Resilience to shocks of micro, small, and medium-sized enterprises in Fiji

G eenland, William; To h, Russell
Wo king Pape
Resilience o shocks o mic o, small, and medium-sized
en e p ises in Fiji
ADB Economics Wo king Pape Se ies, No. 765
P o ided in Coope a ion wi h:
Asian De elopmen Bank (ADB), Manila
Sugges ed Ci a ion: G eenland, William; To h, Russell (2025) : Resilience o shocks o mic o, small,
and medium-sized en e p ises in Fiji, ADB Economics Wo king Pape Se ies, No. 765, Asian
De elopmen Bank (ADB), Manila,
h ps://doi.o g/10.22617/WPS250006-2
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/310445
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 ps://c ea i ecommons.o g/licenses/by/3.0/igo/
ASIAN DEVELOPMENT BANK
ASIAN DEVELOPMENT BANK
6 ADB A enue, Mandaluyong Ci y
1550 Me o Manila, Philippines
www.adb.o g
RESILIENCE TO SHOCKS
OF MICRO, SMALL,
AND MEDIUM-SIZED
ENTERPRISES IN FIJI
William G eenland and Russell To h
ADB ECONOMICS
WORKING PAPER SERIES
NO. 765
Janua y 2025
Resilience o Shocks o Mic o, Small, and Medium-Sized En e p ises in Fiji
This pape examines he esilience o mic o, small, and medium-sized en e p ises (MSMEs) in Fiji o
economic shocks, ocusing on he COVID-19 pandemic and clima e- ela ed haza ds. Based on o iginal
su ey da a om 2,400 MSMEs, he pape e alua es he impac o go e nmen concessional loans on
pandemic eco e y and explo es clima e esilience s a egies. Findings e eal mixed ou comes om policy
in e en ions and eliance on sel - unding by en e p ises impac ed by disas e s. The pape emphasizes
in eg a ing inancial suppo wi h long- e m clima e adap a ion o s eng hen MSME esilience.
Abou he Asian De elopmen Bank
ADB is commi ed o achie ing a p ospe ous, inclusi e, esilien , and sus ainable Asia and he Paci ic,
while sus aining i s e o s o e adica e ex eme po e y. Es ablished in 1966, i is owned by 69 membe s
—49 om he egion. I s main ins umen s o helping i s de eloping membe coun ies a e policy dialogue,
loans, equi y in es men s, gua an ees, g an s, and echnical assis ance.
ASIAN DEVELOPMENT BANK
The ADB Economics Wo king Pape Se ies
p esen s esea ch in p og ess o elici commen s
and encou age deba e on de elopmen issues
in Asia and he Paci ic. The iews exp essed
a e hose o he au ho s and do no necessa ily
e lec he iews and policies o ADB o
i s Boa d o Go e no s o he go e nmen s
hey ep esen .
ADB Economics Wo king Pape Se ies
Resilience o Shocks o Mic o, Small, and Medium-Sized
En e p ises in Fiji
William G eenland and Russell To h
No. 765 | Janua y 2025
William G eenland (william.g [email protected])
is a law s uden and Russell To h ( ussell. o h@sydney.
edu.au) is an associa e p o esso a he Uni e si y o
Sydney.
C ea i e Commons A ibu ion 3.0 IGO license (CC BY 3.0 IGO)
© 2025 Asian De elopmen Bank
6 ADB A enue, Mandaluyong Ci y, 1550 Me o Manila, Philippines
Tel +63 2 8632 4444; Fax +63 2 8636 2444
www.adb.o g
Some igh s ese ed. Published in 2025.
ISSN 2313-6537 (p in ), 2313-6545 (PDF)
Publica ion S ock No. WPS250006-2
DOI: h p://dx.doi.o g/10.22617/WPS250006-2
The iews exp essed in his publica ion a e hose o he au ho s and do no necessa ily e lec he iews and policies
o  he Asian De elopmen Bank (ADB) o i s Boa d o Go e no s o he go e nmen s hey ep esen .
ADB does no gua an ee he accu acy o he da a included in his publica ion and accep s no esponsibili y o any
consequence o hei use. The men ion o speci ic companies o p oduc s o manu ac u e s does no imply ha hey
a e endo sed o ecommended by ADB in p e e ence o o he s o a simila na u e ha a e no men ioned.
By making any designa ion o o e e ence o a pa icula e i o y o geog aphic a ea in his documen , ADB does no
in end o make any judgmen s as o he legal o o he s a us o any e i o y o a ea.
This publica ion is a ailable unde he C ea i e Commons A ibu ion 3.0 IGO license (CC BY 3.0 IGO)
h ps://c ea i ecommons.o g/licenses/by/3.0/igo/. By using he con en o his publica ion, you ag ee o be bound
by he e ms o his license. Fo a ibu ion, ansla ions, adap a ions, and pe missions, please ead he p o isions
and e ms o use a h ps://www.adb.o g/ e ms-use#openaccess.
This CC license does no apply o non-ADB copy igh ma e ials in his publica ion. I he ma e ial is a ibu ed
oano he sou ce, please con ac he copy igh owne o publishe o ha sou ce o pe mission o ep oduce i .
ADB canno be held liable o any claims ha a ise as a esul o you use o he ma e ial.
Please con ac pubsma ke [email protected] g i you ha e ques ions o commen s wi h espec o con en , o i you wish
oob ain copy igh pe mission o you in ended use ha does no all wi hin hese e ms, o o pe mission o use
heADB logo.
Co igenda o ADB publica ions may be ound a h p://www.adb.o g/publica ions/co igenda.
No e:
In his publica ion, “F$” e e s o Fiji dolla s.
ABSTRACT
This pape examines he esilience o mic o, small, and medium-sized en e p ises
(MSMEs) in Fiji agains he economic impac s o majo shocks including he co ona i us
disease (COVID-19) pandemic and clima e- ela ed haza ds, using a su ey o 2,400
MSMEs om ea ly 2023. The analysis ocuses on he e ec i eness o go e nmen
concessional loans in aiding pandemic eco e y and e alua es MSMEs’ s a egies o
clima e esilience. Findings indica e ha while go e nmen suppo has been pi o al in
mi iga ing pandemic- ela ed economic down u ns, i s e ec i eness is a ied, highligh ing
he necessi y o accessible and a ge ed inancial suppo . Fu he mo e, expe iences wi h
na u al haza ds signi ican ly in luence MSMEs’ u u e esilience s a egies, wi h a no able
p e e ence o sel - unding eco e y e o s among MSMEs ha had expe ienced a
signi ican clima e- ela ed disas e . The s udy unde sco es he impo ance o in eg a ing
immedia e inancial suppo wi h long- e m clima e adap a ion s a egies o MSMEs,
o e ing insigh s o policy o mula ion aimed a enhancing economic esilience in
eme ging ma ke s in he Paci ic and beyond.
Keywo ds: mic o, small, and medium-sized en e p ises; COVID-19 pandemic;
concessional loans; clima e esilience; Fiji
JEL codes: Q54, L53, O56
We hank he Asian De elopmen Bank (ADB) Paci ic P i a e Sec o De elopmen Ini ia i e (PSDI), he Fiji
Bu eau o S a is ics (FBoS), and Rese e Bank o Fiji (RBF) and Inno a ions o Po e y Ac ion (IPA) o
hei suppo and collabo a ion on his s udy, especially Je emy Clea e , Miel de Villa, Ma hew Hodge, and
Simon Thompson a ADB/PSDI; Mohamed Hakim, Bimlesh K ishna, Ma ia Masud oka and he Household
Su ey Di ision and he Business Regis y Uni a FBoS; Yuna Liang, Ma ieliz Maines, and Maxine Tong a
IPA; La u Se a Kaukilakeba, Ese ani Muni ai, Sakiusa Nabou, Ch is ina Rokoua, and Lepani Uluina i i a
RBF. E hical e iew o he p ojec was p o ided by he IPA Ins i u ional Re iew Boa d unde p ojec numbe
15524. We acknowledge unding o his s udy om he ADB Impac E alua ion Commi ee and he ADB
PSDI. The analysis and in e p e a ions in his s udy a e exclusi ely he esponsibili y o he au ho s and may
no necessa ily e lec he opinions o posi ions o he unde s o he ins i u ions he au ho s a e a ilia ed wi h.
Co esponding au ho : Russell To h ([email p o ec ed]).

1 INTRODUCTION
Mic o, small, and medium-sized en e p ises (MSMEs) con ibu e signi ican ly o
employmen , income gene a ion, and o e all economic g ow h ac oss a numbe o
eme ging ma ke s, including in Fiji. As o 2020, MSMEs accoun o 82.4% o all
en e p ises in Fiji and abso b 28.3% o he labo o ce (ADB 2023). Howe e , MSMEs in
Fiji can ace signi ican challenges om majo shocks, which can h ea en hei su i al
and he esilience o he MSME sec o . The co ona i us disease (COVID-19) pandemic
p o ided a c i ical shock o en e p ises in Fiji, as he coun y was isola ed o a signi ican
pe iod o ime, dis up ing he ou ism sec o , impo /expo ac i i ies, and domes ic
consump ion. As a small island economy, Fiji is also ecognized o u gen ly ace he
inc easing h ea o clima e change- ela ed e en s, including cyclones and looding. These
isks no only dis up business ope a ions bu also h ea en he li elihoods o hose
dependen on hese sec o s, he eby ampli ying hei ulne abili y o economic shocks.
Unde s anding he esilience mechanisms o MSMEs in Fiji agains such shocks is c i ical
o c a ing policies ha ensu e hei sus ainabili y and, by ex ension, he b oade
economic s abili y o he coun y.
This pape seeks o explo e he esilience o MSMEs in Fiji, ocusing on hei esponses
o majo shocks including he COVID-19 pandemic and clima e change- ela ed disas e s. I
d aws on an o iginal su ey o MSMEs conduc ed in he i s hal o 2023, based on a sample
d awn om he mos comp ehensi e lis ing o o mally- egis e ed MSMEs in Fiji. The su ey
in ol ed a sho phone su ey conduc ed wi h se e al housand MSMEs, ollowed by an
in-pe son su ey o 2,400 MSMEs. The su eys we e a ge ed o co e en e p ises in, and
in he icini y o , he h ee la ges owns in Fiji—Lau oka, Nadi, and Su a—on he main
island o Vi i Le u. These geog aphic egions con ain 80% o o mally egis e ed MSMEs,
and so ou sample, hough geog aphically- a ge ed, is la gely ep esen a i e o MSMEs
ac oss Fiji. The sample also sligh ly o e samples women-owned and ope a ed MSMEs.
Fi s , we ocus on he impac s o he COVID-19 pandemic on MSMEs in Fiji. We ind
ha en e p ises we e s ill eeling he economic e ec s o he COVID-19 pandemic as o
he i s hal o 2023, as jus 8.3% o en e p ises a ec ed by he pandemic said ha hey
had comple ely eco e ed, wi h 65% saying hey had pa ly eco e ed. We hen p o ide
e idence on he impac o he majo in e en ion om he Go e nmen o Fiji o suppo
MSMEs du ing he pandemic, he COVID-19 Concessional Finance Suppo Package.
While MSME access o bank c edi is adi ionally small in Fiji, accoun ing o jus 10.9% o
o al bank lending and 7.7% o g oss domes ic p oduc (GDP) as o 2022, he subsidized
loans and c edi gua an ees caused bank lending o MSMEs o inc ease sha ply in 2021,
o eco d le els (ADB 2023).
Focusing i s on a ge ing, we ind ha he loans we e accessed mos ly by ela i ely
la ge en e p ises wi h g ea e colla e al holdings and p io access o capi al, o by
en e p ises mo e deeply a ec ed by COVID-19. The po en ial o eg essi e a ge ing is
plausible gi en he con ex o he p og am–banks we e deluged wi h housands o
applica ions o concessional loans ela i ely ea ly in he pandemic, and had li le
oppo uni y o sc een bo owe s apa om e iewing loan applica ion pape wo k. I is
plausible ha lending would a o MSMEs wi h a lowe isk p o ile. Focusing on he subse
o MSMEs ha exis ed p io o he onse o he pandemic in 2020, we hen use p opensi y
sco e ma ching o c ea e s a is ically-compa able g oups o MSMEs ha did and did no
ecei e access o concessional loans. We ind ha he concessional loans may ha e
helped soo he he impac s o he pandemic o some en e p ises, wi h en e p ises ha
accessed loans 3.4 pe cen age poin s mo e likely o ha e comple ely o somewha
eco e ed, hough his es ima e is no s a is ically signi ican . I is possible ha since a
signi ican amoun o he inancing wen o ela i ely be e -es ablished en e p ises, while
mo e ulne able en e p ises we e mo e likely o apply in gene al, ha selec ion neu alizes
he in ended impac o he p og am.
Second, we p o ide e idence on clima e esilience o MSMEs in Fiji, ocusing on
di e ences be ween en e p ises ha had and hadn’ p e iously expe ienced a signi ican
clima e- ela ed shock. No ably, we ind ha hese wo g oups ha e di e en expec a ions
abou mechanisms o eco e om a clima e shock. In pa icula , en e p ises ha had
no ye expe ienced a majo clima e shock a e much mo e op imis ic abou he ole o
go e nmen suppo in aiding hei eco e y om a p ospec i e u u e shock, compa ed
o hose ha had expe ienced such an e en . Those ha had p e ious expe ience wi h a
disas e a e mo e likely o expec o ely on hei own sa ings o eco e om such a shock.
This s udy con ibu es o b oade li e a u es on he impac s and esponses o he
COVID-19 pandemic in eme ging ma ke s, he esilience o MSMEs o na u al haza ds ha
may be inc easing in ligh o clima e change, and esponses o majo economic shocks o
he MSME sec o in Fiji.
Fi s , he COVID-19 pandemic has had p o ound impac s on MSMEs in eme ging
ma ke economies, necessi a ing a a ie y o policy esponses. One s and o li e a u e
p o ides p ospec i e analysis ea ly in he pandemic, p o iding ad ice on iscal and mone a y
policies o help suppo economic ac o s, including MSMEs (Baldwin and Wede di Mau o
2020; Gou inchas, Philippon, and Vayanos 2021). O he s documen he impac s o he
COVID-19 pandemic on MSMEs in speci ic eme ging economies (e.g., Sha i, Liu, and
2
Ren 2020), hough much o he e idence is qui e desc ip i e, while o he s documen
en ep eneu ial esponses o he pandemic (e.g., Kucke z e al. 2020). Finally, he e is
a small li e a u e epo ing on go e nmen inancing p og ams in ended o mi iga e he
impac s o he pandemic on MSMEs. Ce da e al. (2023) conduc wo andomized con olled
ials o e alua e he impac o go e nmen -gua an eed loans on MSMEs ha had applied
o loans ea lie in he pandemic, bu p e iously been ejec ed o loans om banks in Chile
and Colombia. They ind ha he loans had ake-up a es o 27% and 29%, espec i ely,
and cause a signi ican posi i e impac on MSMEs’ o al liquidi y wi h he o mal banking
sys em. Ou s udy con ibu es o his li e a u e by p o iding e idence on pandemic impac s
on MSMEs om a small island economy. I also e alua es he ole o a concessional
lending p og am on mi iga ing he impac s o he pandemic, including by p o iding some o
he i s global e idence on he a ge ing o such a p og am.
Second, he impac s o clima e change, especially clima e- ela ed na u al haza ds, on
MSMEs in eme ging ma ke economies ha e been inc easingly s udied wi hin he
economic li e a u e. In seminal wo k, Hallega e and Rozenbe g (2017) explo e he
economic ulne abili y o MSMEs o clima e change, emphasizing he disp opo iona e
impac o disas e s on hese en i ies due o hei limi ed capaci y o abso b and eco e
om shocks. They ad oca e o enhanced inancial and echnical suppo mechanisms o
MSMEs, alongside imp o ed access o in o ma ion ega ding clima e isks. Noy and
Yonson (2018) ocus on he ole o insu ance in economic eco e y ollowing disas e s.
They a gue ha accessible insu ance p oduc s ailo ed o he needs o MSMEs can play a
c i ical ole in hei abili y o wi hs and and eco e om he e ec s o clima e- ela ed
haza ds. These s udies collec i ely unde sco e he u gen need o comp ehensi e
s a egies ha include inancial p oduc s, policy in e en ions, and clima e in o ma ion
dissemina ion o sa egua d MSMEs in eme ging ma ke s agains he ad e se e ec s o
clima e change. We con ibu e o his li e a u e by p o iding e idence om a
comp ehensi e sample o MSMEs in an eme ging ma ke , on pas impac s o na u al
haza ds, and how pas expe ience o na u al haza ds impac s expec a ions o mi iga ion
and eco e y mechanisms.
Finally, we p o ide new e idence on he impac s o economic shocks on MSMEs
in Fiji, and policies o mi iga e disas e isk. MSMEs in Fiji ha e aced signi ican isks
and shocks, pa icula ly om he COVID-19 pandemic and clima e change. These dual
c ises p esen compounded challenges o i s economy, hea ily elian on ou ism and
ag icul u e—sec o s especially ulne able o bo h pandemics and clima e- ela ed e en s.
A s udy by Kuma and Singh (2020) assesses he pandemic’s immedia e economic impac
in Fiji, no ing signi ican declines in e enue and employmen wi hin MSMEs. They s ess
3
he impo ance o go e nmen suppo p og ams o sus ain businesses h ough he c isis.
A epo by he ADB (2019) ou lines he ulne abili y o he Paci ic, including Fiji, o clima e
change, emphasizing he need o adap i e s a egies o sa egua d he u u e o MSMEs
in hese egions. We con ibu e o he li e a u e on he impac s o economic shocks on
MSMEs in Fiji by p o iding new e idence on he impac s and mi iga ion o he COVID-19
pandemic, and esponses o na u al haza ds, om a comp ehensi e su ey o MSMEs in
Fiji.
In Sec ion 2, we p o ide backg ound in o ma ion on he economic impac s o COVID-19
in Fiji, he go e nmen ’s loan assis ance o MSMEs ha was in ended o mi iga e he
impac s o he pandemic, and he p ima y clima e haza ds aced by MSMEs in Fiji. In
Sec ion 3, we desc ibe ou p ima y da a sou ce, a su ey o MSMEs, and p o ide desc ip i e
e idence on COVID-19 impac s, concessional loan a ge ing, and clima e haza ds aced
by MSMEs in ou sample. In Sec ion 4, we desc ibe ou me hodology o iden i ying he
causal impac s o COVID-19 concessional loans on pandemic eco e y, and he impac s o
clima e e en s on MSMEs’ mi iga ion s a egies, while in Sec ion 5, we epo on he main
indings. In Sec ion 6, we p o ide a b ie discussion and policy applica ion o ou indings,
and in Sec ion 7, we p o ide concluding ema ks. Figu es, ables, and some addi ional
in o ma ion is p o ided in he supplemen al Appendix.
2 BACKGROUND AND CONTEXT
In his sec ion we p o ide impo an backg ound in o ma ion ha will in o m he la e sec ions.
Fi s , we b ie ly desc ibe he economic impac s o he COVID-19 pandemic in Fiji. We
hen desc ibe he Go e nmen o Fiji’s COVID-19 Concessional Finance Suppo Package,
which was designed o help mi iga e he impac s o he pandemic on MSMEs. Finally, we
p o ide in o ma ion on he p ima y clima e haza ds and isks aced by MSMEs in Fiji.
2.1 Impac s o he COVID-19 Pandemic in Fiji
The COVID-19 pandemic deli e ed a signi ican blow o Fiji’s economy, p ima ily due o
he coun y’s hea y eliance on ou ism, which accoun s o abou a hi d o GDP and a
simila p opo ion o employmen . The i s case o COVID-19 in Fiji was epo ed in Ma ch
2020, leading o immedia e and s ingen measu es o con ain he i us’s sp ead. The
Go e nmen o Fiji implemen ed i s i s lockdown in Lau oka, one o he coun y’s majo
ci ies, in Ma ch 2020, ollowed by se e al localized lockdowns as cases we e de ec ed in
di e en pa s o he coun y. In addi ion o in e nal es ic ions, Fiji closed i s in e na ional
4
accessing a loan p o ided h ough his suppo package. Figu e A2 demons a es ha
hose en e p ises ha ook ou such a loan ended o be abou 25% mo e likely o ha e
laid o wo ke s, and abou wice as likely o ha e sold asse s, o cope wi h he e ec s o
he pandemic.
Two policy ques ions he e o e a ise om hese obse a ions:
(i)
Fi s ly, which MSMEs accessed concessional loans? Was he a ge ing o he loans
p og essi e o eg essi e (we e loans mo e likely o go o smalle , mo e ulne able
i ms, o o la ge i ms wi h mo e esou ces)? We add ess his ques ion in he
ollowing sub-sec ion.
(ii)
Secondly, ha e hese concessional loans ac ually helped en e p ises eco e om
he e ec s o he pandemic? We add ess his ques ion in Sec ion 5.1 wi h he main
esul s.
3.4 Ta ge ing o COVID-19 Concessional Loans
Table A1 demons a es ha he en e p ises ha accessed a concessional loan ended
o be la ge (bo h in u no e and numbe o employees), and we e mo e likely o ha e
p e iously accessed c edi han en e p ises ha did no ake ou a concessional loan. O
en e p ises ha ecei ed a loan, 53.5% we e mic oen e p ises, while 46.5% we e small
o medium-sized en e p ises. Meanwhile, 62.9% o en e p ises ha didn’ ecei e a loan
we e mic oen e p ises, while only 37.1% o en e p ises ha didn’ ecei e a loan we e
small o medium-sized en e p ises. Loan ecipien en e p ises ha e 4.8 employees on
a e age, compa ed o 4.2 o non- ecipien i ms. Loan ecipien i ms a e also mo e likely
o possess asse s, bo h eal p ope y and mo able colla e al.
This sugges s ha he a ge ing o he concessional loan p og am was somewha
eg essi e. I is impo an o no e ha we don’ ha e da a on loan applica ion beha io , so
we canno disen angle whe he eg essi i y was caused mo e by smalle i ms being less
likely o apply (“demand side”), o mo e by smalle i ms being mo e likely o be ejec ed a e
applying (“supply side”). In any case, somewha eg essi e a ge ing may be a e lec ion
o he eali y o adminis e ing such a p og am, in which loan assessmen needed o be
conduc ed on a ela i ely igh schedule du ing a ime o mobili y es ic ions ha would
ha e made he loan assessmen p ocess mo e di icul . In his case, i was gene ally no
possible o loan assesso s o ha e in-pe son mee ings wi h MSMEs, while a he same
ime he p og am ep esen ed he la ges inc ease in MSME lending in Fiji’s his o y (ADB
2023). I is pe haps no su p ising ha en e p ises ha could p o e c edi wo hiness and
11

likelihood o con inue in o he u u e, on pape , we e a o ed wi h access o suppo . The
social p og essi i y o he p og am is bols e ed somewha by he ac ha loan ecipien
en e p ises had mo e employees, and hence he concessional loan suppo may ha e also
bene i ed wo ke s, no jus en e p ise owne s.
These indings on he e ogenei y o loan a ge ing also sugges ha an analysis o he
impac o loans will need o conside his sys ema ic he e ogenei y in loan a ge ing. We
desc ibe ou app oach in Sec ion 4.1.
3.5 Impac s o Na u al Haza ds on MSMEs in Fiji
The baseline su ey also included a module asking MSMEs abou hei exposu e o na u al
haza ds and clima e isk, ha could lead o empo a y closu e o a signi ican educ ion
in e enue o up o 4 weeks. Table A1 shows ha jus o e hal o en e p ises ha e
expe ienced a signi ican na u al haza d in he pas , and a simila p opo ion belie e hey
ace a se ious isk o being exposed o such a na u al haza d in he u u e.
Figu e A3 p o ides en e p ises’ subjec i e pe cep ion o hei exposu e o signi ican
na u al haza ds, disagg ega ed by u no e , hough esponses end o be ela i ely simila
be ween mic oen e p ises, and small and medium en e p ises. We ind ha nea ly hal o
he en e p ises (48%) s a ed ha a cyclone could lead o empo a y closu e o a signi ican
educ ion in e enue o up o ou weeks. O he equen ly ci ed na u al haza d and
clima e- ela ed isks include looding (35.2%) and ea hquakes (4.1%). Only 1.8% o he
en e p ises pe cei ed sea le el ise as a subs an ial h ea o hei en e p ise, and an e en
smalle numbe pe cei ed d ough as a h ea .5
I any o hese isks had al eady impac ed hei en e p ise, esponden en e p ises we e
asked abou he suppo mechanisms hey elied on o cope wi h he a e ma h. Addi ionally,
en e p ises ha had no ye encoun e ed hese isks we e asked abou which suppo
mechanisms hey would ely on in he e en o u u e occu ences o hese isks. Figu e
A4 summa izes hei esponses. In e es ingly, o he 711 en e p ises ha had p e iously
su e ed a se ious clima e e en , mos (79.3%) used hei own sa ings o mi iga e he
impac o he e en . Only 11.3% accessed go e nmen suppo , while less han a 10 h
ecei ed help om he inancial se ice indus y ei he in he o m o a loan (7.7%) o an
insu ance payou (3.2%). In con as , a smalle p opo ion (62.2%) o he 648 MSMEs
ha had no ye su e ed a se ious clima e e en s a ed ha hey would need o ely on
5
I is wo h e-emphasizing ha he su ey was based on an en e p ise da abase ha excludes p ima y
a ming, ishing and o es y en e p ises. I may include en e p ises in ol ed in o he ag icul u al alue chain
unc ions– anspo , s o age, p ocessing, e c. Howe e , p ima y ag icul u al en e p ises may be e en mo e
exposed o many na u al haza ds, and hey a e no ep esen ed in ou s udy.
12
sa ings. On he o he hand, hey we e much mo e likely o belie e ha go e nmen suppo
(43.8%), o suppo om he inancial sec o h ough loans (31.8%) o insu ance (8.2%),
would assis hei eco e y. This sugges s ha en e p ises wi h less di ec expe ience o a
se ious clima e e en may o e es ima e he ex en o ou side suppo o aid hei eco e y.
This p o ides sugges i e e idence ha en e p ises ha ha e mo e pas expe ience wi h
na u al haza ds end o be mo e sel - elian . Howe e , i does no accoun o he possibili y
ha en e p ises ha display mo e sel - elian beha io s migh ha e selec ed in o sec o s,
geog aphies, o o he cha ac e is ics ha gi e hem mo e exposu e o na u al haza ds.
3.6 Alloca ion o Na u al Haza ds o Sampled MSMEs
In Table A2 we p o ide e idence on he co ela es o ha ing expe ienced a clima e shock,
in ou sample. This helps us weigh he plausibili y o asse ing ha clima e-shocks a e as-i
andom, o whe he en e p ises wi h ce ain cha ac e is ics migh ha e made pas choices
ha gi e hem mo e exposu e o na u al haza ds. I also p o ides insigh s on empi ical
s a egies ha migh allow us o mo e plausibly iden i y he impac s o clima e shocks on
en e p ises.
To do so, we employ a simple linea p obabili y model wi h a numbe o explana o y
a iables cap u ing he age o en e p ise, o al employees, owne ship o p ope y, c edi
access ou comes, gende o i m owne ship (male=1), a dummy o mic oen e p ise, age
o he en e p ise, and a numbe o sec o al dummies. We a y loca ion ixed e ec s by he
h ee main owns, and clus e ing o s anda d e o s by own. Ou p e e ed speci ica ion is
in column (3).
We ind only mild e idence o a sys ema ic ela ionship be ween hese cha ac e is ics
and clima e isk. In ou p e e ed speci ica ion, coe icien s a e o en economically small.
Only wo sec o dummies a e signi ican : ha o a s (wi h a posi i e 10.5% co ela ion wi h
likelihood o expe iencing a na u al haza d), and he eal es a e sec o (nega i e 21.2%
ela ionship wi h expe iencing a na u al haza d); we supp ess he o he , economically and
s a is ically insigni ican sec o dummies, o space. These a e wo o he mo e mino sec o s
in ou da a. Owne ship o eal p ope y and in angible p ope y show up as economically
and s a is ically signi ican , along wi h mic oen e p ise s a us. Hence om his we conclude
ha alloca ion o clima e haza d shocks is la gely idiosync a ic, hough in ou analysis o
he impac s o clima e haza d shocks we con ol o a a ge ed se o co a ia es, including
gende , age, age-squa ed, u no e and colla e al owne ship ( eal, mo able o in angible).
We desc ibe ou me hodology in mo e de ail in Sec ion 4.2.
13
4 METHODOLOGY
In his sec ion we desc ibe ou me hodology o iden i ying he causal impac s o COVID-19
concessional loans on pandemic eco e y, and he impac s o clima e e en s on MSMEs’
mi iga ion s a egies.
4.1 Impac s o COVID-19 Loans
As demons a ed in Sec ion 3.4 and Table A1, he e is he e ogenei y in access o COVID-19
concessional loans, as la ge i ms ( u no e , employees, asse s) we e mo e likely o ecei e
he loans. I is also plausible ha he e a e unobse able ac o s in loan applica ion o
loan app o al, based on he isk, p o ile, epu a ion, o o he aspec s o he en e p ises.
Un o una ely, we lack a clea sou ce o a ia ion in access o loans. None heless, i is
plausible ha he e was some a bi a iness in loan alloca ion, as applica ion illing and
e iew was conduc ed unde a ushed and p essu ed p ocess in he ea ly mon hs o he
pandemic. Hence i is plausible ha he e may be some o e lap in en e p ise cha ac e is ics
be ween MSMEs app o ed o loans, and hose ha did no ecei e loans (whe he due o
being denied access upon applying, o ailing o apply o quali y a all). Fo his analysis,
we limi ou sample o en e p ises ha we e ac i e p io o he onse o he pandemic in
2020.
To add ess he po en ial o endogenous selec ion in o loan access, we employ
p opensi y sco e ma ching (PSM), o impose balance on a ich se o obse ables be ween
loan ecipien s and non- ecipien s (DIME Analy ics, n.d.). The alidi y o PSM es s on he
assump ion ha , condi ional on hese obse able cha ac e is ics, ea ed uni s can be
di ec ly compa ed wi h un ea ed uni s. To do so, we ollow well-es ablished p ocedu es in
he li e a u e. We u ilize a logi eg ession o he ma ching equa ion, gi en he lack o
guidance on unc ional o m (Smi h 1997). In ou p e e ed speci ica ion, uni s a e hen
ma ched h ough nea es neighbo 1-1 ma ching, wi h a calipe o 0.01. “T ea men ” is
de ined as accessing a concessional COVID-19 loan.
The se o obse able cha ac e is ics o e which he p opensi y sco es a e calcula ed
includes he ollowing, which we e selec ed because hey we e ixed p io o he pandemic,
o i ele an o loan eligibili y:
•
En e p ise cha ac e is ics. Age o en e p ise, gende o owne ship, pas exposu e
o signi ican na u al haza ds;
•
C edi access. An indica o a iable showing whe he an en e p ise had an ac i e
loan when COVID-19 hi in Ma ch, 2020; and
14
•Fixed e ec s. Loca ion and sec o .
As shown in Figu e A5, which displays he common suppo his og ams, ma ching
only excludes a iny ac ion o en e p ises (<1%). This sugges s, pe haps su p isingly,
ha selec ion in o loan access may be mode a e. F om his igu e we can also see ha
he e is some he e ogenei y in he sample, as un ea ed en e p ises end o ha e lowe
p opensi y sco es, ela i e o ea ed en e p ises. Howe e , 1:1 ma ching ensu es ha
he e is balance be ween ea men and con ol ac oss he suppo o he p opensi y sco e
dis ibu ion co e ed by ea men uni s.
We a e limi ed in he se o ou comes we can conside , as he su ey was p ima ily
designed o se e as a baseline o an ongoing s udy. We also ha e limi ed access o
inancial da a on en e p ises, as he majo i y o en e p ises ailed o p o ide baseline
inancial da a. Hence we ocus on he causal e ec o accessing a concessional loan on
whe he en e p ises laid o wo ke s o sold asse s o add ess he impac s o he pandemic,
and en e p ises’ own subjec i e assessmen o hei eco e y om he pandemic. These a e
bina y ou comes, wi h he la e a iable aking on wo alues: eco e ed (ei he comple ely
o somewha ), and no eco e ed. I is mo e di icul o in e p e he a iables on laying o
wo ke s o selling asse s, as i ms may ha e al eady been ca ying ou hese ac ions when
conside ing and deciding o apply o concessional loans. I is possible ha he loans may
ha e mi iga ed an e en la ge employee o asse d awdown.
We hen un he ou come eg essions, based on an OLS linea p obabili y model, wi h
obus Abadie-Imbens s anda d e o s (Abadie and Imbens 2016).
yi=α+βTi+γxi+i
whe e ea men ,
Ti
is de ined as being equal o one i a MSME has ecei ed a COVID-19
concessional loan, and ze o i no .
We i s un his model on he unma ched sample (“OLS analysis”). We hen un he
same model on he p e e ed PSM ma ched sample (nea es neighbo 1:1 ma ching, wi h
a calipe o 0.01), which p oduces es ima es o he a e age ea men e ec on he ea ed
(ATT). We also epo on a obus ness check in he adjacen columns o he main PSM
esul s in Table A4, which adop s a 1:3 nea es neighbo ma ching mechanism, and widens
he calipe o 0.02.
4.2 Clima e Resilience
On he assump ion ha na u al haza ds la gely a ec MSMEs idiosync a ically in ou
sample, we quan i y he ex en o which pas exposu e o hese ypes o haza ds has a
15
causal impac on en e p ises’ pe cep ion o he mechanisms a ailable o assis eco e y
om u u e disas e s.
We ollow a simila app oach as in Sec ion 4.1. Fi s , we conduc p opensi y sco e
ma ching wi h 1:1 nea es neighbo ma ching wi h a calipe o 0.01, wi h ma ching a iables
including age o he en e p ise, gende o owne ship, an indica o a iable showing whe he
an en e p ise had an ac i e loan when COVID-19 hi , he amoun o ha loan, as well
as loca ion and sec o ixed e ec s. Figu e A6 shows ha only a small p opo ion o he
sample goes unma ched, hough he con ol g oup has la ge mass on lowe p opensi y
sco es, while he ea men g oup has la ge mass on highe p opensi y sco es.
Second, we un ou come eg essions in a simila s yle o he p e ious, hough we also
con ol o a ec o o co a ia es,
xi
, which includes u no e and colla e al owne ship ( eal,
mo able, o in angible), because o he selec ed imbalance shown in Table A2. Ou come
a iables cap u e whe he he MSMEs epo ha hey plan o ely on loans, go e nmen
suppo , hei own sa ings o local help o suppo hem i a na u al haza d we e o a ec
hem in he u u e (all bina y ou comes). S anda d e o s a e clus e ed a he loca ion le el.
Thi d, we addi ionally epo on a obus ness check wi h 1:3 nea es neighbo ma ching,
wi h a calipe o 0.02.
5 RESULTS
We p esen ou main esul s in his sec ion, i s o he impac o COVID-19 concessional
loans in enabling eco e y om he COVID-19 pandemic, and hen he ex en o which he
expe ience o pas na u al haza ds induces di e en expec a ions abou disas e eco e y
mechanisms by MSMEs.
5.1 Impac s o COVID-19 Loans
Ou main esul s on he impac s o COVID-19 concessional loans a e p esen ed in Tables
A3 and A4.
Table A3 e eals a posi i e co ela ion be ween concessional loans and he likelihood
o laying o wo ke s (s a is ically signi ican a 1%), selling o asse s (s a is ically signi ican
a 10%) and eco e y (no s a is ically signi ican ). This p o ides weak, bu sugges i e
e idence ha he loans wen o en e p ises ha we e mo e ha med by he COVID-19 shock,
and enabled eco e y.
Ou main esul s, which a emp o co ec o selec ion bias in o loan access h ough
p opensi y sco e ma ching, a e p esen ed in Table A4. Ou p e e ed esul s a e in columns
(1), (3), and (5), while we p o ide obus ness checks in columns (2), (4), and (6).
16

Fi s , ega ding d awdowns o wo ke s in esponse o he COVID-19 shock, we ind ha
loan ecipien i ms a e 3.3 pe cen age poin s mo e likely o ha e laid o wo ke s, howe e
his is no s a is ically signi ican . Though ou obus ness check has a la ge magni ude
(5.4%) and is s a is ically signi ican . Gi en ha abou 22% o non-loan ecipien i ms laid
o wo ke s, his sugges s ha loan ecipien i ms we e a leas one-se en h mo e likely o
ha e laid o wo ke s.
Second, ega ding d awdowns o asse s in esponse o he COVID-19 shock, we ind
ha loan eceipien i ms a e 4.0 pe cen age poin s mo e likely o ha e educed asse s in
esponse o he pandemic, howe e his is no s a is ically signi ican . Gi en ha abou
7.1% o non-loan ecipien i m laid o wo ke s, his sugges s ha loan ecipien i ms we e
abou wo- hi ds mo e likely o ha e shed asse s. The obus ness check co obo a es his
main esul , wi h a sligh ly smalle poin es ima e.
Wi h ega ds o d awdowns o wo ke s and asse s, i is possible ha he adminis a ion o
he concessional loan p og am, in spi e o he bes a emp s, was no swi enough o mi iga e
he ea ly impac s o he pandemic. E en a e con olling o sec o and geog aphy, i is
possible ha he mos -impac ed i ms we e mo e likely o apply and quali y o concessional
loans.
We also ind in column (5) ha MSMEs ha accessed a concessional loan a e 3.4
pe cen age poin s mo e likely o say ha hey ha e eco e ed om he impac s o he
pandemic. Howe e , his e ec is no s a is ically signi ican . This e ec is con i med by ou
obus ness check, hough wi h a sligh ly smalle magni ude. As his esul is less likely o be
unde mined as ou wo ke and asse esul s can by unce ain y abou iming o selec i i y
in loan eceip , i is mo e sugges i e ha he p og am only had mild bene i s o ecipien s
o mi iga e he e ec s o he pandemic. This may be ela ed o he eg essi e a ge ing o
he loans ha we documen ed in Sec ion 3.4.
5.2 Clima e Resilience and Mi iga ion Mechanisms
Ou main esul s on he impac s o pas expe ience o na u al haza ds on expec a ion o
u u e mi iga ion mechanisms a e in Tables A5 and A6.
Table A5 e eals a nega i e ela ionship be ween pas expe ience o na u al haza d
e en s and u u e expec a ions o ecei ing ou side help om hi d pa ies o eco e
( h ough loans (-0.22 p.p.) o go e nmen suppo (-0.3 p.p.)). Gi en ha abou 50% o
en e p ises ha e pas expe ience wi h signi ican na u al haza ds, he magni ude o e ec
ep esen s oughly a hal ing in he expec a ion o ecei ing loans o go e nmen suppo .
Meanwhile, he change in expec a ion o ecei ing local help, while posi i e, is ela i ely
17
small (0.03 p.p.) and insigni ican . Meanwhile, i ms a e 14 pe cen age poin s mo e likely
o be sel - elian ( elying on own-sa ings), signi ican a 5%.
Ou main esul s, which a emp o co ec o selec ion bias in o loan access h ough
p opensi y sco e ma ching, a e p esen ed in Table A6. Ou p e e ed esul s a e in columns
(1), (3), (5), and (7), while we p o ide obus ness checks in columns (2), (4), (6), and (8).
Table A6 shows ha en e p ises ha ha e expe ienced a na u al haza d we e 25
pe cen age poin s less likely o say hey would ely on loans i ano he haza d we e o a ise
in he u u e, and 27 pe cen age poin s less likely o say hey would u n o go e nmen
suppo , compa ed o en e p ises ha had no expe ienced a p e ious se ious disas e .
These esul s a e highly signi ican (1%), consis en wi h he obus ness checks, and align
wi h he magni udes o ou co ela ional analysis in A5. Hence, we ha e s ong and
consis en indings ha pas expe ience wi h na u al haza ds makes i ms abou hal as
likely o say hey would ely on help om hi d-pa ies like banks o he go e nmen o
assis ance wi h eco e y in he u u e.
Con e sely, MSMEs wi h pas expe ience o na u al haza ds we e 15 pe cen age poin s
mo e likely o ely on own sa ings in he u u e, abou a 28% inc ease in likelihood. These
esul s a e highly signi ican , consis en wi h he obus ness check, and align wi h he
magni udes o ou co ela ional analysis in A5. We also ind ha he likelihood o elying
on local help 3 pe cen age poin s in ou p e e ed speci ica ion, hough insigni ican in
ou p e e ed speci ica ion. Howe e , i is no able ha he es ima e is la ge (5 p.p.) and
signi ican a 5% in he obus ness check.
These esul s a e sugges i e ha MSMEs expe iencing clima e- ela ed disas e s ha e
ecei ed less suppo om he o mal inancial sec o o he go e nmen han hey migh
ha e o he wise expec ed, which has led hem o be mo e elian on hei own esou ces in
he u u e. This disappoin men could be a ibu ed o a numbe o mechanisms, including
delays in ecei ing suppo , eceip o less suppo han expec ed, o simply no ecei ing
suppo when i was expec ed.
6 DISCUSSION
6.1 Impac s o COVID-19 Loans
The COVID-19 pandemic p ecipi a ed unp eceden ed economic challenges wo ldwide,
p omp ing a ee alua ion o inancial suppo mechanisms o businesses, especially in
eme ging ma ke s. These inancing p og ams aimed o p o ide liquidi y o businesses acing
ope a ional and inancial dis up ions, ensu ing hei su i al and sa egua ding employmen .
18
Concessional loans, cha ac e ized by a o able e ms such as lowe in e es a es and
longe g ace pe iods, eme ged as a key ool in he policy a senal. S udies such as Ci e a
e al. (2021) p o ide a comp ehensi e e iew o policies ac oss se e al eme ging ma ke s.
The au ho s ound ha while concessional loans we e c ucial in p o iding immedia e
elie , hei e ec i eness a ied signi ican ly ac oss di e en economies, depending on
he p og am’s design and implemen a ion e iciency. Mo eo e , s udies ha e poin ed ou
he need o complemen ing inancial suppo wi h measu es such as ax elie and wage
subsidies.
F om hese policies, se e al lessons we e lea ned:
(i)
The e ec i eness o concessional loans hinges on ca e ul design, including eligibili y
c i e ia, o ensu e a ge ed suppo .
(ii)
Implemen a ion e iciency, acili a ed by digi al pla o ms and s eamlined p ocesses,
can enhance he impac o inancial suppo p og ams.
(iii)
Complemen a y measu es, such as ax de e als and wage subsidies, a e c ucial in
p o iding comp ehensi e suppo o businesses.
These indings emphasize he need o agile, well- a ge ed, and mul i- ace ed policy
esponses o suppo he business ecosys em in eme ging ma ke s du ing c ises.
In Fiji, we obse e ha a numbe o hese measu es we e in ac used. Howe e ,
implemen a ion was challenging due o he eali ies o he pandemic, and we ind ha he
p oblem had ela i ely limi ed measu able impac s on helping MSMEs eco e om he
pandemic. Some inancial ins i u ions we e o e whelmed wi h loan applica ions, which
may ha e led o mo e eg essi e a ge ing as implemen e s aded o he speed o loan
decision making wi h he ex ensi eness o he loan applica ion e iews. While he p og am
was success ul in app o ing a eco d-b eaking olume o MSME loans du ing an ex emely
challenging and unp eceden ed ime, he e may be oom o e lec on he lessons lea ned
om he p og am o conside ing wha could be possible in he u u e. Fo example, i
may be wo h conside ing e-calib a ing he balance be ween speed and ex ensi eness
o e iew o applica ions (o e en au oma ic p e-app o al based on ce ain c i e ia) and
he unding amoun s a ailable. I may be ha wha some MSMEs needed was as e
app o al and disbu semen o inancing, while o he s needed la ge amoun s, o ha e a
mo e measu able impac on hei mi iga ion and eco e y om pandemic impac s. These
de ailed design ques ions would bene i om u he esea ch, a ge ed o he design o
pandemic esponse concessional inancing.
19
6.2 Clima e Resilience
To mi iga e he e ec s o clima e change- ela ed na u al haza ds and disas e s, i is
impo an o implemen e ec i e policy measu es, and communica e clea ly abou hem.
We ind ha in Fiji, many en e p ises may o e es ima e he suppo measu es a ailable
a e a disas e . En e p ises ha had p e iously expe ienced a disas e a e less likely o
expec o u n o loans o go e nmen suppo i a disas e we e o occu again in he u u e,
compa ed o en e p ises ha hadn’ p e iously expe ienced such a disas e . Con e sely,
hey a e mo e likely o ely on hei own sa ings and local help.
Mo e e ec i e communica ion abou esponses o disas e s could help o educe
con usion among MSMEs, and educe unce ain y and isk o ca as ophic losses. I
go e nmen and p i a e sec o esou ces a e no a ailable in he magni ude needed o ully
mi iga e he e ec s o clima e- ela ed disas e s, hen i may be bes o encou age MSMEs
o ake s eps o indi idually mi iga e clima e- ela ed isks, such as h ough own-sa ing and
pu chasing p i a e insu ance.
7 CONCLUSION
This pape seeks o explo e he esilience o MSMEs in Fiji o majo shocks including
he COVID-19 pandemic and clima e change- ela ed disas e s. I d aws on an o iginal
su ey o MSMEs conduc ed in he i s hal o 2023. We ind e idence o se e e impac s
o COVID-19 on MSMEs. The go e nmen esponded wi h a concessional loan p og am,
which we ind had some eg essi e a ge ing, howe e his is plausible in ligh o he
unp eceden ed and challenging ea ly pandemic pe iod. We ind ela i ely mild impac s
o concessional loans in mi iga ing he pandemic’s economic impac s on MSMEs. The
s udy e eals ha p e ious expe ience wi h na u al haza ds signi ican ly shapes MSMEs’
s a egies o u u e esilience, wi h a p onounced inc ease in eliance on in e nal esou ces
o e ex e nal suppo mechanisms.
While exis ing policy measu es ha e p o ided a ounda ion o economic eco e y
and esilience-building among MSMEs, he e emains subs an ial oom o imp o emen .
While he pandemic undoub edly p o ided an unp eceden ed challenge in implemen ing
a concessional loan p og am, a mo e e icien assessmen p ocess may ha e be e
a ge ed he loans o MSMEs whe e he unds could ha e he highes ma ginal impac s. In
he domain o na u al haza ds, enhancing awa eness and access o clima e adap a ion
esou ces could signi ican ly bols e MSMEs long- e m sus ainabili y and esilience, as
hey would be be e able o ealis ically plan and p epa e o majo clima e e en s.
20
Figu e A6: Clima e Analysis: Common Suppo His og am
Sou ce: Au ho s’ calcula ions.
27

Table A1: Cha ac e is ics o MSMEs by Concessional Loan Access
(1) (2) (3) (4)
No Loan Loan To al N
Tu no e
<F$50,000 0.629 0.535 0.609 2,380
F$50,000–F$3,000,000 0.371 0.465 0.391 2,380
Age o en e p ise 12.44 11.79 12.30 2341
To al employees 4.236 4.761 4.345 2,372
Clima e Impac s
Fu u e exposu e o na u al haza d 0.561 0.583 0.566 2,400
Pas exposu e o na u al haza d 0.521 0.532 0.523 1,359
Owne ship gende
Male 0.576 0.540 0.568 2,370
Female 0.297 0.270 0.291 2,370
Bo h 0.128 0.190 0.141 2,370
C edi access
Accessed 0.220 0.380 0.254 2,288
Denied 0.0617 0.0551 0.0603 2,288
Ne e applied 0.718 0.565 0.685 2,288
Owne ship o eal p ope y 0.361 0.431 0.376 2,399
Owne ship o mo able p ope y 0.671 0.762 0.690 2,399
Owne ship o in angible p ope y 0.0327 0.0317 0.0325 2,399
Obse a ions 1,896 504 2,400
No e: The i s column p esen s a e ages o cha ac e is ics o mic o, small, and medium-sized en e p ises
(MSMEs) ha did no access a concessional loan (n = 1896); he second column p esen s a e ages o
MSMEs ha did access a concessional loan (n = 504); he hi d column p esen s a e ages ac oss he en i e
sample, d opping en e p ises ha did no epo u no e o c edi access; he ou h column p esen s he
numbe o non-missing obse a ions o each a iable. Wi h he excep ion o age and o al employees, all
a iables a e bina y. Hence, a e ages can be unde s ood as pe cen ages.
Sou ce: Au ho s’ calcula ions.
28
Table A2:
Is he En e p ise Exposed o a Na u al Haza d ha Could Shu he Business o
4 Weeks?
(1) (2) (3)
Age o en e p ise -0.000691 -0.000691 -0.000232
(0.00174) (0.00142) (0.000964)
Age squa ed 0.00000817 0.00000817 0.000000710
(0.0000250) (0.00000926) (0.00000787)
To al employees -0.000123 -0.000123 0.000823
(0.00108) (0.000742) (0.000632)
Owne ship o mo able p ope y 0.0366 0.0366 0.0570
(0.0242) (0.0539) (0.0387)
Accessed c edi (las 5 y s) 0.0443∗0.0443 0.0321
(0.0259) (0.0283) (0.0186)
Denied c edi (las 5 y s) -0.0326 -0.0326 -0.0415
(0.0439) (0.0359) (0.0368)
Male owne ship -0.0215 -0.0215 -0.0260
(0.0214) (0.0229) (0.0185)
Owne ship o eal p ope y 0.126∗∗∗ 0.126∗∗∗ 0.0855∗∗
(0.0234) (0.0360) (0.0354)
Owne ship o in angible p ope y 0.205∗∗∗ 0.205∗∗ 0.158∗
(0.0594) (0.0749) (0.0834)
Mic oen e p ise (2023 u no e ) 0.0657∗∗∗ 0.0657∗∗∗ 0.0355∗
(0.0236) (0.0168) (0.0178)
A s sec o 0.106 0.106∗∗ 0.105∗
(0.0725) (0.0400) (0.0465)
Real sec o -0.242∗∗ -0.242∗∗∗ -0.212∗∗
(0.117) (0.0666) (0.0673)
Loca ion ixed e ec s No No Yes
Clus e ed SEs No Yes Yes
R-squa ed 0.0344 0.0344 0.0923
Obse a ions 2,195 2,195 2,195
No e: Each column ep esen s a di e en speci ica ion o a join eg ession o mic o, small, and medium-sized
en e p ise (MSME) exposu e o na u al haza ds o clima e isks on a ange o co a ia es. Column (1) epo s
s anda d O dina y Leas Squa es (OLS) esul s, while column (2) adds clus e ed s anda d e o s (a he own
le el), and column (3) includes loca ion ixed e ec s. Age e e s o he numbe o yea s since he en e p ise
was es ablished, while bo h colla e al owne ship and c edi access co a ia es a e bina y. The able omi s
he coe icien on sec o ixed e ec s o sec o s ha a e no s a is ically signi ican unde any o hese h ee
speci ica ions. S anda d e o s a e in pa en heses. ∗p < 0.10,∗∗ p < 0.05,∗∗∗ p < 0.01.
Sou ce: Au ho s’ calcula ions.
29
Table A3:
Did Concessional Loans Aid Pos -pandemic Reco e y? (O dina y Leas Squa es
Analysis)
(1) (2) (3)
We e wo ke s laid o ? We e asse s sold o ? Ex en o eco e y?
Concessional Loan 0.037∗∗∗ 0.045∗0.022
(0.012) (0.027) (0.024)
En e p ises 1,710 1,710 1,658
No e: Es ima es a e in pe cen age poin e ms. Analysis is es ic ed o en e p ises ha exis ed p io o he
onse o he co ona i us disease (COVID-19) pandemic in ea ly 2020. O hese en e p ises, 52 ei he did no
answe o e used o commen on whe he he i m had eco e ed om he impac o he pandemic. We
es ima e he e ec o concessional loans on h ee ele an ou comes: (1) whe he a i m laid o wo ke s
in esponse o he impac s o he COVID-19 pandemic, (2) whe he a i m sold asse s, and (3) whe he
mic o, small, and medium-sized en e p ises (MSMEs) hink hei business has “somewha ” o “comple ely”
eco e ed om he impac s o he pandemic. ∗p < 0.10,∗∗ p < 0.05,∗∗∗ p < 0.01.
Sou ce: Au ho s’ calcula ions.
Table A4:
Did Concessional Loans Aid Pos -pandemic Reco e y? (P opensi y Sco e
Ma ching Analysis)
Wo ke s? Asse s? Reco e ed?
(1) (2) (3) (4) (5) (6)
PSM Robus PSM Robus PSM Robus
Concessional Loan 0.033 0.054∗0.040 0.033 0.034 0.030
(0.038) (0.031) (0.025) (0.022) (0.037) (0.030)
En e p ises 1,696 1,703 1,696 1,703 1,640 1,645
No e: Es ima es a e in pe cen age poin e ms. We use p opensi y sco e ma ching (PSM) o es ima e he
e ec o concessional loans on h ee ele an ou comes: (1) whe he a i m laid o wo ke s in esponse o
he impac s o he COVID-19 pandemic, (2) whe he a i m sold asse s, and (3) whe he mic o, small, and
medium-sized en e p ises (MSMEs) hink hei business has “somewha ” o “comple ely” eco e ed om
he impac s o he co ona i us disease (COVID-19) pandemic. P opensi y sco es a e calcula ed using he
ollowing a iables: age, gende o owne ship, business loca ion, whe he he i m had an ac i e loan when
he COVID-19 pandemic hi , whe he he i m was exposed o na u al haza ds, as well as loca ion and sec o
ixed e ec s. En e p ises we e hen ma ched using nea es neighbo ma ching, es ic ed o common suppo .
Robus Abadie-Imbens s anda d e o s a e in pa en heses. ∗p < 0.10,∗∗ p < 0.05,∗∗∗ p < 0.01.
Sou ce: Au ho s’ calcula ions.
30
Table A5:
Does Pas Expe ience wi h Clima e Shocks A ec En e p ises’ Fu u e
Responses? (O dina y Leas Squa es Analysis)
(1) (2) (3) (4)
Loans Go Suppo Sa ings Local Help
Has ha e en occu ed? -0.22∗∗∗ -0.30∗∗∗ 0.14∗∗ 0.03
(0.05) (0.06) (0.05) (0.02)
En e p ises 1,289 1,289 1,289 1,289
Mean o ou come 0.20 0.28 0.71 0.11
No e: Each column epo s he eg ession esul s o a sepa a e ou come. All ou lis ed ou comes a e bina y,
meaning ha eg ession coe icien s can be in e p e ed as pe cen age poin changes. Whe he an en e p ise
had encoun e ed a na u al haza d in he pas is he dependen a iable ac oss all ou eg essions. Analysis
is es ic ed o mic o, small, and medium-sized en e p ises (MSMEs) who ace he isk o a se ious na u al
haza d. Con ol a iables include loca ion, sec o , gende , age, and age-squa ed. Robus Abadie-Imbens
s anda d e o s a e in pa en heses. ∗p < 0.10,∗∗ p < 0.05,∗∗∗ p < 0.01.
Sou ce: Au ho s’ calcula ions.
Table A6:
Does Pas Expe ience wi h Clima e Shocks A ec En e p ises’ Fu u e
Responses? (P opensi y Sco e Ma ching Analysis)
Loans Go Suppo Sa ings Local Help
(1) (2) (3) (4) (5) (6) (7) (8)
Has ha e en -0.25∗∗∗ -0.21∗∗∗ -0.27∗∗∗ -0.25∗∗∗ 0.15∗∗∗ 0.11∗∗∗ 0.03 0.05∗∗
occu ed? (0.04) (0.03) (0.04) (0.03) (0.04) (0.03) (0.03) (0.02)
En e p ises 1,210 1,239 1,210 1,239 1,210 1,239 1,210 1,239
Mean o ou come 0.19 0.19 0.27 0.27 0.72 0.72 0.11 0.11
No e: We use p opensi y sco e ma ching (PSM) o es ima e he e ec o pas exposu e o na u al haza ds
on likelihood o ou dis inc coping mechanisms: (1) loans, (2) go e nmen suppo , (3) sa ings and (4) local
help. P opensi y sco es a e calcula ed h ough a p obi eg ession using he ollowing a iables: age, gende
o owne ship, business loca ion, whe he he i m had an ac i e loan when he COVID-19 pandemic hi , he
amoun o ha loan, whe he he i m was exposed o na u al haza ds, as well as loca ion and sec o ixed
e ec s. En e p ises we e hen ma ched using 1-3 nea es neighbo ma ching, es ic ed o common suppo .
Robus Abadie-Imbens s anda d e o s a e in pa en heses. ∗p < 0.10,∗∗ p < 0.05,∗∗∗ p < 0.01.
Sou ce: Au ho s’ calcula ions.
31
A.2 Sec o s
All de ini ions a e ob ained om he Fiji Bu eau o S a is ics (2010) mos ecen S anda d
Indus ial Classi ica ions.
Accommoda ion and Food Se ices. These include p o ision o sho -s ay
accommoda ion and he p o ision o comple e meals and d inks i o immedia e
consump ion.
Adminis a i e and Suppo Se ice Ac i i ies. These include a a ie y o ac i i ies
ha suppo gene al business ope a ions. These ac i i ies do no equi e he ans e o
specialized knowledge.
A s, En e ainmen and Rec ea ion. These include a wide ange o ac i i ies o mee
a ied cul u al, en e ainmen and ec ea ional in e es s o he gene al public, including li e
pe o mances, ope a ion o museum si es, gambling, spo s, and ec ea ion ac i i ies.
Cons uc ion. These include gene al cons uc ion and specialized cons uc ion
ac i i ies o buildings and ci il enginee ing wo ks. Also includes he epai o buildings
and enginee ing wo ks.
Financial and Insu ance Ac i i ies. These include inancial se ice ac i i ies, including
insu ance, einsu ance and pension unding ac i i ies and ac i i ies o suppo inancial
se ices. I also includes he ac i i ies o holding asse s.
Human Heal h and Social Wo k Ac i i ies. These include he p o ision o heal h
and social wo k ac i i ies. Ac i i ies include a wide ange o ac i i ies, om heal h ca e
p o ided by ained medical p o essionals in hospi als and o he acili ies, o e esiden ial
ca e ac i i ies ha s ill in ol e a deg ee o heal h ca e ac i i ies, o social wo k ac i i ies
wi hou any in ol emen o heal h ca e p o essionals.
In o ma ion and Communica ion. These include he p oduc ion and dis ibu ion o
in o ma ion and cul u al p oduc s, he p o ision o means o ansmi o dis ibu e hese
p oduc s, communica ions, in o ma ion echnology ac i i ies, and he p ocessing o da a
and o he in o ma ion se ice ac i i ies.
Manu ac u ing. These include he physical o chemical ans o ma ion o ma e ials,
subs ances, o componen s in o new p oduc s. Subs an ial al e a ion, eno a ion o
econs uc ion is gene ally conside ed o be manu ac u ing.
O he Se ice Ac i i ies. These include he ac i i ies o membe ship o ganiza ions,
he epai o compu e s and pe sonal and household goods and a a ie y o pe sonal
se ice ac i i ies no co e ed elsewhe e in he classi ica ion.
32

P o essional, Scien i ic and Technical Ac i i ies. These include specialized
p o essional, scien i ic and echnical ac i i ies. These ac i i ies equi e a high deg ee o
aining, and make specialized knowledge and skills a ailable o use s.
Real Es a e Ac i i ies. These include ac ing as lesso s, agen s and/o b oke s in one
o mo e o he ollowing: selling o buying eal es a e, en ing eal es a e, p o iding o he
eal es a e se ices such as app aising eal es a e, o ac ing as eal es a e esc ow agen s.
T anspo and S o age. These include he p o ision o passenge o eigh anspo
by ail, pipeline, oad, wa e o ai , as well as associa ed ac i i ies, such as e minal and
pa king acili ies, ca go handling, s o age e c.
Wholesale and Re ail T ade; Repai o Mo o Vehicles, Mo o cycles’ (WRT). These
include wholesale and e ail sale (i.e., sale wi hou ans o ma ion o goods), and he
ende ing o se ices inciden al o he sale o hese goods. Goods bough and sold a e
also e e ed o as me chandise and wholesaling and e ailing a e he inal s eps in he
dis ibu ion o hese goods. Also included in his sec ion a e he epai o mo o ehicles
and mo o cycles.
33
REFERENCES
Abadie, Albe o, and Guido W. Imbens. 2016. “Ma ching on he Es ima ed P opensi y
Sco e.” Econome ica 84 (2): 781–807.
ADB. 2019. The Economics o Clima e Change in he Paci ic. Asian De elopmen Bank.
. 2023. Asia Small and Medium-Sized En e p ise Moni o 2023: How Small Fi ms
Can Con ibu e o Resilien G ow h in he Paci ic Pos COVID-19 Pandemic. Asian
De elopmen Bank.
Baldwin, Richa d, and Bea ice Wede di Mau o, eds. 2020. Economics in he Time o
COVID-19. CEPR P ess.
Ce da, Maikol, Paul Ge le , Sean Higgins, Ana Ma ía Mon oya, E ic Pa ado, and
Raimundo Undu aga. 2023. “The Causal Impac o COVID-19 Go e nmen -backed
Loans on MSMEs Liquidi y and Ea nings.” IDB Wo king Pape Se ies No.
IDB-WP-1430.
Ci e a, Xa ie , Ma cio C uz, Elwyn Da ies, A i G o e , Leona do Iaco one, Jose E nes o
Lopez Co do a, Denis Med ede , e al. 2021. “Policies o Suppo Businesses h ough
he COVID-19 Shock: A Fi m Le el Pe spec i e.” The Wo ld Bank Resea ch Obse e
36 (1): 41–66.
DIME Analy ics. n.d. “P opensi y Sco e Ma ching.” (accessed: 9.15.2023). h ps://dimewiki.
wo ldbank.o g/P opensi y_Sco e_Ma ching.
Gou inchas, Pie e-Oli ie , Thomas Philippon, and Dimi i Vayanos. 2021. “Pandemic
Relie : Public Policy in Times o C isis.” Jou nal o Economic Pe spec i es 35 (1):
199–224.
Hallega e, S ephane, and Julie Rozenbe g. 2017. “Clima e Change h ough a Po e y
Lens.” Na u e Clima e Change 7 (4): 250–256.
Kucke z, And eas, Lisa B ändle, Anja Gaudig, Sa ah Hinde e , Ca los A u o Mo eno
Reyes, Anas asia P ocho a, Ka ha ina M S einb ink, and Elisabe h S.C. Be ge . 2020.
“S a ups in Times o C isis–-A Rapid Response o he COVID-19 Pandemic.” Jou nal
o Business Ven u ing Insigh s e00169.
34
Kuma , Sunil, and Rup Singh. 2020. “Economic Impac s o he COVID-19 Pandemic on
Mic o, Small, and Medium En e p ises (MSMEs) in Fiji.” Jou nal o Economic Impac
2 (3): 125–140.
Noy, Ilan, and Rio Yonson. 2018. “Economic Reco e y A e Na u al Disas e s: The Role
o Insu ance and O he Financial Resou ces.” Re iew o De elopmen Economics 22
(3): e13097.
Sha i, Mohsin, Jun ong Liu, and Wei Ren. 2020. “Impac o COVID-19 Pandemic on
Mic o, Small, and Medium-sized En e p ises Ope a ing in Pakis an.” Resea ch in
Globaliza ion 2:100018.
Smi h, He be L. 1997. “Ma ching Wi h Mul iple Con ols o Es ima e T ea men E ec s in
Obse a ional S udies.” Sociological Me hodology 27 (1): 325–353.
35
ASIAN DEVELOPMENT BANK
ASIAN DEVELOPMENT BANK
6 ADB A enue, Mandaluyong Ci y
1550 Me o Manila, Philippines
www.adb.o g
RESILIENCE TO SHOCKS
OF MICRO, SMALL,
AND MEDIUM-SIZED
ENTERPRISES IN FIJI
William G eenland and Russell To h
ADB ECONOMICS
WORKING PAPER SERIES
NO. 765
Janua y 2025
Resilience o Shocks o Mic o, Small, and Medium-Sized En e p ises in Fiji
This pape examines he esilience o mic o, small, and medium-sized en e p ises (MSMEs) in Fiji o
economic shocks, ocusing on he COVID-19 pandemic and clima e- ela ed haza ds. Based on o iginal
su ey da a om 2,400 MSMEs, he pape e alua es he impac o go e nmen concessional loans on
pandemic eco e y and explo es clima e esilience s a egies. Findings e eal mixed ou comes om policy
in e en ions and eliance on sel - unding by en e p ises impac ed by disas e s. The pape emphasizes
in eg a ing inancial suppo wi h long- e m clima e adap a ion o s eng hen MSME esilience.
Abou he Asian De elopmen Bank
ADB is commi ed o achie ing a p ospe ous, inclusi e, esilien , and sus ainable Asia and he Paci ic,
while sus aining i s e o s o e adica e ex eme po e y. Es ablished in 1966, i is owned by 69 membe s
—49 om he egion. I s main ins umen s o helping i s de eloping membe coun ies a e policy dialogue,
loans, equi y in es men s, gua an ees, g an s, and echnical assis ance.