Dodd, Olga; Liao, Shushu
A icle — Published Ve sion
The ole o in en o y in i m esilience o he Co id‐19
pandemic
Eu opean Financial Managemen
P o ided in Coope a ion wi h:
John Wiley & Sons
Sugges ed Ci a ion: Dodd, Olga; Liao, Shushu (2024) : The ole o in en o y in i m esilience o he
Co id‐19 pandemic, Eu opean Financial Managemen , ISSN 1468-036X, Wiley, Hoboken, NJ, Vol. 31,
Iss. 2, pp. 786-818,
h ps://doi.o g/10.1111/eu m.12517
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/319327
S anda d-Nu zungsbedingungen:
Die Dokumen e au EconS o dü en zu eigenen wissenscha lichen
Zwecken und zum P i a geb auch gespeiche und kopie we den.
Sie dü en die Dokumen e nich ü ö en liche ode komme zielle
Zwecke e iel äl igen, ö en lich auss ellen, ö en lich zugänglich
machen, e eiben ode ande wei ig nu zen.
So e n die Ve asse die Dokumen e un e Open-Con en -Lizenzen
(insbesonde e CC-Lizenzen) zu Ve ügung ges ell haben soll en,
gel en abweichend on diesen Nu zungsbedingungen die in de do
genann en Lizenz gewäh en Nu zungs ech e.
Te ms o use:
Documen s in EconS o may be sa ed and copied o you pe sonal
and schola ly pu poses.
You a e no o copy documen s o public o comme cial pu poses, o
exhibi he documen s publicly, o make hem publicly a ailable on he
in e ne , o o dis ibu e o o he wise use he documen s in public.
I he documen s ha e been made a ailable unde an Open Con en
Licence (especially C ea i e Commons Licences), you may exe cise
u he usage igh s as speci ied in he indica ed licence.
h p://c ea i ecommons.o g/licenses/by/4.0/
DOI: 10.1111/eu m.12517
ORIGINAL ARTICLE
The ole o in en o y in i m esilience o he
Co id‐19 pandemic
Olga Dodd
1
|Shushu Liao
2
1
Depa men o Finance, Auckland
Uni e si y o Technology, Auckland,
New Zealand
2
Depa men o Leade ship and
Managemen , Kühne Logis ics
Uni e si y, Hambu g, Ge many
Co espondence
Shushu Liao, Depa men o Leade ship
and Managemen , Kühne Logis ics
Uni e si y, Hambu g, Ge many.
Email: [email p o ec ed]
Abs ac
We s udy he ole o in en o y in co po a e esilience
o Co id‐19 in 2020, which igge ed exogenous shocks
o consume demand, commodi y p ices and supply
chains. Unexpec ed d ops in consume demand and
commodi y p ices inc ease he cos s o in en o y.
Con e sely, in en o y holdings can bu e agains
supply dis up ions. Empi ically, US i ms wi h highe
in en o y expe ienced mo e nega i e s ock ma ke
esponses ea ly in he c isis due o alling consume
demand. Howe e , since May 2020, in en o y has
become aluable as a hedge agains supply dis up ions,
imp o ing i m pe o mance. Du ing Co id‐19, unlike
o he c ises, in en o y played a unique ole as a hedge
agains supply dis up ions.
KEYWORDS
commodi y p ice shock, consume demand shock, Co id‐19,
in en o y, supply chain dis up ion
JEL CLASSIFICATION
G31, G32, G01
Eu Financ Manag. 2025;31:786–818.786
|
wileyonlinelib a y.com/jou nal/eu m
EUROPEAN
FINANCIAL MANAGEMENT
This is an open access a icle unde he e ms o he C ea i e Commons A ibu ion License, which pe mi s use, dis ibu ion and
ep oduc ion in any medium, p o ided he o iginal wo k is p ope ly ci ed.
© 2024 The Au ho (s). Eu opean Financial Managemen published by John Wiley & Sons L d.
We would like o hank John A. Doukas ( he edi o ) and an anonymous e e ee o help ul commen s. We also would
like o hank Ma cel P okopczuk, Leibniz Uni e si y Hanno e , Jing Yu, Uni e si y o Sydney, Ali eza Tou ani‐Rad,
Auckland Uni e si y o Technology and o he pa icipan s o he semina a Kühne Logis ics Uni e si y, Ge many, and
2021 New Zealand Finance Mee ings o hei help ul commen s and sugges ions. Open Access unding enabled and
o ganized by P ojek DEAL.
1|INTRODUCTION
Fi ms hold in en o y o manage s ockou and inpu p ice isks (Bianco & Gamba, 2019) and
hedge agains supply chain dis up ions (Gao, 2018; Kulchania & Thomas, 2017). In he las
se e al decades, a signi ican educ ion in US i ms' in en o y holdings, mainly due o supply
chain managemen de egula ion and inno a ion, has inc eased he isk o dis up ions
(Kulchania & Thomas, 2017).
1
Wi h his o ically low in en o y holdings, i ms ace high cos s o
s ockou , inpu p ice luc ua ions, and supply chain dis up ion and ely mo e on hei supply
chains (Bianco & Gamba, 2019; Kulchania & Thomas, 2017). On he lip side, lowe in en o y
holdings educe s o age and se ice cos s, ee up wo king capi al and enable an inc ease in
cash holdings (Ba es e al., 2009). In his s udy, we examine in en o y holdings' ole in co -
po a e esilience o he Co id‐19 pandemic associa ed wi h exogenous shocks o consume
demand, commodi y p ices and supply chains.
The Co id‐19 pandemic in 2020 a ec ed he human popula ion due o he apid sp ead o
he SARS‐CoV‐2 i us a ound he globe. In addi ion o signi ican heal h and social cos s, his
pandemic had subs an ial economic implica ions. Wi h he in oduc ion o measu es o con ain
he sp ead o he i us, including “s ay‐a ‐home”o de s and manda o y social dis ancing in he
i s pa o 2020, consume demand o disc e iona y p oduc s and se ices had plunged.
Bekae e al. (2020) posi ha wo‐ hi ds o he d op in g oss domes ic p oduc in he i s
qua e o 2020 was asc ibed o he nega i e shock o agg ega e demand. High le els o
unce ain y ha e u he con ibu ed o educed consump ion and in es men among con-
sume s and i ms (Ozili & A un, 2020). Wi h he sha p educ ion in demand o oil and he
ollowing oil p ice wa be ween Saudi A abia and Russia, oil p ices collapsed by mo e han 20%
in a single day on 9 Ma ch 2020 (Albulescu, 2020).
Fu he mo e, public heal h measu es, such as “s ay‐a ‐home”o de s, social dis ancing ules
and isola ion equi emen s, led o manu ac u ing acili ies wo king a a educed capaci y o
e en closing down, causing signi ican supply chain dis up ions. In Feb ua y 2020, China was
he i s coun y o shu down ac o ies o p e en he sp ead o he i us, hampe ing global
supply chains, pa icula ly o i ms elying on Chinese supplie s (Ha en & Simchi‐Le i, 2020;
Meie & Pin o, 2020; The Economis , 2020). As he pandemic p og essed, supply chain dis-
up ions became mo e se e e and widesp ead (Helpe & Sol as, 2021), which po en ially had
de as a ing inancial consequences o i ms (Hend icks & Singhal, 2003,2005a,2005b). The
dis inc na u e o supply chain dis up ions du ing he Co id‐19 pandemic se s i apa om
p e ious c ises, p omp ing esea ch on co po a e esilience in he con ex o his global
pandemic.
In his s udy, we examine i m pe o mance du ing di e en s ages o he Co id‐19 pan-
demic o assess he alue o in en o y holdings. Co id‐19 has igge ed unexpec ed exogenous
shocks o consume demand, commodi y p ices and global supply chains.
2
On he one hand,
due o he plunge in consume demand and sales in he i s pa o 2020, he alue o in en o y
as a hedge agains s ockou had signi ican ly diminished. Also, he concu en collapse o
1
Se e al s udies epo a dec ease in in en o y holdings o e he las 50 yea s, o example, Rajagopalan and Malho a
(2001) and Chen e al. (2005).
2
Co id‐19 was an exogenous shock ha had no bea ing on co po a e in en o y holdings be o e he ou b eak. Ramelli
and Wagne (2020) p o ide a imeline o he key e en s. The i s cases o he i us we e epo ed o he WHO on 31
Decembe 2019. Human‐ o‐human ansmission was no con i med un il 20 Janua y 2020, and he WHO issued he
i s epo on he ou b eak on 22 Janua y 2020.
DODD and LIAO EUROPEAN
FINANCIAL MANAGEMENT
|
787
commodi y p ices u he diminished he alue o in en o y by educing i s bene i as a hedge
agains inpu p ice isk. On he o he hand, in en o y was aluable in sa egua ding agains
global supply chain dis up ions du ing he Co id‐19 pandemic. Mo eo e , in en o y ca ies
s o age and oppo uni y cos s. The e o e, he ne e ec o in en o y holdings on i m pe -
o mance du ing he Co id‐19 c isis emains an empi ical ques ion. The ne e ec can be
nega i e i he alue o in en o y holdings as a hedge agains s ockou , ising commodi y p ices
and supply chain dis up ions is ou weighed by i s holding cos s. This is wha we ind in he i s
pa o 2020 when consume demand and commodi y p ices plumme ed, and supply chain
issues jus began o eme ge.
The economic condi ions changed in May 2020, when he US o al business and e ail sales
eco e ed quickly o he p ec isis le els a e hi ing hei lowes le el in Ap il 2020. Following
he eco e y in consume demand, commodi y p ices ebound om May 2020. In he en-
i onmen o ising sales and inpu p ices, we expec he in en o y alue o become posi i e.
Mo eo e , as he Co id‐19 pandemic con inued, supply chain issues became mo e p ominen
(Helpe & Sol as, 2021). In en o y alue as a hedge agains supply chain dis up ions is man-
i es ed mo e du ing his ime. Indeed, we ind ha highe in en o y holdings wa an ed be e
i m pe o mance in May–Decembe 2020, pa icula ly o i ms expe iencing supply chain
dis up ions in 2020.
Ou sample includes all publicly aded US i ms om Compus a wi h a ailable i m‐le el
da a, excluding inancial, eal es a e and u ili ies i ms—3429 i ms in o al. We examine he
de e minan s o he i m inancial and ope a ing pe o mance in 2020. We spli ou analysis
in o wo pa s: (1) an analysis o he Co id‐19 c isis using he Janua y–Ap il 2020 sample, and
(2) a longe ‐ un analysis o he Co id‐19 pandemic using he ull yea 2020 wi h a ocus on he
la e s age o he pandemic in May–Decembe 2020 ha ea u ed a s ong eco e y in consume
demand and commodi y p ices, bu also se e e and widesp ead supply chain dis up ions.
3
We
measu e he se e i y o Co id‐19 using he change in he numbe o daily cases in each US s a e
epo ed by USAFac s. To cons uc ou in en o y holdings a iable, we use he i m's
in en o y posi ion be o e he onse o he Co id‐19 c isis. This app oach add esses he conce n
ha concu en in en o y holdings may be endogenous o unobse able i m‐speci ic ac o s
ha could explain i m pe o mance du ing he Co id‐19 pandemic (see, e.g., Duchin
e al., 2010).
We documen ha in Janua y–Ma ch 2020, i ms wi h highe p ec isis in en o y le els
expe ienced a mo e nega i e s ock ma ke esponse o he g ow h in Co id‐19 cases, sugges ing
ha when consume demand and commodi y p ices a e alling, he cos s o ca ying in en o y
ou weigh i s bene i s. The nega i e impac o in en o y in Janua y–Ma ch 2020 is economically
signi ican . One s anda d de ia ion inc ease in in en o y holdings leads o a 0.024% decline in
daily s ock e u ns, holding he g ow h a e o Co id‐19 cases a he mean, which ep esen s a
15.42% dec ease o e he absolu e alue o he uncondi ional mean o daily s ock e u ns
o 0.156%.
The nega i e impac o in en o y holding on s ock e u ns in Janua y–Ma ch 2020 emains
signi ican when we con ol o he impac o cash holdings and o he i m cha ac e is ics
documen ed in he li e a u e as signi ican de e minan s o s ock ma ke esponse o Co id‐19,
3
We selec he sample ime pe iods based on he economic condi ions such as consume demand, commodi y p ices
and global supply chain dis up ions discussed in Sec ion 2“Economic backd op du ing he Co id‐19 pandemic
in 2020”.
788
|
EUROPEAN
FINANCIAL MANAGEMENT
DODD and LIAO
including cash holdings, le e age, g ow h oppo uni ies, p o i abili y, i m size, cash low (Ding
e al., 2021; Ramelli & Wagne , 2020) and o he a iables ha po en ially in luence s ock
e u ns, including S&P500 index e u n, s ock e u n om he p e ious day, sha e u no e and
daily ange. Ou main esul is also obus o using al e na i e me hods o es ima e s ock
ma ke pe o mance ( isk‐adjus ed e u ns, weekly and mon hly e u ns, and buy‐and‐hold
abno mal e u ns), a mo e es ic i e sample pe iod (1 Janua y–20 Ma ch 2020, be o e he
Fede al Rese e Boa d [Fed] in e en ion announcemen ), and al e na i e in en o y measu es,
including in en o y‐ o‐sales a io, in en o y‐days a io and abno mal in en o y.
The documen ed nega i e impac o in en o y on i m pe o mance du ing he Co id‐19
c isis is a guably d i en by consume demand and commodi y p ice shocks. To es his
p oposi ion, we exploi a signi ican a ia ion ac oss indus ies in he deg ee o he shock o
demand and commodi y p ices du ing he Co id‐19 c isis (Ozili & A un, 2020; Ramelli &
Wagne , 2020). On he basis o he sales changes in Q1 2020, consume disc e iona y, ene gy,
indus ials and ma e ials indus ies a e signi ican ly ad e sely a ec ed by Co id‐19, while
consume s aples, in o ma ion echnology, heal h ca e and communica ion se ices indus ies
a e less a ec ed. The nega i e impac o in en o y holds only o i ms in highly a ec ed
indus ies, showing ha he nega i e alue o in en o y is associa ed wi h he d op in con-
sume demand and commodi y p ices du ing he Co id‐19 c isis.
Nex , o disen angle he e ec s o consume demand shock and commodi y p ice shock, we
conside di e en componen s o in en o y— aw ma e ials, wo k‐in‐p og ess and inished
goods. We ind ha he inished goods componen p edomina es he nega i e impac o
in en o y holdings in Janua y–Ma ch 2020, implying ha he d op in consume demand can
explain he nega i e impac o in en o y holdings du ing he Co id‐19 c isis. To ein o ce ou
indings on he ole o consume demand shocks, we also p o ide e idence (in Suppo ing
In o ma ion Appendix) ha in en o y holdings nega i ely a ec i m pe o mance du ing wo
o he c ises accompanied by signi ican ad e se demand shocks: he 9/11 e o is a acks and
he 2007–2008 Global Financial C isis.
One ad an age o in en o y holdings is p o ec ion agains supply chain dis up ions caused
by Co id‐19 (Ha en & Simchi‐Le i, 2020; Helpe & Sol as, 2021). In he i s pa o 2020, he
Co id‐19 ou b eak o ced many ac o ies in China o shu down, causing dis up ions o i ms
ha ely on Chinese supplies (Ha en & Simchi‐Le i, 2020; Meie & Pin o, 2020). We use he
Hobe g and Moon Tex ‐based O sho ing Ne wo k Da abase (Hobe g & Moon, 2017,2019) o
iden i y i ms wi h Chinese supplie s. We ind ha in Janua y–Ma ch 2020, he nega i e impac
o in en o y is mi iga ed by he bene i s o in en o y holdings as a hedge agains supply chain
dis up ion o i ms wi h Chinese supplie s.
The second pa o ou in es iga ion, he longe ‐ un analysis, co e s he ull yea 2020. I
ocuses on he ole o in en o y holdings in i ms' esilience o he Co id‐19 pandemic in he
la e s age o he Co id‐19 pandemic in May–Decembe 2020. Du ing his pe iod o ebounding
consume demand and ising commodi y p ices bu dis up ed supply chains, we documen a
e e sal in he impac o in en o y holdings. We ind ha i ms wi h highe p ec isis in en o y
holdings expe ience highe s ock ma ke e u ns in May–Decembe 2020. The posi i e impac
o in en o y in May–Decembe 2020 is economically signi ican . One s anda d de ia ion
inc ease in in en o y holdings leads o a 0.063% inc ease in daily s ock e u ns, holding he
g ow h a e o Co id‐19 cases a he mean, which ep esen s a 30.7% inc ease o e he absolu e
alue o he uncondi ional mean o daily s ock e u ns in May–Decembe 2020 o 0.206%.
While he i s pa o 2020 wi nessed a b eakdown o global supply chains caused mainly by
shu downs o ac o ies in China, la e in 2020, wi h he sp ead o he pandemic in he Uni ed
DODD and LIAO EUROPEAN
FINANCIAL MANAGEMENT
|
789
S a es and globally, supply chain issues a e no limi ed o i ms wi h Chinese supplie s.To
cap u e global supply chain dis up ions, we cons uc a b oade measu e, he numbe o
men ions o “supply chain”in he i m's annual 10‐K ile in 2020. We ind ha i ms ha
expe ience signi ican supply chain issues bene i mo e om highe in en o y holdings in
May–Decembe 2020, when supply chain issues became mo e se e e and widesp ead. This
inding con i ms he i al ole o in en o y holdings as a isk managemen ool agains supply
chain dis up ions in he la e s age o he Co id‐19 pandemic.
Finally, we examine he impac o in en o y holdings on i ms' ope a ing pe o mance in
2020, measu ed using qua e ly seasonally adjus ed e u n on asse s and sales g ow h. In line
wi h he indings on s ock ma ke pe o mance, p ec isis in en o y is a posi i e de e minan o
i ms' ope a ing pe o mance in qua e s wo, h ee and ou o 2020. The ope a ing pe -
o mance analysis ein o ces ou a gumen ha in en o y holdings became aluable in he la e
s age o he Co id‐19 pandemic in 2020 when consume demand and commodi y p ices
eco e ed, bu supply chain issues wo sened.
Ou s udy con ibu es o wo s ands o li e a u e. Fi s , i con ibu es o he li e a u e on he
economic impac s o he Co id‐19 pandemic, pa icula ly on co po a e ac o s ha de e mine a
i m's esilience o he Co id‐19 pandemic (e.g., Albuque que e al., 2020; Ding e al., 2021;
Fahlenb ach e al., 2020; Ramelli & Wagne , 2020). Se e al s udies on he economic conse-
quences o Co id‐19 poin o he impo ance o in en o y and global supply chain managemen .
Deme s e al. (2021) epo ha he indus y‐adjus ed in en o y u no e a io (cos s o goods
sold di ided by in en o y holdings) is a signi ican posi i e de e minan o a i m's s ock
pe o mance esilience in he i s pa o 2020. This e idence aligns wi h ou inding ha lowe
in en o y le els a e alue‐adding in he ea ly days o he Co id‐19 pandemic. Ramelli and
Wagne (2020) epo ha in e na ionally o ien ed US i ms, especially hose exposed o China,
expe ienced wo se s ock ma ke pe o mance in he ea ly s age o he Co id‐19 pandemic in
Janua y–Feb ua y 2020 when China had lockdown es ic ions in place. Fu he mo e, hey ind
he e ec o exposu e o China became posi i e and signi ican in Feb ua y–Ma ch 2020 when
he pandemic in China was ge ing mo e unde con ol. Fo a global sample, Ding e al. (2021)
show ha i ms wi h supplie s loca ed in coun ies mo e a ec ed by Co id‐19 expe ienced mo e
signi ican s ock p ice declines in he i s qua e o 2020, highligh ing he impo ance o
exposu e o global supply chains du ing he Co id‐19 pandemic. Meie and Pin o (2020)
show ha US indus ies wi h high exposu e o impo s om China expe ienced a signi ican
decline in economic ac i i ies in Ma ch–Ap il 2020 due o supply chain dis up ion issues.
Cheema‐Fox e al. (2021) examine companies' media esponses o Co id‐19 in Feb ua y–Ma ch
2020 ega ding hei supply chain, among o he ac o s. They ind ha companies wi h mo e
posi i e sen imen in hei esponses expe ience less nega i e s ock ma ke e u ns. They a gue
ha companies mo e commi ed o supplie ela ionships may espond mo e quickly o modi y
hei supply chains o minimise he ad e se e ec s o supply chain dis up ions. We con ibu e
o his li e a u e by p o iding an in‐dep h analysis o he impo ance o in en o y holdings
condi ional on he exposu e o Co id‐19 shocks, including supply chain dis up ions, du ing he
Co id‐19 pandemic.
Second, ou s udy con ibu es o he li e a u e on wo king capi al managemen ha
explo es he ole o in en o y as a isk managemen ool. In en o y managemen is ecognised
as i al o imp o ing ope a ional lexibili y and business g ow h (e.g., Chalo a, 2013; P a e
e al., 2001). Fo ins ance, Wang (2019) epo s ha a high cash con e sion cycle (i.e., he ime a
i m akes o sell i s in en o y o collec i s ecei ables) leads o poo s ock ma ke pe o mance.
In con as , Ca pen e e al. (1994) and Kashyap e al. (1994) documen ha in en o y holdings
790
|
EUROPEAN
FINANCIAL MANAGEMENT
DODD and LIAO
ha e liquidi y alue o inancially cons ained i ms. Mo e ecen ly, Dasgup a e al. (2019) ind
ha cons ained i ms deple e in en o y mo e agg essi ely in esponse o ad e se shocks.
Bianco and Gamba (2019) show ha i ms hold in en o y o mi iga e commodi y inpu p ice
and cash low isks. Bo (2001) and Caglayan e al. (2012) posi ha i ms wi h heigh ened
demand unce ain y build up in en o y o a oid s ockou . Resea ch also documen s he
oppo uni y cos s o holding in en o y due o a subs i u ion e ec be ween in en o y and cash
holdings. Fo ins ance, Ba es e al. (2009) and Kulchania and Thomas (2017) a gue ha he
d ama ic decline in in en o y explains he end o inc easing cash holdings o US i ms. Gao
(2018) shows ha i ms can shi esou ces om in en o y o cash holdings due o swi ching o
a jus ‐in‐ ime (JIT) in en o y sys em. Ou s udy con ibu es o his li e a u e by ocusing on he
cos s and bene i s o in en o y holdings in co po a e esilience o a global pandemic.
The es o he pape is o ganised as ollows. Sec ion 2discusses s ock ma ke p ices,
consume demand, commodi y p ices and supply chain issues as an economic backd op o he
Co id‐19 pandemic. Sec ion 3p o ides he heo e ical backg ound and expec a ions on he ole
o in en o y in gene al and du ing he Co id‐19 pandemic. Sec ion 4desc ibes ou da a and
sample and epo summa y s a is ics. Sec ion 5discusses he empi ical s a egy and he esul s
in de ail. Sec ion 6concludes.
2|ECONOMIC BACKDROP DURING THE COVID‐19
PANDEMIC IN 2020
The sp ead o he Co id‐19 pandemic has igge ed a d op in s ock p ices in he i s pa o
2020. Panel (a) o Figu e 1plo s he S&P500 daily p ices om Compus a om 1 Decembe
2019 o 31 Decembe 2020. We obse e a sha p and conside able d op in s ock p ices in
Feb ua y–Ma ch 2020, wi h a s ong eco e y in he second pa o 2020.
As a esul o “s ay‐a ‐home”o de s,
4
business ac i i ies and consume demand d opped
subs an ially in Ap il 2020. Panel (b) o Figu e 1plo s mon hly o al business sales in 2020
epo ed by he U.S. Census Bu eau. The o al business ( e ail) sales declined om $1,347,262
($418,734) million in Janua y 2020 o $1,165,203 ($377,210) in Ap il 2020 be o e ebounding o
$1,274,361 ($462,286) in May 2020 (The U.S. Census Bu eau). Acco ding o he Na ional
Bu eau o Economic Resea ch, he US economy was in a deep bu sho ecession in Ma ch–
Ap il 2020 and s a ed expanding in May 2020.
5
The dep essed demands du ing his pe iod led o commodi y p ices, pa icula ly c ude oil
used o gasoline and uel, collapsing (e.g., Albulescu, 2020). Panel (c) o Figu e 1plo s he
Bloombe g Commodi y index and Wes Texas In e media e (WTI) c ude oil p ices om
1 Decembe 2019 o 31 Decembe 2020. I shows ha he Commodi y Index and he WTI c ude
oil p ices eco ded a con inuous decline since he beginning o he Co id‐19 ou b eak and a
c ash in Ma ch 2020. Fu he mo e, he oil p ices plunged below ze o on 20 Ap il 2020, alling
4
In Ma ch–Ap il 2020, mos US s a es issued “s ay‐a ‐home”o de s p esc ibing ha people limi hei mo emen s o
essen ial ac i i ies and o de ing nonessen ial businesses o shu down. On 3 Ap il 2020, 90% o he US popula ion was
li ing unde “s ay‐a ‐home”o de s. Sou ce:h ps://www.pbs.o g/newshou /poli ics/mos -s a es-ha e-issued-s ay-a -
home-o de s-bu -en o cemen - a ies-widely.
5
h ps://www.nbe .o g/news/business-cycle-da ing-commi ee-announcemen -july-19-2021 and also h ps://www.
cnbc.com/2021/07/19/i s-o icial- he-co id- ecession-las ed-jus - wo-mon hs- he-sho es -in-us-his o y.h ml
DODD and LIAO EUROPEAN
FINANCIAL MANAGEMENT
|
791
in o nega i e oil p ice e i o y o he i s ime in his o y. Commodi y p ices eco e y s a ed
in May 2020.
Supply chain dis up ions caused by Co id‐19 we e e lec ed in he global anspo a ion
cos s. Panel (d) o Figu e 1plo s he weekly mo emen s o wo global anspo a ion cos s
indices: he Ha pex index (Ha pe Pe e sen Cha e Ra es Index), which e lec s he wo ldwide
con aine shipping a e changes in he cha e ma ke o con aine ships, and he Bal ic D y
Index (BDI), which is a global a e age cos o anspo ing d y bulk ma e ials. Bo h indices ose
in he second pa o 2020, indica ing a subs an ial inc ease in global anspo a ion cos s due o
supply chain dis up ions du ing he Co id‐19 pandemic. Acco ding o he epo by GEP, in
FIGURE 1 Economic condi ions du ing he Co id‐19 pandemic in 2020. Panel (a) S ock ma ke
pe o mance. Panel (a) plo s he S&P500 index daily p ices om 1 Decembe o 31 Decembe 2020 (Sou ce:
Thomson Reu e s). Panel (b) Consume demand and sales. Panel (b) plo s o al business sales in Janua y–
Decembe 2020 epo ed by he U.S. Census Bu eau. Sales a e in millions o dolla s (Sou ce: The U.S. Census
Bu eau h ps://www.census.go /m is/index.h ml). Panel (c) Commodi y p ice index and WTI oil p ices. Panel
(c) plo s he Bloombe g Commodi y index (le y‐axis) and WTI c ude oil p ices ( igh y‐axis) om 1 Decembe
2019 o 31 Decembe 2020 (Sou ce: Thomson Reu e s' websi e). Panel (d) Supply chains in 2020—global
anspo a ion cos s. Panel (d) plo s he weekly mo emen s o he Ha pex index in he solid line and Bal ic D y
Index (BDI) in he dashed line. The Ha pex (Ha pe Pe e sen Cha e Ra es Index) e lec s he wo ldwide
con aine shipping a e changes in he cha e ma ke o con aine ships. The BDI is he global a e age cos o
anspo ing d y bulk ma e ials (Sou ces: Ha pe Pe e sen Holding GmbH and Bal ic Exchanges). WTI, Wes
Texas In e media e. [Colo igu e can be iewed a wileyonlinelib a y.com]
792
|
EUROPEAN
FINANCIAL MANAGEMENT
DODD and LIAO
2020, he o al cos o supply chain dis up ions o US and Eu opean businesses was $4 n, wi h
45% o i ms epo ing ha Co id‐19 signi ican ly dis up ed hei supply chain.
6
3|THEORETICAL BACKGROUND AND PREDICTIONS
3.1 |The ole o in en o y holdings
Fi ms hold in en o y o a oid s ockou , hedge agains ising inpu p ices and mi iga e supply
chain dis up ions. Acco ding o he s ockou ‐a oidance heo y, i ms in es in in en o y o a oid
s ockou and loss o p ospec i e sales when hey expe ience an unan icipa ed inc ease in demand
since p oduc ion akes ime (e.g., Dasgup a e al., 2019; Eichenbaum, 1989;Wen,2005).
7
Fo hedging pu poses, i ms hold mo e in en o y when an icipa ing a ise in inpu p ices.
Chen e al. (2005) a gue ha high in la ion incen i ises i ms o buy inpu s ea lie . Bianco and
Gamba (2019) posi ha i ms hold in en o y as an ope a ional hedge, and using in en o y as a
isk managemen ool adds mo e alue when commodi y p ices a e ising.
Finally, i ms hold in en o y o hedge agains supply chain dis up ions (Gao, 2018;
Kulchania & Thomas, 2017; Tomlin, 2006). Supply chain dis up ions can be e y cos ly o
i ms ha a e unp o ec ed. Hend icks and Singhal (2003,2005a,2005b) documen a signi ican
de e io a ion in inancial and ope a ing pe o mance and a las ing inc ease in he cos o capi al
o i ms ha expe ience supply chain dis up ion e en s. One s a egy o hedge supply isk is o
hold highe in en o y le els as a bu e . Howe e , in he las se e al decades, US i ms sig-
ni ican ly educed hei in en o y holdings due o supply chain managemen de egula ion and
inno a ion and he use o JIT in en o y managemen p ac ices (Gao, 2018; Kulchania &
Thomas, 2017). Mo eo e , US i ms a e less e ically in eg a ed han in he pas and mo e
elian on o sho e supplie s (Snyde e al., 2016). Low in en o y holdings and high eliance on
supplie s imply ha i ms may ace subs an ial supply chain dis up ion isks.
Economic li e a u e de ines in en o y cos as a unc ion o he dis ance be ween he ac ual
in en o y holdings and he a ge in en o y le el de e mined by he i m's sales (e.g.,
Blancha d, 1983; Eichenbaum, 1989).
8
This de ini ion e lec s wo main ypes o in en o y
cos s. The i s ype is he cos s o holding in en o y ha inc ease wi h in en o y le els.
In en o y holding cos s include in es men oppo uni y cos s, physical s o age cos s, s a ing
cos s, in en o y se ice cos s (e.g., insu ance and axes) and in en o y isk o dep ecia ion cos s
(e.g., obsolescence o he o in en o y) (La Londe & Lambe , 1977). The second ype is
s ockou cos s, which a e high when he in en o y le els a e low, o he sales le els (and
he eby a ge in en o y le els) a e high. The e o e, in en o y holding is a ade‐o be ween
he bene i s o a oiding s ockou and he cos s o holding in en o y.
This de ini ion can be ex ended o include cos s o ising inpu p ices and cos s o po en ial
supply chain dis up ions. Fi ms hold addi ional in en o y as a bu e agains ising inpu cos s
6
h ps://www.cips.o g/supply-managemen /news/2021/ma ch/ o al-cos -o -supply-chain-dis up ion-in-2020-was-4 n/
7
Fi ms wi h con ex p oduc ion cos s ace a mo e apid ise in p oduc ion cos s when demand is a ou able. The e o e,
i ms need o accumula e in en o y as hey would unde p oduce when demand is high and o e p oduce when demand
is low.
8
Blancha d (1983) de ines he cos s o holding in en o y as
()
G
dII=/2 −*
,
2whe e
I
is he in en o y holdings, and
IaS
*=
is he a ge in en o y le el de e mined by sales
S
. When
I
is signi ican ly highe han I
*
, i ms ace he
holding cos s. When
I
is signi ican ly lowe han I
*
, i ms ace he cos s o s ockou .
DODD and LIAO EUROPEAN
FINANCIAL MANAGEMENT
|
793
TABLE 3 In en o y and sho ‐ un s ock ma ke esponse o he Co id‐19 c isis.
This able epo s he es ima es o O dina y Leas Squa es panel eg essions o he impac o in en o y on he
esponse o s ock e u ns o he g ow h o Co id‐19 cases using daily e u ns in Panel A and weekly, mon hly and
buy‐and‐hold abno mal e u ns in Panel B. The sample pe iod is om 1 Janua y o 30 Ap il 2020, excep o
Model 4 in Panel A, which employs he sample om 1 Janua y o 20 Ma ch 2020. In Panel A, he dependen
a iable is he daily s ock e u n (Models 1, 2, 3 and 5) and isk‐adjus ed daily s ock e u n (Model 4). In Panel B,
in Models 1 and 2 (3 and 4), he dependen a iable is he weekly (mon hly) s ock e u ns, and he independen
a iables a e all measu ed on a weekly (mon hly) basis.In Model 5 o Panel B, he dependen a iable is buy‐and‐
hold abno mal e u ns o e Janua y–Ap il 2020, compu ed using he Fama–F ench and Ca ha ou ‐ ac o
model. Co id19 is he g ow h a e o Co id‐19 cases by s a e. In en o y is he a e age o al in en o y o o al asse s
in 2019. All a iables a e de ined in Appendix A.1. In Models 2–5 in Panel A and Models 1–4 in Panel B, i m‐le el
a iables a e abso bed by i m ixed e ec s. In pa en heses, we epo obus s anda d e o s clus e ed a he i m
le el. ***, ** and * indica e signi icance a he 1%, 5% and 10% le els, espec i ely.
Panel A: Daily e u ns
(1) (2) (3) (4) (5)
Indus y
and s a e
ixed e ec s
Fi m ixed
e ec s
Wi h e u n
ac o
con ols
Fou ‐ ac o
model
1
Janua y–20
Ma ch 2020
In en o y ×Co id19 −2.212** −2.159** −2.210** −2.390** −2.283**
(1.01) (1.02) (1.03) (1.05) (1.10)
Co id19 −1.360*** −1.396*** −1.556*** −1.128** −1.579***
(0.46) (0.46) (0.48) (0.49) (0.49)
Cash ×Co id19 0.727* 0.791* 0.948** 0.314
(0.44) (0.44) (0.48) (0.48)
Le e age ×Co id19 0.001 0.001 −0.002 0.002
(0.01) (0.01) (0.01) (0.01)
MTB ×Co id19 0.093 0.094 −0.068 0.280
(0.21) (0.22) (0.26) (0.23)
ROA ×Co id19 0.131** 0.135** 0.072 0.090
(0.05) (0.05) (0.06) (0.06)
Fi m size ×Co id19 0.056 0.031 0.071 0.142
(0.14) (0.14) (0.17) (0.16)
Cash low ×Co id19 −0.100 −0.103 0.086 −0.169
(0.16) (0.17) (0.20) (0.17)
SP500 e u n 0.798***
(1.13)
Lag e u n −0.133***
(0.01)
Sha e u no e 0.831
(0.60)
800
|
EUROPEAN
FINANCIAL MANAGEMENT
DODD and LIAO
TABLE 3 (Con inued)
Panel A: Daily e u ns
(1) (2) (3) (4) (5)
Indus y
and s a e
ixed e ec s
Fi m ixed
e ec s
Wi h e u n
ac o
con ols
Fou ‐ ac o
model
1
Janua y–20
Ma ch 2020
Daily ange −0.127
(0.07)
In en o y 0.141
(0.14)
Cash 0.081
(0.08)
Le e age −0.009
(0.02)
MTB −0.000
(0.00)
ROA −0.054
(0.04)
Fi m size 0.006
(0.01)
Cash low 0.040
(0.03)
Indus y and s a e ixed
e ec s
Yes No No No No
Fi m ixed e ec s No Yes Yes Yes Yes
Obse a ions 203,930 203,930 198,822 203,930 133,915
R
2
0.002 0.008 0.141 0.010 0.016
Panel B: Weekly, mon hly o buy‐and‐hold e u ns
(1) (2) (3) (4) (5)
Weekly s ock e u ns Mon hly s ock e u ns Buy‐and‐hold
abno mal e u ns
In en o y ×Co id19 −2.558** −2.304** −0.022*** −0.014* In en o y −0.201**
(1.00) (1.10) (0.01) (0.01) (0.08)
Cash ×Co id19 −2.423* 1.302 −0.123 0.088 Cash 0.008
(1.40) (1.47) (0.10) (0.10) (0.04)
Le e age ×Co id19 0.594 1.014 0.112 0.097 Le e age −0.001
(1.50) (1.53) (0.11) (0.11) (0.01)
MTB ×Co id19 −0.586*** −0.187** −0.012** 0.004 MTB −0.000
(0.09) (0.09) (0.01) (0.00) (0.00)
(Con inues)
DODD and LIAO EUROPEAN
FINANCIAL MANAGEMENT
|
801
The economic magni ude o he coe icien es ima es on he in e ac ion a iable In en-
o y ×Co id19 is la ge. Fo example, in Model 2, one s anda d de ia ion inc ease in In en o y
leads o a 0.024% (2.4 basis poin ) decline in daily s ock e u ns holding he g ow h a e o
Co id‐19 cases, Co id19, a he mean (0.087 × 0.128 × (−2.16) = −0.024). This esul is eco-
nomically signi ican as i ep esen s a 15.42% dec ease o e he absolu e alue o he
uncondi ional mean o daily s ock e u ns o 0.156%.
Model 3 o Table 3addi ionally includes ma ke mic os uc u e con ol a iables. Daily
s ock e u ns a e posi i ely associa ed wi h S&P 500 e u ns and nega i ely wi h lag e u ns.
Sha e u no e and daily ange a e insigni ican de e minan s o s ock e u ns. No ably, he
coe icien on he in e ac ion a iable In en o y ×Co id19 emains nega i e and signi ican .
O e all, ou baseline eg ession esul s indica e ha i ms wi h high p e‐Co id in en o y
holdings pe o med wo se in he s ock ma ke in he sho un du ing he Co id‐19 c isis. The
esul s a e consis en wi h ou a gumen s ha high amoun s o in en o y du ing he c isis a e
associa ed wi h educed bene i s o a oiding s ockou and managing p ice isk and inc eased
in en o y holding cos s.
TABLE 3 (Con inued)
Panel B: Weekly, mon hly o buy‐and‐hold e u ns
(1) (2) (3) (4) (5)
Weekly s ock e u ns Mon hly s ock e u ns Buy‐and‐hold
abno mal e u ns
ROA ×Co id19 −1.622*** −1.380** −0.028 −0.018 ROA −0.004
(0.54) (0.57) (0.04) (0.04) (0.02)
Fi m size ×Co id19 −0.953 −1.523 −0.079 −0.087 Fi m size 0.010***
(1.21) (1.23) (0.09) (0.09) (0.00)
Cash low ×Co id19 −2.423* 1.302 −0.123 0.088 Cash low −0.006
(1.40) (1.47) (0.10) (0.10) (0.01)
Co id19 −1.328*** −1.310*** −0.013*** −0.010***
(0.20) (0.22) (0.00) (0.00)
SP500 e u n 0.784*** 0.538***
(1.71) (0.02)
Lag e u n −0.171*** −0.289***
(0.01) (0.01)
Sha e u no e 0.201 0.000
(0.17) (0.00)
Fi m ixed e ec s Yes Yes Yes Yes No
Indus y and s a e ixed
e ec s
No No No No Yes
Obse a ions 55,722 52,507 12,245 12,245 3072
R
2
0.054 0.134 0.274 0.385 0.145
802
|
EUROPEAN
FINANCIAL MANAGEMENT
DODD and LIAO
Al e na i e me hods o es ima ing s ock e u ns
To show he obus ness o ou indings, we employ al e na i e me hods o es ima e s ock
ma ke pe o mance, including es ima ing isk‐adjus ed e u ns and using an al e na i e
sample pe iod be o e he cen al bank in e en ions in esponse o he Co id‐19 c isis in Ma ch
2020. Addi ionally, we es ima e s ock e u ns using di e en equencies o da a, including
weekly, mon hly and buy‐and‐hold abno mal e u ns.
Ou baseline model uses aw e u ns a he han isk‐adjus ed e u ns because adjus ed
e u ns ely on s ic assump ions ha exposu es o isk ac o s emain unchanged (Ramelli &
Wagne , 2020). As a obus ness es , we es ima e Equa ion (1) wi h isk‐adjus ed s ock e u ns
as he dependen a iable es ima ed using Fama–F ench and Ca ha ou ‐ ac o model
(Model 4).
18
We es ima e a i m's ac o loading by eg essing daily e u ns on isk ac o s in
2019 and sub ac ing ac o exposu es imes he ac o e u ns om he aw e u ns. We ind
ha he nega i e impac o in en o y holdings emains signi ican when we use isk‐adjus ed
s ock e u ns o measu e i m pe o mance.
Nex , we e‐examine he e ec s o in en o y du ing he Co id‐19 c isis in he absence o
cen al bank in e en ions. On Monday, 23 Ma ch 2020, he Fed announced wo new acili ies,
a P ima y Ma ke Co po a e C edi Facili y and a Seconda y Ma ke Co po a e C edi Facili y,
o p o ide c edi o la ge co po a ions and ease liquidi y s ains (see he imeline desc ibed in
Ramelli & Wagne , 2020). We e‐es ima e he baseline eg ession o an al e na i e sample
pe iod, om 1 Janua y o 20 Ma ch 2020 (F iday be o e he Fed's announcemen ) and epo
he es ima ion esul s in Model 5 o Table 3. The coe icien es ima e on In en o y ×Co id19
emains nega i e and s a is ically signi ican a he 5% le el, con i ming he obus ness o ou
main inding.
19
Finally, in Panel B o Table 3, we epo he es ima ion esul s o models ha use weekly,
mon hly and buy‐and‐hold abno mal e u ns. In Models 1–4 o Panel B, he explana o y and
con ol a iables, he same as in Models 2 and 3 o Panel A, a e all measu ed on a weekly
(mon hly) basis. The es ima ion esul s o weekly and mon hly e u ns a e simila o hose
o daily e u ns epo ed in Panel A o Table 3. Model 5 o Panel B is a c oss‐sec ional
eg ession wi h he dependen a iable as buy‐and‐hold abno mal e u ns o e he pe iod
Janua y–Ap il 2020, compu ed using he Fama–F ench and Ca ha ou ‐ ac o model. The
nega i e and s a is ically signi ican coe icien on In en o y econ i ms ou main inding. The
posi i e and signi ican coe icien on Fi m size shows ha la ge i ms a e mo e immune o
he pandemic.
O e all, ou main inding ha in en o y holdings nega i ely impac ed he s ock ma ke
pe o mance du ing he Co id‐19 c isis is obus o using al e na i e es ima ion me hods o
s ock ma ke pe o mance.
18
The es ima ion esul s a e simila when we use isk‐adjus ed e u ns es ima ed using capi al asse ing p icing
model.
19
We also es ima e he impac o in en o y holdings using (1) an al e na i e de ini ion o he Co id‐19 c isis as
he pe iod om 20 Feb ua y o 23 Ma ch 2020 when he la ge‐scale decline in e u ns occu ed (Cheema‐Fox
e al., 2021) and (2) mon hly s ock e u ns (ins ead o daily e u ns) as he dependen a iable o an ex ended
pe iod om 1 Sep embe 2019 o 30 Ap il 2020 wi h he Co id‐19 a iable de ined as a dummy a iable ha
equals one o Feb ua y and Ma ch 2020 and ze o o he wise. The es ima ion esul s o he pe iods 20
Feb ua y–23 Ma ch 2020 and 1 Sep embe 2019–30 Ap il 2020 (no epo ed) con i m ou main esul ha highe
p ec isis in en o y holdings a e associa ed wi h a mo e nega i e s ock ma ke esponse o he Co id‐19 c isis.
DODD and LIAO EUROPEAN
FINANCIAL MANAGEMENT
|
803
5.1.2 |Explaining he nega i e impac o in en o y holdings du ing he
Co id‐19 c isis
This sec ion examines po en ial explana ions o he documen ed nega i e impac o in en o y
holdings in Janua y–Ma ch 2020. We exploi c oss‐sec ional he e ogenei y in i ms' exposu e o
Co id‐19 shocks and dis up ions and examine he ole o di e en componen s o in en o y
holdings o disen angle he ole in en o y holdings play in he ace o consume demand,
commodi y p ice and supply shocks.
Fi m's exposu e o Co id‐19: Consume demand shock and commodi y p ice shock
Fi s , we es he p oposi ion ha he ad e se shocks o consume demand and commodi y
p ices du ing Co id‐19 can explain he nega i e impac o in en o y holdings du ing his c isis.
To es his p oposi ion, we use indus y‐le el a ia ion in he deg ee o exposu e o Co id‐19,
discussed in Suppo ing In o ma ion Appendix SA.1. To empi ically assess how se e ely
Co id‐19 a ec s di e en indus ies, we calcula e he pe cen age sales changes om Q1 2019
o Q1 2020 by indus y based on GICS wo‐digi indus y codes ( epo ed in Figu e SA1 in
Suppo ing In o ma ion Appendix SA.1). We documen a signi ican d op in sales o consume
disc e iona y, ene gy, ma e ials and indus ials indus ies (GICS indus y codes 25, 10, 15 and
20, espec i ely); he e o e, we classi y hese indus ies as “High Co id‐19 shock”. The “Low
Co id‐19 shock”indus ies ha e a less signi ican d op o an inc ease in sales in Q1 2020; hey
include consume s aples, in o ma ion echnology, heal h ca e and communica ion se ices
(GICS indus y codes 30, 45, 35 and 50, espec i ely). We expec he nega i e impac o
in en o y holdings o be mo e p onounced o “High shock” han “Low shock”indus ies.
Table 4 epo s he es ima ion esul s o he baseline eg ession o he wo subsamples: (1)
i ms ope a ing in “High Co id‐19 shock”indus ies (Model 1) and (2) i ms ope a ing in “Low
Co id‐19 shock”indus ies (Model 2). As expec ed, in Model 1, he in e ac ion e m In en-
o y ×Co id19 has a nega i e and signi ican coe icien es ima e, indica ing ha he nega i e
impac o in en o y holdings is signi ican o i ms ha expe ience signi ican demand and
commodi y p ice shocks du ing he Co id‐19 c isis. In Model 2, he coe icien es ima e on
In en o y ×Co id19 is insigni ican , meaning ha o i ms less a ec ed by Co id‐19, in en o y
holdings do no ha e a signi ican ly nega i e bea ing on s ock pe o mance du ing he Co id‐19
c isis. O e all, ou esul s show ha he nega i e ole o in en o y is associa ed wi h shocks o
consume demand and commodi y p ices in he i s pa o 2020.
Di e en componen s o in en o y: Consume demand shock e sus commodi y p ice shock
To u he unde s and he ole o in en o y du ing he Co id‐19 c isis, we examine whe he he
nega i e impac o in en o y is p ima ily d i en by he consume demand shock o he com-
modi y p ices shock. To es his, we dis inguish be ween di e en componen s o in en o y
holdings, including aw ma e ials, wo k‐in‐p og ess and inished goods.
20
We assume ha he
consume demand shock is impounded in inished goods while he commodi y p ice shock is in
aw ma e ials.
In Table 5, we epo he es ima ion esul s o Equa ion (1), whe e we employ di e en
componen s o in en o y a iables ins ead o he o al in en o y holdings: Raw Ma e ials
(In en o y_RM) (Models 1 and 2), Wo k‐in‐P og ess (In en o y_WIP) (Models 3 and 4) and
20
This analysis has a educed sample size due o he limi ed a ailabili y o da a on indi idual in en o y componen s.
804
|
EUROPEAN
FINANCIAL MANAGEMENT
DODD and LIAO
Finished Goods (In en o y_FG) (Models 5 and 6). We ind ha he nega i e impac is con-
cen a ed in he Finished Goods componen o in en o y, sugges ing ha he documen ed
nega i e impac o in en o y holdings du ing he Co id‐19 c isis is mainly d i en by he d op in
consume demand in he ace o Co id‐19. The collapse in consume demand inc eases
in en o y holding cos s and educes he impo ance o in en o y as a s ockou hedge.
O he consume demand shocks
To ein o ce ou indings on he ole o a consume demand shock, we also examine he ole o
in en o y holdings du ing wo o he c ises ha we e accompanied by signi ican ad e se
demand shocks: (1) he 9/11 e o is a acks and (2) he 2007–2008 Global Financial C isis.
These es s ( epo ed in Suppo ing In o ma ion Appendix SA.2) p o ide addi ional empi ical
e idence ha he nega i e impac o in en o y holdings can be a ibu ed o ad e se consume
demand shocks.
Shock o global supply chains
We ha e shown ha he ad e se shock o consume demand educes he alue o in en o y
holdings o he a ec ed i ms. On he lip side, in en o y holdings may be aluable o i ms
TABLE 4 The ole o in en o y du ing Co id‐19: High e sus low Co id‐19 shock.
This able epo s he i m ixed e ec s panel eg ession es ima es explaining he impac o in en o y on he
esponse o daily s ock e u ns o he g ow h o Co id‐19 cases o wo subsamples: “High”and “Low”Co id‐19
shock based on he sales dec ease in Q1 2020 (Figu e SA1). Fi ms in indus ies ha su e ed a signi ican
dec ease in sales, ha is, consume disc e iona y, ene gy, indus ials and ma e ials (Global Indus y
Classi ica ion S anda d codes 10, 15, 20 and 25, espec i ely) a e iden i ied as “High Co id‐19 shock”. The i ms
in consume s aples, in o ma ion echnology, heal h ca e and communica ion se ices indus ies a e iden i ied
as “Low Co id‐19 shock”. The sample pe iod is om 1 Janua y o 30 Ap il 2020. The dependen a iable is he
daily s ock e u n. Co id‐19 is he g ow h a e o Co id‐19 cases by s a e. In en o y is he a e age o al
in en o y o o al asse s in 2019. In en o y a iable on i s own is abso bed by i m ixed e ec s. Fi m‐le el
con ols include Cash,Le e age,MTB,ROA,Fi m size and Cash low. Re u n ac o con ols include SP500
e u n,Lag e u n,Sha e u no e and Daily ange. All a iables a e de ined in Appendix A.1. In pa en heses,
we epo obus s anda d e o s clus e ed a he i m le el. ***, ** and * indica e signi icance a he 1%, 5% and
10% p obabili y le els, espec i ely.
(1) (2)
High Co id‐19 shock Low Co id‐19 shock
In en o y ×Co id19 −2.736** −0.592
(1.33) (1.68)
Co id19 −1.589** −1.796**
(0.72) (0.63)
Fi m‐le el con ols ×Co id19 Yes Yes
Re u n ac o con ols Yes Yes
Fi m ixed e ec s Yes Yes
Nobse a ions 87,460 111,362
R
2
0.141 0.113
DODD and LIAO EUROPEAN
FINANCIAL MANAGEMENT
|
805
exposed o he dis up ions o global supply chains caused by Co id‐19 (Ha en & Simchi‐Le i,
2020). P ec isis le els o in en o y holdings could bu e agains supply sho ages du ing he
c isis. To es his p oposi ion, we examine i ms ha a e mo e exposed o global supply chain
dis up ions e sus i ms less exposed o global supply chain dis up ions.
O e he pas se e al decades, China has isen as he wo ld's majo ading pa ne . Du ing
he Co id‐19 ou b eak, many ac o ies in China shu down, causing global supply chain dis-
up ions (Ha en & Simchi‐Le i, 2020). The e o e, we expec ha in en o y holdings bene i
i ms ha ely on Chinese supplie s. Fo i ms wi h Chinese supplie s, he bene i s o in en o y
as a hedge agains supply chain dis up ions can o se he nega i e impac o in en o y holdings
due o he demand shock. Fi ms ha do no ha e Chinese supplie s a e less likely o be a ec ed
by global supply chain dis up ions and, he e o e, de i e less alue om in en o y holdings as a
hedge agains supply chain dis up ions.
We use an ex an e measu e o i ms' eliance on Chinese supplie s o cap u e he impac o
supply chain dis up ions. We e e o Hobe g and Moon Tex ‐based O sho ing Ne wo k Da-
abase (Hobe g & Moon, 2017,2019) and de ine i ms ha men ion in hei 10‐K iles “China”
in ela ion o impo ing ac i i ies in he las decade as i ms wi h Chinese supplie s. We use wo
a iables om his da abase: (1) INPUT, which is he numbe o men ions o he i m
TABLE 5 The ole o in en o y componen s du ing he Co id‐19 c isis.
This able epo s he i m ixed e ec s panel eg ession es ima es explaining he impac o di e en
componen s o p ec isis in en o y (Raw ma e ials (In en o y_RM) in Models 1 and 2, wo k‐in‐p og ess
(In en o y_WIP) in Models 3 and 4, and Finished goods (In en o y_FG) in Models 5 and 6 on s ock ma ke
esponse o he g ow h o Co id‐19 cases. The sample pe iod is om 1 Janua y o 30 Ap il 2020. The dependen
a iable is he daily s ock e u n. Co id19 is he g ow h a e o Co id‐19 cases by s a e. Fi m‐le el con ols
include Cash,Le e age,MTB,ROA,Fi m size and Cash low. Re u n ac o con ols include SP500 e u n,Lag
e u n,Sha e u no e and Daily ange. All a iables a e de ined in Appendix A.1. Fi m‐le el a iables a e
abso bed by i m ixed e ec s. In pa en heses, we epo obus s anda d e o s clus e ed a he i m le el. ***, **
and * indica e signi icance a he 1%, 5% and 10% p obabili y le els, espec i ely.
(1) (2) (3) (4) (5) (6)
In en o y_RM ×Co id19 −1.967 −0.631
(2.59) (2.79)
In en o y_WIP ×Co id19 −1.852 0.082
(3.33) (3.55)
In en o y_FG ×Co id19 −2.911** −2.388*
(1.31) (1.43)
Co id19 −1.019*** −1.654*** −1.060*** −1.766*** −0.941*** −1.649***
(0.13) (0.50) (0.13) (0.49) (0.13) (0.49)
Fi m‐le el con ols ×Co id19 No Yes No Yes No Yes
Re u n ac o con ols Yes Yes Yes Yes Yes Yes
Fi m ixed e ec s Yes Yes Yes Yes Yes Yes
Nobse a ions 169,452 160,569 169,039 160,194 171,293 162,324
R
2
0.113 0.111 0.113 0.111 0.114 0.112
806
|
EUROPEAN
FINANCIAL MANAGEMENT
DODD and LIAO
pu chasing inpu s om China, and (2) ININ, which is he numbe o men ions o he i m
pu chasing inpu s om China when he i m also men ions owning asse s in China. We
iden i y one‐ hi d o ou sample i ms wi h nonmissing alues in INPUT and ININ a iables as
wi h Chinese supplie s and he es as wi hou Chinese supplie s.
We es ima e Equa ion (1) o he wo subsamples, (1) i ms wi h Chinese supplie s and (2)
i ms wi hou Chinese supplie s, and epo he es ima ion esul s in Table 6. Models 1 and 2
p esen he eg ession es ima es o he wo subsamples based on he ull sample. We obse e
ha he “wi hou Chinese supplie s”subsample size is wice as la ge as ha o “wi h Chinese
supplie s”. To mi iga e he impac o unbalanced subsamples, we e un he es ima ion using a
ma ched sample. We ma ch each i m “wi h Chinese supplie s”wi h a i m “wi hou Chinese
supplie s”based on hei GICS indus y sec o code, Cash holdings,Fi m size,MTB a io, ROA
and Le e age (de ined in Appendix A.1). Models 3 and 4 o Table 6p esen he es ima ion
TABLE 6 The ole o in en o y du ing he Co id‐19 c isis o i ms wi h and wi hou Chinese supplie s.
This able epo s he i m ixed e ec s panel eg ession es ima es explaining he impac o in en o y on he
esponses o daily s ock e u ns o he g ow h a e o Co id‐19 cases o wo subsamples: (1) i ms wi h Chinese
supplie s and (2) i ms wi hou Chinese supplie s. We classi y a i m as wi h Chinese supplie s i i men ions
China in i s 10‐K ile in ela ion o impo ing ac i i ies, ha is, he i m has nonmissing alues in INPUT and
ININ o China in Hobe g and Moon Tex ‐based O sho ing Ne wo k Da abase (Hobe g & Moon, 2017,2019).
We classi y he es o he i ms as “wi hou Chinese supplie s”. Models 1 and 2 p esen eg ession es ima es o
he wo subsamples based on he ull sample. Models 3 and 4 p esen he es ima es o he subsamples o i ms
ma ched based on he Global Indus y Classi ica ion S anda d indus y sec o code and Cash holdings,Fi m size,
MTB a io, ROA and Le e age (de ined in Appendix A.1). The sample pe iod is om 1 Janua y o 30 Ap il 2020.
The dependen a iable is he daily s ock e u n. Co id19 is he g ow h a e o Co id‐19 cases by s a e. In en o y
is he a e age o al in en o y o o al asse s in 2019. In en o y a iable on i s own is abso bed by i m ixed
e ec s. Fi m‐le el con ols include Cash,Le e age,MTB,ROA,Fi m size and Cash low. Re u n ac o con ols
include SP500 e u n,Lag e u n,Sha e u no e and Daily ange. All a iables a e de ined in Appendix A.1.In
pa en heses, we epo obus s anda d e o s clus e ed a he i m le el. ***, ** and * indica e signi icance a he
1%, 5% and 10% p obabili y le els, espec i ely.
(1) (2) (3) (4)
Full sample Ma ched sample
wi h Chinese
supplie s
wi hou Chinese
supplie s
wi h Chinese
supplie s
wi hou Chinese
supplie s
In en o y ×Co id19 0.028 −2.877** −3.365** 0.028
(2.27) (1.16) (1.39) (2.27)
Co id19 −3.068*** −1.146** −0.438 −3.068***
(1.04) (0.56) (0.98) (1.04)
Fi m‐le el con ols ×Co id19 Yes Yes Yes Yes
Re u n ac o con ols Yes Yes Yes Yes
Fi m ixed e ec s Yes Yes Yes Yes
Nobse a ions 63,487 135,332 63,487 63,337
R
2
0.164 0.100 0.136 0.164
DODD and LIAO EUROPEAN
FINANCIAL MANAGEMENT
|
807
esul s o he subsample “wi h Chinese supplie s”and he ma ched “wi hou Chinese supplie s”
subsample, espec i ely.
The analysis based on he ull sample (Models 1 and 2) and he ma ched sample
(Models 3 and 4) shows ha he nega i e impac o in en o y on he s ock ma ke esponse
o he Co id‐19 c isis is mo e p onounced o i ms wi hou Chinese supplie s han o i ms
wi h Chinese supplie s. This inding is consis en wi h ou p edic ion ha i ms wi hou
Chinese supplie s gain less om in en o y holdings as a hedge agains supply chain dis-
up ions. Howe e , o i ms wi h Chinese supplie s ha a e exposed o global supply chain
dis up ions in he i s pa o 2020, he nega i e impac o in en o y holdings due o he
demand shock is o se by he posi i e alue o in en o y as a hedge agains supply chain
dis up ions.
5.1.3 |Robus ness es s
Al e na e measu es o in en o y
Ou in en o y holdings a iable is he in en o y‐ o‐asse s a io, which is widely used in
inance li e a u e (e.g., Ca pen e e al., 1994; Dasgup a e al., 2019;Kulchania&
Thomas, 2017).Asa obus ness es ,we e‐es ima e he baseline eg ession wi h di e en
measu es o in en o y holdings, ollowing Chen e al. (2005). Fi s , we conside he
in en o y‐ o‐sales a io (In en o y_sales), calcula ed as he o al in en o y di ided by sales;
his a io ma e s mos o s ockou . Second, we calcula e he in en o y‐days a io (In-
en o y_days) as 365 imes he in en o y di ided by he cos s o goods sold; his a io
measu es how many days i akes o u n o e he in en o y in o cos s o goods sold and
indica es in en o y managemen e iciency.Thi d,wees ima eabno malin en o y(In-
en o y_abno mal) based on a no malised in en o y‐ o‐asse s a io o accoun o he
indus y‐and i m size‐d i en di e ences ha may a ec in en o y holdings. We so ou
sample i ms based on i m size in o quin iles and compu e In en o y_abno mal as he
de ia ion o he i m's in en o y om he minimum alue o i ms' in en o y in he same
GICS indus y sec o and i m size quin ile, di ided by he dis ance be ween he maximum
and he minimum alue (min–max no malisa ion).
Table 7 epo s he es ima ion esul s wi h he al e na i e in en o y measu es. The
coe icien es ima es on he in e ac ion e m o in en o y measu es wi h Co id19 emain
nega i e and s a is ically signi ican in all models. I indica es ha ou esul s a e obus o
using al e na i e measu es o in en o y holdings.
Placebo es
In en o y holdings in he p e ious yea could be nega i ely co ela ed wi h i ms' g ow h
oppo uni ies and s ock ma ke pe o mance in he ollowing yea , i espec i e o he Co id‐19
pandemic. To add ess his conce n, we un Co id‐19 “expe imen s”a ound placebo (a andom
nonc isis) pe iods assigned in yea s p eceding he Co id‐19 c isis. Suppo ing In o ma ion
Appendix SA.3 epo s he placebo es de ails and he es ima ion esul s. We ind ha he
nega i e e ec s o in en o y holdings do no appea in nonc isis yea s when he e a e no
nega i e demand shocks. The e o e, we can ule ou he explana ion ha some unobse able
i m cha ac e is ics d i e he nega i e ela ionship be ween p ec isis in en o y holdings and
s ock ma ke esponse o Co id‐19 in Janua y–Ma ch 2020.
808
|
EUROPEAN
FINANCIAL MANAGEMENT
DODD and LIAO
5.2 |Longe ‐ un analysis o he Co id‐19 pandemic in 2020
In he i s pa o ou analysis, we ha e es ablished ha in Janua y–Ap il 2020, in en o y
holdings ha e a nega i e alue o i ms due o he signi ican d op in consume demand. In
his sec ion, we ex end ou analysis o include he ull yea 2020 and examine he impac o
in en o y holdings on i m pe o mance in May–Decembe 2020. In he longe ‐ un analysis, we
use he same mul i a ia e analysis amewo k as in he sho ‐ un analysis. We es ima e
Equa ion (1) and epo he es ima ion esul s in Table 8.
5.2.1 |In en o y holdings and s ock e u ns in May–Decembe 2020
Model 1 o Table 8p esen s he es ima ion esul s o he eg ession o daily s ock e u ns o all
sample i ms using he sample pe iod om 1 Janua y o 31 Decembe 2020. In his pa o he
TABLE 7 Al e na i e measu es o in en o y.
This able epo s he i m ixed e ec s eg ession es ima es explaining he impac o in en o y on he esponse
o s ock ma ke e u ns o he Co id‐19 c isis. The sample pe iod is om 1 Janua y o 30 Ap il 2020. The
dependen a iable is he daily s ock e u n. Co id19 is he g ow h a e o Co id‐19 cases by s a e.
In en o y_sales is he a io o o al in en o y o sales. In en o y_days is he numbe o days i akes o he
in en o y o u n o e and is calcula ed as 365 imes he o al in en o y di ided by he cos s o goods sold.
In en o y_abno mal is he a io o o al in en o y o o al asse s no malised by indus y wi hin he i m size
quin ile. All in en o y a iables a e calcula ed as he a e age alues in 2019. All in en o y a iables on hei
own a e abso bed by i m ixed e ec s. Fi m‐le el con ols include Cash,Le e age,MTB,ROA,Fi m size and
Cash low. Re u n ac o con ols include SP500 e u n,Lag e u n,Sha e u no e and Daily ange. All a iables
a e de ined in Appendix A.1. In pa en heses, we epo obus s anda d e o s clus e ed a he i m le el. ***, **
and * indica e signi icance a he 1%, 5% and 10% p obabili y le els, espec i ely.
(1) (2) (3)
In en o y_sale ×Co id19 −1.074*
(0.65)
In en o y_days ×Co id19 −0.002***
(0.00)
In en o y_abno mal ×Co id19 −0.978**
(0.50)
Co id19 −1.783*** −1.772*** −1.649***
(0.46) (0.45) (0.48)
Fi m‐le el con ols ×Co id19 Yes Yes Yes
Re u n ac o con ols Yes Yes Yes
Fi m ixed e ec s Yes Yes Yes
Nobse a ions 203,930 198,822 202,095
R
2
0.116 0.123 0.116
DODD and LIAO EUROPEAN
FINANCIAL MANAGEMENT
|
809
Helpe , S., & Sol as, E. (2021). Why he pandemic has dis up ed supply chains. The Whi e House Blog, 17 June
2021. h ps://www.whi ehouse.go /cea/w i en-ma e ials/2021/06/17/why- he-pandemic-has-dis up ed-
supply-chains/
Hend icks, K. B., & Singhal, V. R. (2003). The e ec o supply chain gli ches on sha eholde weal h. Jou nal o
Ope a ions Managemen ,21(5), 501–522.
Hend icks, K. B., & Singhal, V. R. (2005a). Associa ion be ween supply chain gli ches and ope a ing pe o m-
ance. Managemen Science,51(5), 695–711.
Hend icks, K. B., & Singhal, V. R. (2005b). An empi ical analysis o he e ec o supply chain dis up ions on
long‐ un s ock p ice pe o mance and equi y isk o he i m. P oduc ion & Ope a ions Managemen ,14(1),
35–52.
Hi shlei e , D., & Jiang, D. (2010). A inancing‐based mis alua ion ac o and he c oss‐sec ion o expec ed
e u ns. Re iew o Financial S udies,23(9), 3401–3436.
Hobe g, G., & Moon, S. K. (2017). O sho e ac i i ies and inancial s ope a ional hedging. Jou nal o Financial
Economics,125(2), 217–244.
Hobe g, G., & Moon, S. K. (2019). The o sho ing e u n p emium. Managemen Science,65(6), 2876–2899.
Kashyap, A. K., Lamon , O. A., & S ein, J. C. (1994). C edi condi ions and he cyclical beha io o in en o ies.
The Qua e ly Jou nal o Economics,109(3), 565–592.
Kulchania, M., & Thomas, S. (2017). Cash ese es as a hedge agains supply‐chain isk. Jou nal o Financial &
Quan i a i e Analysis,52(5), 1951–1988.
La Londe, B. J., & Lambe , D. M. (1977). A me hodology o calcula ing in en o y ca ying cos s. In e na ional
Jou nal o Physical Dis ibu ion,7(4), 193–231.
Meie , M., & Pin o, E. (2020). Co id‐19 supply chain dis up ions. Co id Economics,48, 139–170.
Ozili, P. K., & A un, T. (2020). Spillo e o COVID‐19: Impac on he global economy. In Managing in la ion and
supply chain dis up ions in he global economy (pp. 41–61). IGI Global.
Pa kinson, M. (1980). The ex eme alue me hod o es ima ing he a iance o he a e o e u n. Jou nal o
Business,61–65.
P a e , E., Biehl, M., & Smi h, M. A. (2001). In e na ional supply chain agili y‐ adeo s be ween lexibili y and
unce ain y. In e na ional Jou nal o Ope a ions & P oduc ion managemen ,21(5–6), 823–839.
Rajagopalan, S., & Malho a, A. (2001). Ha e US manu ac u ing in en o ies eally dec eased? An empi ical
s udy. Manu ac u ing & Se ice Ope a ions Managemen ,3(1), 14–24.
Ramelli, S., & Wagne , A. F. (2020). Fe e ish s ock p ice eac ions o COVID‐19. The Re iew o Co po a e
Finance S udies,9(3), 622–655.
Snyde , L. V., A an, Z., Peng, P., Rong, Y., Schmi , A. J., & Sinsoysal, B. (2016). OR/MS models o supply chain
dis up ions: A e iew. IIE T ansac ions,48(2), 89–109.
The Economis . (2020). Co id ca nage: Much o global comme ce has g ound o a hal . The Economis , Ma ch 21.
s h ps://www.economis .com/business/2020/03/21/much-o -global-comme ce-has-g ound- o-a-hal
Tomlin, B. (2006). On he alue o mi iga ion and con ingency s a egies o managing supply chain dis up ion
isks. Managemen Science,52(5), 639–657.
Wang, B. (2019). The cash con e sion cycle sp ead. Jou nal o Financial Economics,133(2), 472–497.
Wen, Y. (2005). Unde s anding he in en o y cycle. Jou nal o Mone a y Economics,52(8), 1533–1555.
SUPPORTING INFORMATION
Addi ional suppo ing in o ma ion can be ound online in he Suppo ing In o ma ion sec ion
a he end o his a icle.
How o ci e his a icle: Dodd, O., Liao, S. (2025). The ole o in en o y in i m
esilience o he Co id‐19 pandemic. Eu opean Financial Managemen ,31, 786–818.
h ps://doi.o g/10.1111/eu m.12517
816
|
EUROPEAN
FINANCIAL MANAGEMENT
DODD and LIAO
APPENDIX A1: VARIABLE DEFINITIONS
The able p o ides he de ini ion o he a iables. Compus a i ems a e in i alic on .
Va iable Va iable de ini ion
In en o y To al in en o y (in ) di ided by o al asse s (a ), whe e o al in en o y includes
aw ma e ials, inished goods, wo k‐in‐p og ess and o he in en o y. The a iable
is he a e age o he beginning‐and end‐o ‐yea in en o y a io alues in 2019.
Co id19 The daily g ow h a e o Co id‐19 cases by s a e in Janua y–Ap il 2020,
measu ed as [log(1 + #Cases
)−log(1 + #Cases
−1
)].
Re u n Daily s ock log e u n. S ock p ices a e adjus ed o di idends using he daily
mul iplica ion ac o and he p ice adjus men ac o s p o ided by Compus a .
Buy‐and‐hold abno mal
e u n
Accumula ed s ock e u ns minus accumula ed expec ed e u ns o e he
pe iod o Janua y–Ap il 2020. Expec ed e u ns a e compu ed using
Fama–F ench and Ca ha ou ‐ ac o model wi h coe icien s es ima ed using
s ock e u ns in he yea 2019.
Cash Cash and ma ke able secu i ies (che) di ided by o al asse s (a ). The a iable is
he a e age o he beginning‐and end‐o ‐yea alues in 2019.
Le e age The sum o o al long‐ e m deb (dl ) and deb in cu en liabili ies (dlcc) scaled
by o al asse s (a ). The a iable is he a e age o he beginning‐and end‐o ‐yea
alues in 2019.
MTB The ma ke alue o asse s di ided by book alue o o al asse s (a ), whe e he
ma ke alue o asse s is calcula ed as o al asse (a ) plus he ma ke alue o
common equi y (p cc_ ×csho) minus he book alue o common equi y (ceq),
and minus de e ed axes ( xdb). The a iable is he a e age o he beginning‐
and end‐o ‐yea alues in 2019.
ROA Ope a ing income be o e dep ecia ion (oibdp) di ided by o al asse s (a ). The
a iable is he a e age o he beginning‐and end‐o ‐yea alues in 2019.
Fi m size The na u al loga i hm o o al asse s (a ). The a iable is he a e age o he
beginning‐and end‐o ‐yea alues in 2019.
Cash low Income be o e ex ao dina y i ems (ib) plus dep ecia ion and amo isa ion (dp)
di ided by o al asse s (a ). The a iable is he a e age o he beginning‐and
end‐o ‐yea alues in 2019.
SP500 e u n Re u ns on he S&P500 index.
Lag e u n Re u n om he p e ious day.
Sha e u no e T ading olumes scaled by o al sha es ou s anding.
Daily ange Di e ence be ween he daily high p ice and daily low p ice, scaled by he
closing s ock p ice.
In en o y_RM Raw ma e ials (in m) di ided by o al asse s (a ). The a iable is he a e age o
he beginning‐and end‐o ‐yea alues in 2019.
In en o y_FG Finished goods (in g) di ided by o al asse s (a ). The a iable is he a e age o
he beginning‐and end‐o ‐yea alues in 2019.
In en o y_WIP Wo k‐in‐p og ess (in wp) di ided by o al asse s (a ). The a iable is he
a e age o he beginning‐and end‐o ‐yea alues in 2019.
(Con inues)
DODD and LIAO EUROPEAN
FINANCIAL MANAGEMENT
|
817
Va iable Va iable de ini ion
In en o y_sale To al in en o y (in ) di ided by sales (sale). The a iable is he a e age o he
beginning‐and end‐o ‐yea alues in 2019.
In en o y_days 365 imes To al in en o y (in ) di ided by he cos s o goods sold (cogs). The
a iable is he a e age o he beginning‐and end‐o ‐yea alues in 2019.
In en o y_abno mal The de ia ion o he i m's o al in en o y (in ) om he minimum alue o
in en o y in he same Global Indus y Classi ica ion S anda d indus y sec o
and i m size quin ile, di ided by he dis ance be ween he maximum and he
minimum alues. The a iable is he a e age o he beginning‐and end‐o ‐yea
alues in 2019.
SCD_10K The o al numbe o “supply chain”men ions in he i m's 10‐K ile du ing he
ull yea 2020.
ROA_q Qua e ly seasonally adjus ed e u n on asse s.
Sales g ow h_q Qua e ly seasonally adjus ed pe cen age changes in sales (saleq).
818
|
EUROPEAN
FINANCIAL MANAGEMENT
DODD and LIAO
