

2000

4000

6000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


anda

te fa

ce m

ask

use

by e

mpl

oyee

s in

pub

lic fa

cing

bus

ines

ses

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus5000

0

5000

10000

15000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

0

4000

8000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



minus002

000

002

004

006

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


minus01

00

01

02

03

04

05

Mar Apr Maydate

rela

tive

incr

ease

in c

ases







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus03

minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus5000

minus4000

minus3000

minus2000

minus1000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus03

minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus20000

minus15000

minus10000

minus5000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek






Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus02

minus01

00

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus10000

minus5000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


minus06

minus04

minus02

00

Mar Apr Maydate

rela

tive

incr

ease

in c

ases






6 Conclusion






References


















































1 2 3NYT 100 029 038JHU 029 100 033CTP 038 033 100












1

10

100

1000

10000

Apr May Jundate

New

cas

es

NYT

1

10

100

1000

10000

Apr May Jundate

New

cas

es

JHU

1

10

100

1000

10000

Apr May Jundate

New

cas

es

CTP













1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

NYT

1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

JHU

1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

CTP



1

10

100

1000

10000

Apr May Jundate

Dea

ths

NYT

1

10

100

1000

10000

Apr May Jundate

Dea

ths

JHU

1

10

100

1000

Apr May Jundate

Dea

ths

CTP







1e+01

1e+03

1e+05

Apr May Jundate

Test

s


1e+01

1e+03

1e+05

Apr May Jundate

Test

s



























o reuse allowed w

ithout permission

(which w


ho has granted medR



ay 30 2020

httpsdoiorg1011012020052720115139doi

medR

xiv preprint










1 Introduction








32 Counterfactuals


41 Data







53 Mask Mandate

6 Conclusion

References



A2 Deaths

A3 Tests


A5 Policy Variables

A6 Timing

CA

US

AL

IMP

AC

TO

FM

AS

KS

P

OL

ICIE

S

BE

HA

VIO

R17


Dependent variable


(1) (2) (3) (4) (5) (6) (7) (8)













(2074) (3061) (1685) (3793)


j Policyj -2698 -3822 -3190 -3580 -2135 -3200 -2792 -3090






o reuse allowed w

ithout permission

(which w


ho has granted medR



ay 30 2020

httpsdoiorg1011012020052720115139doi

medR

xiv preprint








∆Cit

+εyit (12)

where Bit = (B1it B













Dependent variable


(1) (2) (3) (4)


(0141) (0137)


(0050) (0055) (0045) (0048)



(0068) (0077)


(0056) (0061)



(0049) (0057)




(0003) (0003) (0003) (0003)lag(transit 14) 0002 0002 0004 0005

(0003) (0003) (0004) (0004)


(0020) (0022) (0019) (0019)



(1562) (1504) (1268) (1379)


(0049) (0049)


(0043) (0042)∆D∆C


(0103) (0094)


















masks -004 -009(006) (005)sum



(015) (019)














Dependent variable


(1) (2) (3) (4)


(0145) (0150)


(0095) (0067) (0063) (0071)



(0067) (0076)




(0049) (0058)


(0029) (0024) (0023) (0022)




(0061) (0048)


(0042) (0036)∆D∆C


(0112) (0087)


(018) (020)

Observations 3433 3433 3427 3427R2 0666 0707 0730 0757

Adjusted R2 0665 0705 0729 0755












closed K-12 schools -013 -019 -032 -052 019(013) (009) (014) (014) (007)

stay at home -005 -010 -015 -020 004



(010) (004) (009) (009) (004)


all -021 -042 -063 -101 038(022) (018) (020) (023) (018)






















0

2

4

Mar Apr Maydate

∆log

∆C

Estimated

Observed

Without Policies

∆log∆C

Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C



anda

te fa

ce m

ask

use

by e

mpl

oyee

s in

pub

lic fa

cing

bus

ines

ses

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

0

500000

1000000

1500000

2000000

2500000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek








Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

0e+00

2e+06

4e+06

6e+06

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

0e+00

1e+06

2e+06

3e+06

4e+06

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek






00

02

04

06

08

Mar Apr Maydate

Cha

nge

in ∆

log∆

C



00e+00

25e+07

50e+07

75e+07

10e+08

Mar Apr Maydate

case

s in

the

past

wee

k


0

50

100

150

200

Mar Apr Maydate

rela

tive

incr

ease

in c

ases







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus2000

0

2000

4000

6000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


anda

te fa

ce m

ask

use

by e

mpl

oyee

s in

pub

lic fa

cing

bus

ines

ses

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus5000

0

5000

10000

15000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

0

4000

8000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



minus002

000

002

004

006

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


minus01

00

01

02

03

04

05

Mar Apr Maydate

rela

tive

incr

ease

in c

ases







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus03

minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus5000

minus4000

minus3000

minus2000

minus1000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus03

minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus20000

minus15000

minus10000

minus5000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek






Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus02

minus01

00

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus10000

minus5000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


minus06

minus04

minus02

00

Mar Apr Maydate

rela

tive

incr

ease

in c

ases






6 Conclusion






References


















































1 2 3NYT 100 029 038JHU 029 100 033CTP 038 033 100












1

10

100

1000

10000

Apr May Jundate

New

cas

es

NYT

1

10

100

1000

10000

Apr May Jundate

New

cas

es

JHU

1

10

100

1000

10000

Apr May Jundate

New

cas

es

CTP













1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

NYT

1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

JHU

1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

CTP



1

10

100

1000

10000

Apr May Jundate

Dea

ths

NYT

1

10

100

1000

10000

Apr May Jundate

Dea

ths

JHU

1

10

100

1000

Apr May Jundate

Dea

ths

CTP







1e+01

1e+03

1e+05

Apr May Jundate

Test

s


1e+01

1e+03

1e+05

Apr May Jundate

Test

s



























o reuse allowed w

ithout permission

(which w


ho has granted medR



ay 30 2020

httpsdoiorg1011012020052720115139doi

medR

xiv preprint










1 Introduction








32 Counterfactuals


41 Data







53 Mask Mandate

6 Conclusion

References



A2 Deaths

A3 Tests


A5 Policy Variables

A6 Timing








∆Cit

+εyit (12)

where Bit = (B1it B













Dependent variable


(1) (2) (3) (4)


(0141) (0137)


(0050) (0055) (0045) (0048)



(0068) (0077)


(0056) (0061)



(0049) (0057)




(0003) (0003) (0003) (0003)lag(transit 14) 0002 0002 0004 0005

(0003) (0003) (0004) (0004)


(0020) (0022) (0019) (0019)



(1562) (1504) (1268) (1379)


(0049) (0049)


(0043) (0042)∆D∆C


(0103) (0094)


















masks -004 -009(006) (005)sum



(015) (019)














Dependent variable


(1) (2) (3) (4)


(0145) (0150)


(0095) (0067) (0063) (0071)



(0067) (0076)




(0049) (0058)


(0029) (0024) (0023) (0022)




(0061) (0048)


(0042) (0036)∆D∆C


(0112) (0087)


(018) (020)

Observations 3433 3433 3427 3427R2 0666 0707 0730 0757

Adjusted R2 0665 0705 0729 0755












closed K-12 schools -013 -019 -032 -052 019(013) (009) (014) (014) (007)

stay at home -005 -010 -015 -020 004



(010) (004) (009) (009) (004)


all -021 -042 -063 -101 038(022) (018) (020) (023) (018)






















0

2

4

Mar Apr Maydate

∆log

∆C

Estimated

Observed

Without Policies

∆log∆C

Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C



anda

te fa

ce m

ask

use

by e

mpl

oyee

s in

pub

lic fa

cing

bus

ines

ses

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

0

500000

1000000

1500000

2000000

2500000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek








Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

0e+00

2e+06

4e+06

6e+06

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

0e+00

1e+06

2e+06

3e+06

4e+06

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek






00

02

04

06

08

Mar Apr Maydate

Cha

nge

in ∆

log∆

C



00e+00

25e+07

50e+07

75e+07

10e+08

Mar Apr Maydate

case

s in

the

past

wee

k


0

50

100

150

200

Mar Apr Maydate

rela

tive

incr

ease

in c

ases







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus2000

0

2000

4000

6000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


anda

te fa

ce m

ask

use

by e

mpl

oyee

s in

pub

lic fa

cing

bus

ines

ses

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus5000

0

5000

10000

15000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

0

4000

8000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



minus002

000

002

004

006

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


minus01

00

01

02

03

04

05

Mar Apr Maydate

rela

tive

incr

ease

in c

ases







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus03

minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus5000

minus4000

minus3000

minus2000

minus1000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus03

minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus20000

minus15000

minus10000

minus5000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek






Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus02

minus01

00

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus10000

minus5000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


minus06

minus04

minus02

00

Mar Apr Maydate

rela

tive

incr

ease

in c

ases






6 Conclusion






References


















































1 2 3NYT 100 029 038JHU 029 100 033CTP 038 033 100












1

10

100

1000

10000

Apr May Jundate

New

cas

es

NYT

1

10

100

1000

10000

Apr May Jundate

New

cas

es

JHU

1

10

100

1000

10000

Apr May Jundate

New

cas

es

CTP













1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

NYT

1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

JHU

1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

CTP



1

10

100

1000

10000

Apr May Jundate

Dea

ths

NYT

1

10

100

1000

10000

Apr May Jundate

Dea

ths

JHU

1

10

100

1000

Apr May Jundate

Dea

ths

CTP







1e+01

1e+03

1e+05

Apr May Jundate

Test

s


1e+01

1e+03

1e+05

Apr May Jundate

Test

s



























o reuse allowed w

ithout permission

(which w


ho has granted medR



ay 30 2020

httpsdoiorg1011012020052720115139doi

medR

xiv preprint










1 Introduction








32 Counterfactuals


41 Data







53 Mask Mandate

6 Conclusion

References



A2 Deaths

A3 Tests


A5 Policy Variables

A6 Timing



Dependent variable


(1) (2) (3) (4)


(0141) (0137)


(0050) (0055) (0045) (0048)



(0068) (0077)


(0056) (0061)



(0049) (0057)




(0003) (0003) (0003) (0003)lag(transit 14) 0002 0002 0004 0005

(0003) (0003) (0004) (0004)


(0020) (0022) (0019) (0019)



(1562) (1504) (1268) (1379)


(0049) (0049)


(0043) (0042)∆D∆C


(0103) (0094)


















masks -004 -009(006) (005)sum



(015) (019)














Dependent variable


(1) (2) (3) (4)


(0145) (0150)


(0095) (0067) (0063) (0071)



(0067) (0076)




(0049) (0058)


(0029) (0024) (0023) (0022)




(0061) (0048)


(0042) (0036)∆D∆C


(0112) (0087)


(018) (020)

Observations 3433 3433 3427 3427R2 0666 0707 0730 0757

Adjusted R2 0665 0705 0729 0755












closed K-12 schools -013 -019 -032 -052 019(013) (009) (014) (014) (007)

stay at home -005 -010 -015 -020 004



(010) (004) (009) (009) (004)


all -021 -042 -063 -101 038(022) (018) (020) (023) (018)






















0

2

4

Mar Apr Maydate

∆log

∆C

Estimated

Observed

Without Policies

∆log∆C

Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C



anda

te fa

ce m

ask

use

by e

mpl

oyee

s in

pub

lic fa

cing

bus

ines

ses

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

0

500000

1000000

1500000

2000000

2500000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek








Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

0e+00

2e+06

4e+06

6e+06

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

0e+00

1e+06

2e+06

3e+06

4e+06

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek






00

02

04

06

08

Mar Apr Maydate

Cha

nge

in ∆

log∆

C



00e+00

25e+07

50e+07

75e+07

10e+08

Mar Apr Maydate

case

s in

the

past

wee

k


0

50

100

150

200

Mar Apr Maydate

rela

tive

incr

ease

in c

ases







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus2000

0

2000

4000

6000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


anda

te fa

ce m

ask

use

by e

mpl

oyee

s in

pub

lic fa

cing

bus

ines

ses

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus5000

0

5000

10000

15000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

0

4000

8000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



minus002

000

002

004

006

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


minus01

00

01

02

03

04

05

Mar Apr Maydate

rela

tive

incr

ease

in c

ases







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus03

minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus5000

minus4000

minus3000

minus2000

minus1000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus03

minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus20000

minus15000

minus10000

minus5000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek






Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus02

minus01

00

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus10000

minus5000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


minus06

minus04

minus02

00

Mar Apr Maydate

rela

tive

incr

ease

in c

ases






6 Conclusion






References


















































1 2 3NYT 100 029 038JHU 029 100 033CTP 038 033 100












1

10

100

1000

10000

Apr May Jundate

New

cas

es

NYT

1

10

100

1000

10000

Apr May Jundate

New

cas

es

JHU

1

10

100

1000

10000

Apr May Jundate

New

cas

es

CTP













1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

NYT

1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

JHU

1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

CTP



1

10

100

1000

10000

Apr May Jundate

Dea

ths

NYT

1

10

100

1000

10000

Apr May Jundate

Dea

ths

JHU

1

10

100

1000

Apr May Jundate

Dea

ths

CTP







1e+01

1e+03

1e+05

Apr May Jundate

Test

s


1e+01

1e+03

1e+05

Apr May Jundate

Test

s



























o reuse allowed w

ithout permission

(which w


ho has granted medR



ay 30 2020

httpsdoiorg1011012020052720115139doi

medR

xiv preprint










1 Introduction








32 Counterfactuals


41 Data







53 Mask Mandate

6 Conclusion

References



A2 Deaths

A3 Tests


A5 Policy Variables

A6 Timing




masks -004 -009(006) (005)sum



(015) (019)














Dependent variable


(1) (2) (3) (4)


(0145) (0150)


(0095) (0067) (0063) (0071)



(0067) (0076)




(0049) (0058)


(0029) (0024) (0023) (0022)




(0061) (0048)


(0042) (0036)∆D∆C


(0112) (0087)


(018) (020)

Observations 3433 3433 3427 3427R2 0666 0707 0730 0757

Adjusted R2 0665 0705 0729 0755












closed K-12 schools -013 -019 -032 -052 019(013) (009) (014) (014) (007)

stay at home -005 -010 -015 -020 004



(010) (004) (009) (009) (004)


all -021 -042 -063 -101 038(022) (018) (020) (023) (018)






















0

2

4

Mar Apr Maydate

∆log

∆C

Estimated

Observed

Without Policies

∆log∆C

Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C



anda

te fa

ce m

ask

use

by e

mpl

oyee

s in

pub

lic fa

cing

bus

ines

ses

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

0

500000

1000000

1500000

2000000

2500000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek








Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

0e+00

2e+06

4e+06

6e+06

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

0e+00

1e+06

2e+06

3e+06

4e+06

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek






00

02

04

06

08

Mar Apr Maydate

Cha

nge

in ∆

log∆

C



00e+00

25e+07

50e+07

75e+07

10e+08

Mar Apr Maydate

case

s in

the

past

wee

k


0

50

100

150

200

Mar Apr Maydate

rela

tive

incr

ease

in c

ases







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus2000

0

2000

4000

6000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


anda

te fa

ce m

ask

use

by e

mpl

oyee

s in

pub

lic fa

cing

bus

ines

ses

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus5000

0

5000

10000

15000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

0

4000

8000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



minus002

000

002

004

006

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


minus01

00

01

02

03

04

05

Mar Apr Maydate

rela

tive

incr

ease

in c

ases







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus03

minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus5000

minus4000

minus3000

minus2000

minus1000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus03

minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus20000

minus15000

minus10000

minus5000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek






Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus02

minus01

00

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus10000

minus5000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


minus06

minus04

minus02

00

Mar Apr Maydate

rela

tive

incr

ease

in c

ases






6 Conclusion






References


















































1 2 3NYT 100 029 038JHU 029 100 033CTP 038 033 100












1

10

100

1000

10000

Apr May Jundate

New

cas

es

NYT

1

10

100

1000

10000

Apr May Jundate

New

cas

es

JHU

1

10

100

1000

10000

Apr May Jundate

New

cas

es

CTP













1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

NYT

1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

JHU

1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

CTP



1

10

100

1000

10000

Apr May Jundate

Dea

ths

NYT

1

10

100

1000

10000

Apr May Jundate

Dea

ths

JHU

1

10

100

1000

Apr May Jundate

Dea

ths

CTP







1e+01

1e+03

1e+05

Apr May Jundate

Test

s


1e+01

1e+03

1e+05

Apr May Jundate

Test

s



























o reuse allowed w

ithout permission

(which w


ho has granted medR



ay 30 2020

httpsdoiorg1011012020052720115139doi

medR

xiv preprint










1 Introduction








32 Counterfactuals


41 Data







53 Mask Mandate

6 Conclusion

References



A2 Deaths

A3 Tests


A5 Policy Variables

A6 Timing



Dependent variable


(1) (2) (3) (4)


(0145) (0150)


(0095) (0067) (0063) (0071)



(0067) (0076)




(0049) (0058)


(0029) (0024) (0023) (0022)




(0061) (0048)


(0042) (0036)∆D∆C


(0112) (0087)


(018) (020)

Observations 3433 3433 3427 3427R2 0666 0707 0730 0757

Adjusted R2 0665 0705 0729 0755












closed K-12 schools -013 -019 -032 -052 019(013) (009) (014) (014) (007)

stay at home -005 -010 -015 -020 004



(010) (004) (009) (009) (004)


all -021 -042 -063 -101 038(022) (018) (020) (023) (018)






















0

2

4

Mar Apr Maydate

∆log

∆C

Estimated

Observed

Without Policies

∆log∆C

Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C



anda

te fa

ce m

ask

use

by e

mpl

oyee

s in

pub

lic fa

cing

bus

ines

ses

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

0

500000

1000000

1500000

2000000

2500000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek








Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

0e+00

2e+06

4e+06

6e+06

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

0e+00

1e+06

2e+06

3e+06

4e+06

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek






00

02

04

06

08

Mar Apr Maydate

Cha

nge

in ∆

log∆

C



00e+00

25e+07

50e+07

75e+07

10e+08

Mar Apr Maydate

case

s in

the

past

wee

k


0

50

100

150

200

Mar Apr Maydate

rela

tive

incr

ease

in c

ases







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus2000

0

2000

4000

6000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


anda

te fa

ce m

ask

use

by e

mpl

oyee

s in

pub

lic fa

cing

bus

ines

ses

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus5000

0

5000

10000

15000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

0

4000

8000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



minus002

000

002

004

006

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


minus01

00

01

02

03

04

05

Mar Apr Maydate

rela

tive

incr

ease

in c

ases







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus03

minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus5000

minus4000

minus3000

minus2000

minus1000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus03

minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus20000

minus15000

minus10000

minus5000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek






Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus02

minus01

00

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus10000

minus5000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


minus06

minus04

minus02

00

Mar Apr Maydate

rela

tive

incr

ease

in c

ases






6 Conclusion






References


















































1 2 3NYT 100 029 038JHU 029 100 033CTP 038 033 100












1

10

100

1000

10000

Apr May Jundate

New

cas

es

NYT

1

10

100

1000

10000

Apr May Jundate

New

cas

es

JHU

1

10

100

1000

10000

Apr May Jundate

New

cas

es

CTP













1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

NYT

1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

JHU

1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

CTP



1

10

100

1000

10000

Apr May Jundate

Dea

ths

NYT

1

10

100

1000

10000

Apr May Jundate

Dea

ths

JHU

1

10

100

1000

Apr May Jundate

Dea

ths

CTP







1e+01

1e+03

1e+05

Apr May Jundate

Test

s


1e+01

1e+03

1e+05

Apr May Jundate

Test

s



























o reuse allowed w

ithout permission

(which w


ho has granted medR



ay 30 2020

httpsdoiorg1011012020052720115139doi

medR

xiv preprint










1 Introduction








32 Counterfactuals


41 Data







53 Mask Mandate

6 Conclusion

References



A2 Deaths

A3 Tests


A5 Policy Variables

A6 Timing





closed K-12 schools -013 -019 -032 -052 019(013) (009) (014) (014) (007)

stay at home -005 -010 -015 -020 004



(010) (004) (009) (009) (004)


all -021 -042 -063 -101 038(022) (018) (020) (023) (018)






















0

2

4

Mar Apr Maydate

∆log

∆C

Estimated

Observed

Without Policies

∆log∆C

Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C



anda

te fa

ce m

ask

use

by e

mpl

oyee

s in

pub

lic fa

cing

bus

ines

ses

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

0

500000

1000000

1500000

2000000

2500000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek








Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

0e+00

2e+06

4e+06

6e+06

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

0e+00

1e+06

2e+06

3e+06

4e+06

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek






00

02

04

06

08

Mar Apr Maydate

Cha

nge

in ∆

log∆

C



00e+00

25e+07

50e+07

75e+07

10e+08

Mar Apr Maydate

case

s in

the

past

wee

k


0

50

100

150

200

Mar Apr Maydate

rela

tive

incr

ease

in c

ases







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus2000

0

2000

4000

6000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


anda

te fa

ce m

ask

use

by e

mpl

oyee

s in

pub

lic fa

cing

bus

ines

ses

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus5000

0

5000

10000

15000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

0

4000

8000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



minus002

000

002

004

006

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


minus01

00

01

02

03

04

05

Mar Apr Maydate

rela

tive

incr

ease

in c

ases







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus03

minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus5000

minus4000

minus3000

minus2000

minus1000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus03

minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus20000

minus15000

minus10000

minus5000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek






Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus02

minus01

00

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus10000

minus5000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


minus06

minus04

minus02

00

Mar Apr Maydate

rela

tive

incr

ease

in c

ases






6 Conclusion






References


















































1 2 3NYT 100 029 038JHU 029 100 033CTP 038 033 100












1

10

100

1000

10000

Apr May Jundate

New

cas

es

NYT

1

10

100

1000

10000

Apr May Jundate

New

cas

es

JHU

1

10

100

1000

10000

Apr May Jundate

New

cas

es

CTP













1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

NYT

1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

JHU

1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

CTP



1

10

100

1000

10000

Apr May Jundate

Dea

ths

NYT

1

10

100

1000

10000

Apr May Jundate

Dea

ths

JHU

1

10

100

1000

Apr May Jundate

Dea

ths

CTP







1e+01

1e+03

1e+05

Apr May Jundate

Test

s


1e+01

1e+03

1e+05

Apr May Jundate

Test

s



























o reuse allowed w

ithout permission

(which w


ho has granted medR



ay 30 2020

httpsdoiorg1011012020052720115139doi

medR

xiv preprint










1 Introduction








32 Counterfactuals


41 Data







53 Mask Mandate

6 Conclusion

References



A2 Deaths

A3 Tests


A5 Policy Variables

A6 Timing













0

2

4

Mar Apr Maydate

∆log

∆C

Estimated

Observed

Without Policies

∆log∆C

Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C



anda

te fa

ce m

ask

use

by e

mpl

oyee

s in

pub

lic fa

cing

bus

ines

ses

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

0

500000

1000000

1500000

2000000

2500000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek








Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

0e+00

2e+06

4e+06

6e+06

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

0e+00

1e+06

2e+06

3e+06

4e+06

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek






00

02

04

06

08

Mar Apr Maydate

Cha

nge

in ∆

log∆

C



00e+00

25e+07

50e+07

75e+07

10e+08

Mar Apr Maydate

case

s in

the

past

wee

k


0

50

100

150

200

Mar Apr Maydate

rela

tive

incr

ease

in c

ases







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus2000

0

2000

4000

6000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


anda

te fa

ce m

ask

use

by e

mpl

oyee

s in

pub

lic fa

cing

bus

ines

ses

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus5000

0

5000

10000

15000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

0

4000

8000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



minus002

000

002

004

006

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


minus01

00

01

02

03

04

05

Mar Apr Maydate

rela

tive

incr

ease

in c

ases







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus03

minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus5000

minus4000

minus3000

minus2000

minus1000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus03

minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus20000

minus15000

minus10000

minus5000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek






Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus02

minus01

00

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus10000

minus5000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


minus06

minus04

minus02

00

Mar Apr Maydate

rela

tive

incr

ease

in c

ases






6 Conclusion






References


















































1 2 3NYT 100 029 038JHU 029 100 033CTP 038 033 100












1

10

100

1000

10000

Apr May Jundate

New

cas

es

NYT

1

10

100

1000

10000

Apr May Jundate

New

cas

es

JHU

1

10

100

1000

10000

Apr May Jundate

New

cas

es

CTP













1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

NYT

1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

JHU

1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

CTP



1

10

100

1000

10000

Apr May Jundate

Dea

ths

NYT

1

10

100

1000

10000

Apr May Jundate

Dea

ths

JHU

1

10

100

1000

Apr May Jundate

Dea

ths

CTP







1e+01

1e+03

1e+05

Apr May Jundate

Test

s


1e+01

1e+03

1e+05

Apr May Jundate

Test

s



























o reuse allowed w

ithout permission

(which w


ho has granted medR



ay 30 2020

httpsdoiorg1011012020052720115139doi

medR

xiv preprint










1 Introduction








32 Counterfactuals


41 Data







53 Mask Mandate

6 Conclusion

References



A2 Deaths

A3 Tests


A5 Policy Variables

A6 Timing



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

0e+00

2e+06

4e+06

6e+06

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

000

025

050

075

100

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

0e+00

1e+06

2e+06

3e+06

4e+06

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek






00

02

04

06

08

Mar Apr Maydate

Cha

nge

in ∆

log∆

C



00e+00

25e+07

50e+07

75e+07

10e+08

Mar Apr Maydate

case

s in

the

past

wee

k


0

50

100

150

200

Mar Apr Maydate

rela

tive

incr

ease

in c

ases







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus2000

0

2000

4000

6000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


anda

te fa

ce m

ask

use

by e

mpl

oyee

s in

pub

lic fa

cing

bus

ines

ses

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus5000

0

5000

10000

15000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

0

4000

8000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



minus002

000

002

004

006

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


minus01

00

01

02

03

04

05

Mar Apr Maydate

rela

tive

incr

ease

in c

ases







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus03

minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus5000

minus4000

minus3000

minus2000

minus1000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus03

minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus20000

minus15000

minus10000

minus5000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek






Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus02

minus01

00

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus10000

minus5000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


minus06

minus04

minus02

00

Mar Apr Maydate

rela

tive

incr

ease

in c

ases






6 Conclusion






References


















































1 2 3NYT 100 029 038JHU 029 100 033CTP 038 033 100












1

10

100

1000

10000

Apr May Jundate

New

cas

es

NYT

1

10

100

1000

10000

Apr May Jundate

New

cas

es

JHU

1

10

100

1000

10000

Apr May Jundate

New

cas

es

CTP













1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

NYT

1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

JHU

1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

CTP



1

10

100

1000

10000

Apr May Jundate

Dea

ths

NYT

1

10

100

1000

10000

Apr May Jundate

Dea

ths

JHU

1

10

100

1000

Apr May Jundate

Dea

ths

CTP







1e+01

1e+03

1e+05

Apr May Jundate

Test

s


1e+01

1e+03

1e+05

Apr May Jundate

Test

s



























o reuse allowed w

ithout permission

(which w


ho has granted medR



ay 30 2020

httpsdoiorg1011012020052720115139doi

medR

xiv preprint










1 Introduction








32 Counterfactuals


41 Data







53 Mask Mandate

6 Conclusion

References



A2 Deaths

A3 Tests


A5 Policy Variables

A6 Timing



00

02

04

06

08

Mar Apr Maydate

Cha

nge

in ∆

log∆

C



00e+00

25e+07

50e+07

75e+07

10e+08

Mar Apr Maydate

case

s in

the

past

wee

k


0

50

100

150

200

Mar Apr Maydate

rela

tive

incr

ease

in c

ases







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus2000

0

2000

4000

6000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


anda

te fa

ce m

ask

use

by e

mpl

oyee

s in

pub

lic fa

cing

bus

ines

ses

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus5000

0

5000

10000

15000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

0

4000

8000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



minus002

000

002

004

006

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


minus01

00

01

02

03

04

05

Mar Apr Maydate

rela

tive

incr

ease

in c

ases







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus03

minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus5000

minus4000

minus3000

minus2000

minus1000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus03

minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus20000

minus15000

minus10000

minus5000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek






Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus02

minus01

00

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus10000

minus5000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


minus06

minus04

minus02

00

Mar Apr Maydate

rela

tive

incr

ease

in c

ases






6 Conclusion






References


















































1 2 3NYT 100 029 038JHU 029 100 033CTP 038 033 100












1

10

100

1000

10000

Apr May Jundate

New

cas

es

NYT

1

10

100

1000

10000

Apr May Jundate

New

cas

es

JHU

1

10

100

1000

10000

Apr May Jundate

New

cas

es

CTP













1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

NYT

1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

JHU

1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

CTP



1

10

100

1000

10000

Apr May Jundate

Dea

ths

NYT

1

10

100

1000

10000

Apr May Jundate

Dea

ths

JHU

1

10

100

1000

Apr May Jundate

Dea

ths

CTP







1e+01

1e+03

1e+05

Apr May Jundate

Test

s


1e+01

1e+03

1e+05

Apr May Jundate

Test

s



























o reuse allowed w

ithout permission

(which w


ho has granted medR



ay 30 2020

httpsdoiorg1011012020052720115139doi

medR

xiv preprint










1 Introduction








32 Counterfactuals


41 Data







53 Mask Mandate

6 Conclusion

References



A2 Deaths

A3 Tests


A5 Policy Variables

A6 Timing




Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus2000

0

2000

4000

6000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


anda

te fa

ce m

ask

use

by e

mpl

oyee

s in

pub

lic fa

cing

bus

ines

ses

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus5000

0

5000

10000

15000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

0

4000

8000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



minus002

000

002

004

006

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


minus01

00

01

02

03

04

05

Mar Apr Maydate

rela

tive

incr

ease

in c

ases







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus03

minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus5000

minus4000

minus3000

minus2000

minus1000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus03

minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus20000

minus15000

minus10000

minus5000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek






Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus02

minus01

00

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus10000

minus5000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


minus06

minus04

minus02

00

Mar Apr Maydate

rela

tive

incr

ease

in c

ases






6 Conclusion






References


















































1 2 3NYT 100 029 038JHU 029 100 033CTP 038 033 100












1

10

100

1000

10000

Apr May Jundate

New

cas

es

NYT

1

10

100

1000

10000

Apr May Jundate

New

cas

es

JHU

1

10

100

1000

10000

Apr May Jundate

New

cas

es

CTP













1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

NYT

1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

JHU

1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

CTP



1

10

100

1000

10000

Apr May Jundate

Dea

ths

NYT

1

10

100

1000

10000

Apr May Jundate

Dea

ths

JHU

1

10

100

1000

Apr May Jundate

Dea

ths

CTP







1e+01

1e+03

1e+05

Apr May Jundate

Test

s


1e+01

1e+03

1e+05

Apr May Jundate

Test

s



























o reuse allowed w

ithout permission

(which w


ho has granted medR



ay 30 2020

httpsdoiorg1011012020052720115139doi

medR

xiv preprint










1 Introduction








32 Counterfactuals


41 Data







53 Mask Mandate

6 Conclusion

References



A2 Deaths

A3 Tests


A5 Policy Variables

A6 Timing



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus005

000

005

010

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

0

4000

8000

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



minus002

000

002

004

006

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


minus01

00

01

02

03

04

05

Mar Apr Maydate

rela

tive

incr

ease

in c

ases







Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus03

minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus5000

minus4000

minus3000

minus2000

minus1000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus03

minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus20000

minus15000

minus10000

minus5000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek






Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus02

minus01

00

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus10000

minus5000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


minus06

minus04

minus02

00

Mar Apr Maydate

rela

tive

incr

ease

in c

ases






6 Conclusion






References


















































1 2 3NYT 100 029 038JHU 029 100 033CTP 038 033 100












1

10

100

1000

10000

Apr May Jundate

New

cas

es

NYT

1

10

100

1000

10000

Apr May Jundate

New

cas

es

JHU

1

10

100

1000

10000

Apr May Jundate

New

cas

es

CTP













1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

NYT

1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

JHU

1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

CTP



1

10

100

1000

10000

Apr May Jundate

Dea

ths

NYT

1

10

100

1000

10000

Apr May Jundate

Dea

ths

JHU

1

10

100

1000

Apr May Jundate

Dea

ths

CTP







1e+01

1e+03

1e+05

Apr May Jundate

Test

s


1e+01

1e+03

1e+05

Apr May Jundate

Test

s



























o reuse allowed w

ithout permission

(which w


ho has granted medR



ay 30 2020

httpsdoiorg1011012020052720115139doi

medR

xiv preprint










1 Introduction








32 Counterfactuals


41 Data







53 Mask Mandate

6 Conclusion

References



A2 Deaths

A3 Tests


A5 Policy Variables

A6 Timing



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus03

minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce

Clo

sed

mov

ie th

eate

rs C

lose

d re

stau

rant

s ex

cept

take

out

Clo

sed

non

esse

ntia

l bus

ines

ses

minus5000

minus4000

minus3000

minus2000

minus1000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus03

minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

Clo

sed

rest

aura

nts

exce

pt ta

ke o

ut

minus20000

minus15000

minus10000

minus5000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek






Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus02

minus01

00

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus10000

minus5000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


minus06

minus04

minus02

00

Mar Apr Maydate

rela

tive

incr

ease

in c

ases






6 Conclusion






References


















































1 2 3NYT 100 029 038JHU 029 100 033CTP 038 033 100












1

10

100

1000

10000

Apr May Jundate

New

cas

es

NYT

1

10

100

1000

10000

Apr May Jundate

New

cas

es

JHU

1

10

100

1000

10000

Apr May Jundate

New

cas

es

CTP













1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

NYT

1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

JHU

1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

CTP



1

10

100

1000

10000

Apr May Jundate

Dea

ths

NYT

1

10

100

1000

10000

Apr May Jundate

Dea

ths

JHU

1

10

100

1000

Apr May Jundate

Dea

ths

CTP







1e+01

1e+03

1e+05

Apr May Jundate

Test

s


1e+01

1e+03

1e+05

Apr May Jundate

Test

s



























o reuse allowed w

ithout permission

(which w


ho has granted medR



ay 30 2020

httpsdoiorg1011012020052720115139doi

medR

xiv preprint










1 Introduction








32 Counterfactuals


41 Data







53 Mask Mandate

6 Conclusion

References



A2 Deaths

A3 Tests


A5 Policy Variables

A6 Timing



Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus02

minus01

00

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


Man

date

face

mas

k us

e by

em

ploy

ees

in p

ublic

faci

ng b

usin

esse

s

Dat

e cl

osed

K 1

2 sc

hool

s C

lose

d re

stau

rant

s ex

cept

take

out

Sta

y at

hom

e s

helte

r in

pla

ce C

lose

d m

ovie

thea

ters

Clo

sed

non

esse

ntia

l bus

ines

ses

minus10000

minus5000

0

Mar Apr Maydate

Cha

nge

in c

ases

in p

ast w

eek



minus02

minus01

00

01

Mar Apr Maydate

Cha

nge

in ∆

log∆

C


minus06

minus04

minus02

00

Mar Apr Maydate

rela

tive

incr

ease

in c

ases






6 Conclusion






References


















































1 2 3NYT 100 029 038JHU 029 100 033CTP 038 033 100












1

10

100

1000

10000

Apr May Jundate

New

cas

es

NYT

1

10

100

1000

10000

Apr May Jundate

New

cas

es

JHU

1

10

100

1000

10000

Apr May Jundate

New

cas

es

CTP













1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

NYT

1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

JHU

1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

CTP



1

10

100

1000

10000

Apr May Jundate

Dea

ths

NYT

1

10

100

1000

10000

Apr May Jundate

Dea

ths

JHU

1

10

100

1000

Apr May Jundate

Dea

ths

CTP







1e+01

1e+03

1e+05

Apr May Jundate

Test

s


1e+01

1e+03

1e+05

Apr May Jundate

Test

s



























o reuse allowed w

ithout permission

(which w


ho has granted medR



ay 30 2020

httpsdoiorg1011012020052720115139doi

medR

xiv preprint










1 Introduction








32 Counterfactuals


41 Data







53 Mask Mandate

6 Conclusion

References



A2 Deaths

A3 Tests


A5 Policy Variables

A6 Timing



6 Conclusion






References


















































1 2 3NYT 100 029 038JHU 029 100 033CTP 038 033 100












1

10

100

1000

10000

Apr May Jundate

New

cas

es

NYT

1

10

100

1000

10000

Apr May Jundate

New

cas

es

JHU

1

10

100

1000

10000

Apr May Jundate

New

cas

es

CTP













1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

NYT

1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

JHU

1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

CTP



1

10

100

1000

10000

Apr May Jundate

Dea

ths

NYT

1

10

100

1000

10000

Apr May Jundate

Dea

ths

JHU

1

10

100

1000

Apr May Jundate

Dea

ths

CTP







1e+01

1e+03

1e+05

Apr May Jundate

Test

s


1e+01

1e+03

1e+05

Apr May Jundate

Test

s



























o reuse allowed w

ithout permission

(which w


ho has granted medR



ay 30 2020

httpsdoiorg1011012020052720115139doi

medR

xiv preprint










1 Introduction








32 Counterfactuals


41 Data







53 Mask Mandate

6 Conclusion

References



A2 Deaths

A3 Tests


A5 Policy Variables

A6 Timing













































1 2 3NYT 100 029 038JHU 029 100 033CTP 038 033 100












1

10

100

1000

10000

Apr May Jundate

New

cas

es

NYT

1

10

100

1000

10000

Apr May Jundate

New

cas

es

JHU

1

10

100

1000

10000

Apr May Jundate

New

cas

es

CTP













1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

NYT

1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

JHU

1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

CTP



1

10

100

1000

10000

Apr May Jundate

Dea

ths

NYT

1

10

100

1000

10000

Apr May Jundate

Dea

ths

JHU

1

10

100

1000

Apr May Jundate

Dea

ths

CTP







1e+01

1e+03

1e+05

Apr May Jundate

Test

s


1e+01

1e+03

1e+05

Apr May Jundate

Test

s



























o reuse allowed w

ithout permission

(which w


ho has granted medR



ay 30 2020

httpsdoiorg1011012020052720115139doi

medR

xiv preprint










1 Introduction








32 Counterfactuals


41 Data







53 Mask Mandate

6 Conclusion

References



A2 Deaths

A3 Tests


A5 Policy Variables

A6 Timing



1

10

100

1000

10000

Apr May Jundate

New

cas

es

NYT

1

10

100

1000

10000

Apr May Jundate

New

cas

es

JHU

1

10

100

1000

10000

Apr May Jundate

New

cas

es

CTP













1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

NYT

1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

JHU

1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

CTP



1

10

100

1000

10000

Apr May Jundate

Dea

ths

NYT

1

10

100

1000

10000

Apr May Jundate

Dea

ths

JHU

1

10

100

1000

Apr May Jundate

Dea

ths

CTP







1e+01

1e+03

1e+05

Apr May Jundate

Test

s


1e+01

1e+03

1e+05

Apr May Jundate

Test

s



























o reuse allowed w

ithout permission

(which w


ho has granted medR



ay 30 2020

httpsdoiorg1011012020052720115139doi

medR

xiv preprint










1 Introduction








32 Counterfactuals


41 Data







53 Mask Mandate

6 Conclusion

References



A2 Deaths

A3 Tests


A5 Policy Variables

A6 Timing



1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

NYT

1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

JHU

1e+01

1e+03

1e+05

Apr May Jundate

New

cas

es

CTP



1

10

100

1000

10000

Apr May Jundate

Dea

ths

NYT

1

10

100

1000

10000

Apr May Jundate

Dea

ths

JHU

1

10

100

1000

Apr May Jundate

Dea

ths

CTP







1e+01

1e+03

1e+05

Apr May Jundate

Test

s


1e+01

1e+03

1e+05

Apr May Jundate

Test

s



























o reuse allowed w

ithout permission

(which w


ho has granted medR



ay 30 2020

httpsdoiorg1011012020052720115139doi

medR

xiv preprint










1 Introduction








32 Counterfactuals


41 Data







53 Mask Mandate

6 Conclusion

References



A2 Deaths

A3 Tests


A5 Policy Variables

A6 Timing



1e+01

1e+03

1e+05

Apr May Jundate

Test

s


1e+01

1e+03

1e+05

Apr May Jundate

Test

s



























o reuse allowed w

ithout permission

(which w


ho has granted medR



ay 30 2020

httpsdoiorg1011012020052720115139doi

medR

xiv preprint










1 Introduction








32 Counterfactuals


41 Data







53 Mask Mandate

6 Conclusion

References



A2 Deaths

A3 Tests


A5 Policy Variables

A6 Timing




















o reuse allowed w

ithout permission

(which w


ho has granted medR



ay 30 2020

httpsdoiorg1011012020052720115139doi

medR

xiv preprint










1 Introduction








32 Counterfactuals


41 Data







53 Mask Mandate

6 Conclusion

References



A2 Deaths

A3 Tests


A5 Policy Variables

A6 Timing










1 Introduction








32 Counterfactuals


41 Data







53 Mask Mandate

6 Conclusion

References



A2 Deaths

A3 Tests


A5 Policy Variables

A6 Timing

causal impact of masks, policies, behavior on early covid ...€¦ · 27/05/2020 · causal impact...

Documents