Effects of COVID-19 on catastrophic health expenditures and inequality in Benin: A microsimulation approach

Honlonkou, Albe N.; Bassongui, Nassibou; Da a é, Co inne B.
A icle
E ec s o COVID-19 on ca as ophic heal h expendi u es
and inequali y in Benin: A mic osimula ion app oach
Economies
P o ided in Coope a ion wi h:
MDPI – Mul idisciplina y Digi al Publishing Ins i u e, Basel
Sugges ed Ci a ion: Honlonkou, Albe N.; Bassongui, Nassibou; Da a é, Co inne B. (2025) : E ec s o
COVID-19 on ca as ophic heal h expendi u es and inequali y in Benin: A mic osimula ion app oach,
Economies, ISSN 2227-7099, MDPI, Basel, Vol. 13, Iss. 8, pp. 1-27,
h ps://doi.o g/10.3390/economies13080222
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/329502
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/4.0/
Academic Edi o : Fabio Clemen i
Recei ed: 22 Feb ua y 2025
Re ised: 6 July 2025
Accep ed: 9 July 2025
Published: 29 July 2025
Ci a ion: Honlonkou, A. N.,
Bassongui, N., & Da a é, C. B. (2025).
E ec s o COVID-19 on Ca as ophic
Heal h Expendi u es and Inequali y in
Benin: A Mic osimula ion App oach.
Economies,13(8), 222. h ps://doi.o g/
10.3390/economies13080222
Copy igh : © 2025 by he au ho s.
Licensee MDPI, Basel, Swi ze land.
This a icle is an open access a icle
dis ibu ed unde he e ms and
condi ions o he C ea i e Commons
A ibu ion (CC BY) license
(h ps://c ea i ecommons.o g/
licenses/by/4.0/).
A icle
E ec s o COVID-19 on Ca as ophic Heal h Expendi u es and
Inequali y in Benin: A Mic osimula ion App oach
Albe N. Honlonkou *, Nassibou Bassongui and Co inne B. Da a é
Na ional School o Applied Economics and Managemen (ENEAM), Uni e si y o Abomey-Cala i,
Co onou BP 358, Benin; [email p o ec ed] (N.B.); [email p o ec ed] (C.B.D.)
*Co espondence: [email p o ec ed]
Abs ac
This s udy assesses he e ec s o he COVID-19 pandemic on ca as ophic heal h expendi-
u es and income inequali y in Benin. A mic osimula ion was calib a ed o es ima e he
impac o he pandemic unde h ee di e en shock scena ios: low, mode a e, and se e e.
The analysis elies on seconda y da a om household li ing condi ion su eys. The esul s
indica e ha he COVID-19 c isis would lead o a signi ican a e age income loss o up
o 20% and income inequali y, while he numbe o households wi h ca as ophic heal h
expendi u es would inc ease by 4%. Mo e impo an ly, he indings e eal he e ogeneous
impac s ac oss households, wi h u ban esiden s, younge indi iduals, mo e educa ed
households, and male-headed households expe iencing he g ea es income decline. These
indings unde sco e he need o a ge ed heal h co e age and employmen policies o
be e p o ec ulne able popula ions in Benin in he ace o u u e shocks.
Keywo ds: heal h expendi u es; inequali ies; mic osimula ion; COVID-19
JEL Classi ica ion: C53; D31; I18; I31
1. In oduc ion
In 2019, he wo ld expe ienced an unp eceden ed heal h shock, ha o he COVID-19
pandemic. This deadly in ec ious disease caused by se e e acu e espi a o y diso de s o
co ona i us synd ome was decla ed a global pandemic by he Wo ld Heal h O ganisa ion
(WHO) on 11 Ma ch 2020 (WHO,2020). The magni ude o he pandemic was es ima ed o
be nea ly hal a billion people in ec ed by 11 Ma ch 2022, and mo e han 6.5 million dea hs
wo ldwide, and 12 million people in ec ed and 254 housand dea hs in A ica (WHO,
2022a). Thus, he numbe o people in ec ed by he COVID-19 pandemic alone ep esen s
mo e han 71 imes he numbe o people in ec ed by he i e la ges epidemics he wo ld
has expe ienced in he las wo decades, namely Zika in 2016, Ebola i us in 2014, Middle
Eas Respi a o y Synd ome Co ona i us (MERS-CoV) in 2012, H1N1 i us in 2009, and
Se e e Acu e Respi a o y Synd ome (SARS) in 2003 (Loungani e al.,2021).
Benin eco ded he i s con i med case o COVID-19 on 16 Ma ch 2020, and he second
case on 17 Ma ch 2020. The i s dea h was eco ded on 6 Ap il 2020. These i s e en s
ma ked he beginning o he COVID-19 c isis in Benin. Figu es 1and 2highligh ha he
c isis was cha ac e ised by wo wa es. The i s wa e indica es ha he pandemic g ew
apidly du ing he pe iod om May o Sep embe 2020 be o e slowing down. The second
wa e spanned he pe iod om June o Augus 2021, whe e bo h he numbe o new cases
and he numbe o dea hs inc eased mo e han in he i s wa e.
Economies 2025,13, 222 h ps://doi.o g/10.3390/economies13080222
Economies 2025,13, 222 2 o 27

0
10
20
30
40
50
60
70
80
90
2020/3/16
2020/4/16
2020/5/16
2020/6/16
2020/7/16
2020/8/16
2020/9/16
2020/10/16
2020/11/16
2020/12/16
2021/1/16
2021/2/16
2021/3/16
2021/4/16
2021/5/16
2021/6/16
2021/7/16
2021/8/16
2021/9/16
2021/10/16
2021/11/16
2021/12/16
2022/1/16
2022/2/16
2022/3/16
Figu e 1. New con i med COVID-19 cases pe million in Benin. Sou ce: Au ho s, using Johns
Hopkins Uni e si y CSSE COVID-19 da a, 2022.

0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
2020/3/16
2020/4/16
2020/5/16
2020/6/16
2020/7/16
2020/8/16
2020/9/16
2020/10/16
2020/11/16
2020/12/16
2021/1/16
2021/2/16
2021/3/16
2021/4/16
2021/5/16
2021/6/16
2021/7/16
2021/8/16
2021/9/16
2021/10/16
2021/11/16
2021/12/16
2022/1/16
2022/2/16
2022/3/16
Figu e 2. Dea hs pe million o he popula ion in Benin. Sou ce: Au ho s, using Johns Hopkins
Uni e si y CSSE COVID-19 da a, 2022.
Following con i ma ion o he i s cases o COVID-19, he Beninese au ho i ies ook
measu es including he closu e o land bo de s, sys ema ic qua an ine o all a ele s
en e ing by ai , school closu es, p ohibi ion o public ga he ings, es ic ions on public
anspo a ion, he closu e o ce ain comme cial cen e s such as es au an s and ba s,
and o cou se he sys ema ic wea ing o masks, hand disin ec ion and he s ic espec o
social dis ancing. On 13 Sep embe 2021, he Benin go e nmen made COVID-19 es ing
con enien o USD 45 and la e decided ha COVID-19 es ing o accina ion would
be manda o y o access o all public se ices, including decen alised se ices, in he
coun y. On 1 Ap il 2020, he go e nmen ’s i s social measu es o he popula ion we e
adop ed h ough he sale o masks a a go e nmen -subsidised p ice o USD 0.36, implying
a subsidy a e o 60% in wel e high- isk dis ic s in Benin (Gou e nemen Benin,2022).
On 20 Ap il 2020, social measu es con inued wi h he go e nmen -subsidised sale o 250
mg chlo oquine able s o he ea men and p e en ion o COVID-19 a a subsidised p ice
o USD 0.9 and a subsidy a e o 64%.
Economies 2025,13, 222 3 o 27
All o hese measu es o con ain he pandemic had many economic and social implica-
ions. A he mac o-le el, COVID-19 led o a decline in economic g ow h o 1.9% in 2020 in
he Sub-Saha an A ica egion (A DB,2022). In e ms o iscal policy, public spending on
heal h ca e, p i a e sec o subsidy p og ammes, and ax cu s led o an inc ease in he budge
de ici o 8.4 pe cen o GDP in 2020, double i s 2019 le el o 4.6 pe cen o GDP. Beyond
hese mac oeconomic implica ions, exis ing s udies ha e elied on mul iple eg ession and
simula ion app oaches o show ha he pandemic inc eased he po e y a e by 22.7% o
38.5% in a sample o 13 La in Ame ican coun ies (B acco e al.,2024). In compa ison, i
was a ound 60% in Ghana (Abu & Issahaku,2020), 22% in E hiopia (Yime & Alemayehu,
2021), 9% in Nige ia (Andam e al.,2020), and 4% in Bu kina Faso (Ouoba & Sawadogo,
2022). O he s udies ha e ocused on heal h expendi u e; o example, Rajalakshmi e al.
(2023) showed ha he pandemic led o a 26% inc ease in heal h expendi u es in Sou h
India. Simila ly, Ayano e e al. (2024) demons a ed ha 52% o households in Ghana spen
mo e han 5% o hei o al expendi u es on COVID-19- ela ed heal h cos s. These indings
demons a e he d as ic impac o COVID-19 ac oss he globe. Howe e , he magni ude
o he impac also a ies d as ically, aising he issue o coun y-speci ic cha ac e is ics.
Mo eo e , he unde lying mechanisms h ough which he pandemic inc eased po e y, as
well as he gende aspec s o he impac s, a e o c ucial impo ance o o mula ing policy-
ele an ecommenda ions. Though he COVID-19 pandemic has passed, ou indings
emain ele an o be e managemen o u u e pandemics.
The main ques ion unde lying his s udy is whe he and o wha ex en he COVID-19
pandemic a ec ed households’ heal h expendi u es and po e y in Benin. The objec i es o
he s udy a e h ee old: (i) o es ima e he e ec o COVID-19 on household income; (ii) o
es ima e he e ec o COVID-19 on household heal h expendi u es; and (iii) o e alua e he
dis ibu ional e ec s o COVID-19. Fo he sake o ha ing an in-dep h unde s anding o he
social implica ions o COVID-19 on household wel a e, we analysed he e ec s o COVID-
19 on he sha e o heal h expendi u es in ela ion o households’ income o de e mine
whe he he pandemic led o ca as ophic heal h expendi u es. Heal h expendi u es a e
quali ied as ca as ophic when he ou -o -pocke paymen s spen by households o heal h
ca e nega i ely a ec he households’ capaci y o sa is y hei basic needs (WHO,2001).
Though he e is no consensus on he h eshold sha e o heal h expendi u es o ca as ophic
cha ac e isa ion, mos o he de ini ions a e based on h eshold alues anging om 10% o
40% o a household’s income o non- ood expendi u es (Kockaya e al.,2021;WHO,2001;
Xu,2005).
The es o he pape de elops as ollows. The li e a u e e iew is p esen ed in
Sec ion 2. The concep ual amewo k is p esen ed in Sec ion 3. In Sec ion 4, we p esen
he empi ical s a egy and da a used. The main indings a e p esen ed and discussed in
Sec ion 5. Las ly, he concluding ema ks a e p esen ed in Sec ion 6.
2. Li e a u e Re iew
Empi ical s udies widely ag ee on he de imen al e ec s o he COVID-19 pandemic
on wel a e and po e y, bu hey di e signi ican ly in me hodology, ocus, and explana o y
powe . One dominan s and o he li e a u e adop s quan i a i e app oaches ( eg essions
and mic osimula ions), e ealing ha he pandemic has wo sened income inequali y,
inc eased household ulne abili y, and in ensi ied mul idimensional po e y.
B acco e al. (2024), using ha monised mic oda a om 13 La in Ame ican coun ies,
employed mic osimula ions o show ha he COVID-19 c isis caused a po e y inc ease
anging om 22.7% o 38.5%, and hey ad oca ed o s ong compensa o y measu es o
mi iga e long- e m impac s. Simila ly, Kang e al. (2023) analysed he impac o COVID-19
on household income using na ionally ep esen a i e su eys in Chad om 2020 and 2021.
Economies 2025,13, 222 4 o 27
Based on mul i a ia e eg ession, hey ound ha wo- hi ds o households, bo h u al and
u ban, epo ed income educ ions, wi h u ban a eas hi ha de in 2020 and u al a eas
in 2021.
In he heal h domain, mul iple s udies epo ising ca as ophic heal h expendi u es
du ing he pandemic. Haakens ad e al. (2023), using OLS eg essions on mic oda a om
Mexico and Pe u, es ima ed 5.6% and 13.5% inc eases in such expendi u es, espec i ely.
Simila ly, Rajalakshmi e al. (2023) showed ha , in Sou h India, 26% o households expe i-
enced ca as ophic heal h spending. Ayano e e al. (2024) epo ed ha , in Ghana, 52.2% o
households spen mo e han 5% o hei o al expendi u es on COVID-19- ela ed heal h
cos s, and 4.2% exceeded he 40% h eshold, highligh ing inequali y in he inancial bu den
o he pandemic.
Beyond de eloping coun ies, many s udies ha e also con i med he de imen al
impac o COVID-19 on households. Mili a u e al. (2024) used he EUROMOD ax–
bene i mic osimula ion model o s udy income changes du ing he COVID-19 pe iod
(2019–2021) and he subsequen in la ion c isis (2021–2023) in Romania. They ound ha
al hough disposable income ose ini ially, he poo es bene i ed he leas , and la e in la ion
disp opo iona ely impac ed lowe -income g oups. Howe e , hei indings e ealed ha
o e all income inequali y declined, emphasising he need o a ge ed policy suppo
o p o ec ulne able households. Kala e al. (2023) elied on he same me hodology
o in es iga e he impac o COVID-19 on income in he Eu opean Union. This s udy
inds ha COVID-19 con ainmen and mi iga ion measu es had a eg essi e impac on
income dis ibu ion, bene i ing lowe -income g oups mo e, while hei e ec i eness a ied
by coun y con ex , wi h old democ acies gene ally achie ing be e ou comes han new
democ acies, highligh ing he need o ailo ed policy mixes. Al ani e al. (2024) elied
on Recen e ed In luence Func ion eg ession o show ha COVID-19 inc eased income
inequali y in he US and B azil. Mo eo e , hey highligh ed ha he dispa i ies we e
pe sis en o e ime. Fu he mo e, Shen and Zhong (2023) ocused on bo h human and
animals and showed ha he COVID-19 pandemic nega i ely a ec ed bo h household
income as well as animal ood consump ion in China.
Despi e he obus ness o hese quan i a i e indings, a key limi a ion is hei limi ed
a en ion o causal mechanisms. These s udies documen income loss, inc eased ood
insecu i y, o educed heal h access, bu o en do no explain hei mechanisms. Mo eo e ,
he e ec s a y widely ac oss and wi hin coun ies, poin ing o he impo ance o con ex ual
and ins i u ional ac o s ha a e no su icien ly unpacked in hese analyses.
Complemen ing his esea ch is a second s and o s udies ha ely p ima ily on
quali a i e me hods, o e ing a mo e nuanced iew o he pandemic’s impac on li ed
expe iences, hough o en a he expense o gene alisabili y. Musoke e al. (2024) use
ocus g oup discussions, key in o man in e iews, and household na a i es o analyse
he social and economic e ec s o he COVID-19 lockdown in Uganda. Thei indings
e eal amily b eakdowns, inc eased gende -based iolence, ising child labou , and ood
insecu i y, along wi h de e io a ed educa ional ou comes due o school closu es. Simila ly,
Ke schbaume e al. (2024) employed quali a i e me hods wi h 151 pa icipan s in Aus-
ia o documen heigh ened po e y, social exclusion, and psychological dis ess du ing
he pandemic. In Indonesia, Ma uha e al. (2025) used a non- andom sample o 100
households ac oss 30 dis ic s o explo e bo h ma e ial and spi i ual po e y, inding ha
he pandemic has deepened bo h. In Spain, O ega-Ma in and Al a ez-Gal ez (2025)
conduc ed semi-s uc u ed in e iews wi h 23 pa icipan s o highligh how COVID-19’s
e ec s ex ended beyond physical heal h o include economic s ess, legal unce ain y, and
weakened social cohesion. Finally, Ka una a hne e al. (2025), using 22 yea s o panel da a
om 20 low-income coun ies, es ima ed he long- unning impac o he pandemic on li e

Economies 2025,13, 222 5 o 27
expec ancy, concluding ha i has signi ican ly educed li e expec ancy, pa icula ly in
coun ies wi h agile heal h sys ems.
Toge he , hese quali a i e and mixed-me hod s udies o e c ucial insigh in o he so-
cial, psychological, and gende ed dimensions o COVID-19’s impac on wel a e. Howe e ,
hei me hodological limi a ions, including hei small, non- ep esen a i e samples and he
lack o causal in e ence, limi he ex e nal alidi y and policy applicabili y o hei indings.
In sum, while he exis ing li e a u e con incingly demons a es he nega i e impac
o COVID-19 on po e y and wel a e, wo majo gaps pe sis . Fi s , he impo ance o
he impac s om quan i a i e s udies widely a ies, and hese s udies ail o iden i y he
mechanisms h ough which hese e ec s un old, especially ac oss di e en social g oups o
policy en i onmen s. Second, quali a i e s udies, while ich in con ex , su e om limi ed
quan i ica ion and gene alisabili y, p e en ing b oade policy conclusions.
3. Concep ual F amewo k
COVID-19 a ec ed household income and heal h expendi u es h ough labou ma -
ke s (demand and supply), mode a ed by go e nmen measu es ( ans e s and ba ie
measu es), and p e en i e and cu a i e heal h expendi u es by households. Figu e 3
desc ibes he channels.
Figu e 3. E ec o COVID-19 on household heal h expendi u es. Sou ce: au ho s.
The popula ion a ec ed by he COVID-19 ou b eak can be classi ied in o wo g oups,
namely he in ec ed and he suscep ible, wi h he possibili y o ha ing bo h g oups o
people wi hin he same household. On he one hand, hose in ec ed a e d awn ou o he
labou ma ke , inc easing unemploymen by educing he labou supply and leading o
a dec ease in household income. The go e nmen in e enes o empe he dec ease in
income by cash ans e s and ax edis ibu ion. The ea men expenses inc ease di ec ly
wi h heal h expendi u e (cu a i e and p e en i e). Howe e , since household income
dec eases, heal h expendi u es may dec ease as a consequence. On he o he hand, he
Economies 2025,13, 222 6 o 27
suscep ible popula ion akes p e en ion measu es ha inc ease heal h expenses while
simul aneously expe iencing income losses. Mo eo e , o p e en con amina ion, he
go e nmen akes ba ie measu es (e.g., lockdowns, in e nal and ex e nal bo de closu es,
and bans on public ga he ings) ha slow down economic ac i i ies and cause a ecession
ha a ec s he labou ma ke s h ough unemploymen due o he dec ease in labou
demand. These go e nmen al measu es lead o a dec ease in income and ul ima ely o
a dec ease in heal h expendi u es a he household le el. The channels h ough which
COVID-19 a ec s income and heal h expendi u es a e ins umen al o he e alua ion o he
e ec s o he pandemic.
4. Empi ical S a egy
The bes s a egy o e alua ing he e ec o COVID-19 on household heal h expen-
di u es is o use expe imen al—o by de aul quasi-expe imen al—impac assessmen
me hods. These me hods assume he exis ence o bo h con ol and ea men g oups. The
use o hese me hodologies is no easible in he case o COVID-19 because no one could
be excluded om exposu e o he epidemic, esul ing in he absence o an app op ia e
con ol g oup. Also, he lack o household and indi idual da a be o e and a e COVID-19
p ecludes he use o quasi-expe imen al impac assessmen me hods. In he ace o hese
empi ical cons ain s, we use a mic osimula ion app oach o assess he e ec o COVID-19.
This app oach is based on he beha io al unc ion o indi iduals h ough which hei
eac ions o a shock o policy a e e alua ed in e ms o a e ages and dis ibu ional e -
ec s, conside ing he en i e dis ibu ion (especially he ails) o he shock o policy e ec s.
Unlike Compu able Gene al Equilib ium (CGE) models, which a e c i icised o agg ega -
ing economic phenomena (Hansen & Heckman,1996), mic osimula ion models ha e he
ad an age o analysing in de ail he e ec o a shock o policy a he indi idual le el.
Mic osimula ion models ha e been widely used in ecen decades o he ex an e
e alua ion o po e y educ ion policies and exogenous shocks, including he e ec s o
COVID-19 (Andam e al.,2020;O’Donoghue e al.,2021;Ouoba & Sawadogo,2022;Yime
& Alemayehu,2021) and mos ly o social p og ams and iscal policies (Benczú e al.,2018;
Bo e e al.,2017;Maskae a e al.,2019,2021). Ou es ima ion s a egy o he sho - e m
e ec o COVID-19 ollows ha o A an e al. (2021). Ou mic osimula ion s a egy en ails
ou s eps as s a ed below.
4.1. Calcula ing he E ec s o COVID-19
4.1.1. Calcula ing COVID-19 Household Income and Heal h Expendi u e
We compu e he mon hly household heal h expendi u es and mon hly household
income be o e he COVID-19 pandemic using he Ha monized Su ey on Li ing Condi ions
o Households (EHCVM) da ase .
Conce ning household income, ou income gene a ion p ocess ollows Sologon e al.’s
(2021) app oach. The o al income is gi en by Equa ion (1):
y=yL+yK+yO+yB(1)
whe e yis he household mon hly income, y
L
e e s o he labou income, y
K
measu es he
capi al income, y
o
is o he household income, and y
B
is public bene i s ( ans e s). Con a y
o Sologon e al. (2021), who used household disposable income, we used g oss income.
This choice is mo i a ed by he ac ha he ax in o ma ion and public bene i s a e no
a ailable in ou da ase . Howe e , he consequences o his o e sigh o axes a e likely o
be negligible, because, in de eloping coun ies like Benin, labou income is ligh ly axed
and he ax cos is p obably balanced by public ans e s.
Economies 2025,13, 222 7 o 27
The labou income y
L
in a household equals he sum o indi iduals employed and
sel -employed wi hin he household, agg ega ed as ollows:
yL=
n
∑
i=1Iemployed
i∗yemployed
i+Isel −employed
i∗ysel −employed
i(2)
whe e i e e s o he indi idual membe o he household and n he size o he household.
Indi idual iis employed (
Iemployed
i
= 1) and (
Iemployed
i
= 0), wi h an income equal o
yemployed
i
.
Isel .employed
i
and
ysel .employed
a e simila ly de ined. This agg ega e o mula also add esses
he si ua ion o indi iduals who combine sala ied wo k and sel -employmen .
Household capi al income comes om in es men income and p ope y income, mos
no ably en al income. Capi al income is hen modelled as ollows:
yK=∑n
i=1Iin es men
iyin es men
i+Ip ope y
iyp ope y
i(3)
Apa om labou and capi al income, households may bene i om o he incomes,
such as p i a e pensions and ans e s ecei ed om ela i es and non ela i es. Such
sou ces o income a e compu ed as ollows:
yO=∑n
i=1Ip i a e.pension
iyp i a e.pension
i+I an e s. ela i es
iy an e s. ela i es
i+I an e s.non− ela i es
iy an e s.non− ela i es
i(4)
Finally, household income comes om public bene i s (i a ailable), bo h a he indi id-
ual le el (pensions, sickness o disabili y, unemploymen insu ance) and a he household
le el such as social secu i y p og ams ( ood dis ibu ion, cash ans e s, ee heal hca e
o child en and p egnan women). The agg ega ed income om public bene i s a he
household le el is compu ed as ollows:
yB=ysocial. sec u i y +∑n
iIpensions
iypensions
i(5)
The Wo ld Heal h O ganisa ion de ines heal h expendi u es based on heal hca e
unc ions as ou -o -pocke money spen by households in sa is ying hei co e heal hca e
needs. Those heal hca e unc ions included co e i ems such as cu a i e ca e, ehabili a i e
ca e, inpa ien ca e, ou pa ien ca e, day ca e, long- e m ca e, home-based ca e, ancilla y
se ices, pha maceu ical goods, he apeu ic appliances, and p e en i e ca e (WHO,2022b).
Following his de ini ion, we modelled he household heal h expendi u e equa ion in o
h ee componen s, namely cu a i e expendi u es
eC
, p e en i e expendi u es
eP
, and
nonmedical expendi u es eO, as ollows:
e=∑n
i=1∑jej
i, wi hi =1, 2, . . . n and j =C,P,O (6)
whe e eis he agg ega ed household mon hly heal h expendi u es o all he household
membe s. All h ee componen s o heal h expendi u es ake in o conside a ion bo h mode n
and adi ional medicine. Cu a i e heal h expendi u es a e composed o doc o consul a ion
ees
econs
, medicines
emed
, diagnos ic es ing
ediagn
, and bed cha ges o hospi alisa ion
ebed
as ollows:
eC=econs +emed +ediagn1+ebed (7)
P e en i e heal h expendi u es equal he sum o p e en a i e diagnos ic es ing
ediagn2
,
accines e ac, and o he nonmedical goods eo he such as ace masks.
eP=ediagn2+e ac +eo he (8)
Economies 2025,13, 222 8 o 27
Finally, household heal h expendi u es encompass o he nonmedical expendi u es
eo
,
no ably anspo e ans and communica ion ees ecom as ollows:
eO=e ans +ecom (9)
I is wo h no ing ha he measu e o heal h expendi u es used in his s udy e e s
exclusi ely o ou -o -pocke paymen s epo ed by households. I does no include heal h
spending co e ed by go e nmen subsidies, public insu ance schemes, o dono - unded
p og ammes. Da a on hese a iables we e no a ailable.
4.1.2. Calcula ing he COVID-19 Job Loss Index A e he Ou b eak
The COVID-19 job loss index is calcula ed by mul iplying he isk o job loss condi-
ional o he sec o o ac i i y, gende , and esidence as ollows1:
Job loss index =1=Risk(job loss|sec o k)∗Risk(job loss|gende i)∗Risk(job loss| esidencej)(10)
whe e = 1 e e s o he COVID-19 pe iod. The isk o job loss condi ional o sec o o
ac i i ies and esidence is calcula ed using he esul s o he apid e alua ion conduc ed
du ing he COVID-19 pandemic (Ma ch–July 2020) by he Na ional Ins i u e o S a is ics
and Economic Analysis (INSAE BENIN,2020). This s udy epo ed ha 80%, 77% and 74%
o Co onou ci y esiden s, o he u ban esiden s, and u al esiden s, espec i ely, did no
wo k du ing he pe iod conside ed. Equally, he isk o job loss in he wholesale and e ail
ade was he highes (37.29%), while i was 0% in he ag icul u e sec o . The ull job loss
p obabili ies acco ding o he sec o o ac i i ies, a e epo ed in Table A1 in Appendix A.
Fu he mo e, his epo indica ed ha 50% o households los hei job du ing he pe iod,
which accoun s o 87.5% o men and 13% o women. Hence, we calcula ed he p obabili ies
o job loss based on gende as ollows2:
P (job loss|gende i) = P (job loss ∩gende i)
P (gende i)(11)
The esul s om Equa ion (11) indica ed p obabili ies o job loss o 0.55 and 0.30,
espec i ely, o men and women. Then, we calcula ed he isk o job loss condi ional o
gende and esidence using a mul iplie as ollows:
Risk(job loss|gende i) = 2P (job loss|gende i)
∑2
i=12P (job loss|gende j), (12)
and
Risk(job loss| esidencek) = 3P (job loss| esidencek)
∑3
i=12P (job loss| esidencel)
The isk o job loss coe icien s by gende and esidence a e epo ed in Table A2 in
he Appendix A. This isk ac o s a e no p obabili ies, bu mul iplie s add essing he ac
ha ce ain indi iduals a e mo e suscep ible o job loss han some o he s depending on
hei cha ac e is ics (gende o place o esidence).
We conside ed he job loss indices calcula ed in Equa ion (12) as he mean shocks, so
hey a e se as he baseline assump ions (mode a e shock). The low and se e e assump ions
a e se by conside ing ha low shock is 0.5 imes he baseline and se e e is wice he
baseline assump ion. The job loss indices in he low, mode a e, and se e e assump ions a e
epo ed in Table 1.
Economies 2025,13, 222 15 o 27
5.3. E ec o COVID-19 on Income Inequali y
To assess whe he he e ec s o COVID-19 a e signi ican ly he e ogeneous, we calcu-
la ed he Gini index. The esul s epo ed in Table 7indica e ha COVID-19 would inc ease
income inequali y in Benin by 0.1 pe cen age poin . Mo eo e , he inc ease in inequali y
is mo e p onounced among men, younge indi iduals, and hose wi h highe le els o
educa ion, who appea o be he mos a ec ed g oups, as ou lined in Table 8. These indings
a e consis en wi h he magni ude o he income decline obse ed wi hin hese subg oups,
as discussed in he p e ious sec ion.
Table 7. Gini index by age and se e i y o shock.
GINI Index BEFORE LOW MODERATE SEVERE
O e all 0.622 0.623 0.624 0.628
Young 0.577 0.579 0.583 0.598
Adul 0.550 0.548 0.546 0.544
Elde ly 0.640 0.640 0.641 0.647
Table 8. Gini index by gende and se e i y o shock.
(a) Gini index by gende and se e i y o shock.
GINI index BEFORE LOW MODERATE SEVERE
O e all 0.622 0.623 0.624 0.628
Women 0.672 0.672 0.672 0.673
Men 0.562 0.562 0.563 0.571
(b) Gini index by educa ion le el and se e i y o shock.
GINI index BEFORE LOW MODERATE SEVERE
O e all 0.622 0.623 0.624 0.628
P ima y and below 0.490 0.488 0.487 0.485
Seconda y 0.573 0.574 0.577 0.586
Highe 0.377 0.383 0.393 0.405
(c) Gini index by place o esidence and se e i y o shock.
GINI index BEFORE LOW MODERATE SEVERE
O e all 0.622 0.623 0.624 0.628
Co onou 0.695 0.698 0.703 0.715
U ban 0.559 0.561 0.563 0.575
Ru al 0.599 0.599 0.600 0.606
Sou ce: au ho s.
5.4. E ec o COVID-19 on Household Heal h Expendi u es
As depic ed in ou concep ual amewo k, i is a gued ha he COVID-19 c isis
has a ec ed (inc eased) household heal h expendi u es h ough di ec and indi ec chan-
nels. Indeed, he heal h expendi u es o households inc ease wi h exposu e o COVID-19
con amina ion isk. Thus, he expendi u es o a e ing measu es (hand sani ise , ace
masks, accine, COVID-19 es s, e c.) and money spen ou -o -pocke on medical isi s
and medicine di ec ly inc ease household heal h expendi u es. The e ec s o COVID-19 on
heal h expendi u e we e examined by es ima ing he heal h expendi u e equa ion in (14).
The d i e s o heal h expendi u e we e es ima ed using mul ile el mixed-e ec s linea
eg ession (Table 9). The alida ion pa ame e s o he model show ha he andom in e cep
a he household le el signi ican ly a ies, as indica ed by a a iance coe icien equal o

Economies 2025,13, 222 16 o 27
1.38 wi h a s anda d e o coe icien equal o 0.40. This esul indica es ha igno ing he
andom e ec by es ima ing only he ixed e ec would lead o biased es ima es. Mo eo e ,
he likelihood a io (LR) es s ongly suppo s ha he mixed-e ec s model be e i s
ou da a han he o dina y leas squa es (OLS) model, as con i med by he p- alue o
0.000. Thus, he esul s o he ixed-e ec s componen s e ealed in Table 9indica e ha
income, age, and isk o diseases such as cough, high blood p essu e, mala ia, and oad
acciden s ha e posi i e and signi ican e ec s on heal h expendi u e. No ably, a 10% income
imp o emen inc eases heal h expendi u e by 3%, con i ming p e ious indings in A ica.
Indeed, Olasehinde and Olaniyan (2017) elied on he o dina y leas squa es echnique
o es ima e he de e minan s o household heal h expendi u e in Nige ia. They ound
ha a 1% inc ease in household income inc eases heal h expendi u es by 0.57%. Ampaw
e al. (2019) used a na ionally ep esen a i e sample o 16,772 households o analyse
he e ec o income on heal h expendi u e in Ghana. Thei indings om o dina y leas
squa es eg ession con i med a posi i e and signi ican e ec o household o al income
on heal h expendi u e. Equally, Houenin o and Assou o (2023) ecen ly used a sample
o 20 low-income A ican coun ies o e 1995–2018 o calcula e income elas ici ies. Based
on panel-pooled mean g oup es ima es, hey concluded ha a 1% imp o emen in pe
capi a income inc eases p i a e heal h expendi u e by 0.54% in he long un. Fu he mo e,
ou indings e ealed ha households esiding in u al and u ban a eas a e less likely o
ha e highe heal h expendi u es han hose in he capi al ci y o Co onou. Ou esul s
u he con i med he nonlinea ela ionship be ween age and heal h expendi u e, as ound
in p e ious s udies such as hose o Ampaw e al. (2019) in Ghana and Olasehinde and
Olaniyan (2017) in Nige ia.
Table 9. Mul ile el mixed-e ec s linea eg ession model o household heal h expendi u e.
Log (Heal h Expendi u e) Coe . S . E . [95% Con In e al]
Log (income) 0.292 *** 0.089 0.117 0.466
Gende head household ( e = emale)
Male −0.191 0.181 −0.546 0.165
Log (Age) 0.035 ** 0.017 0.002 0.067
Log (Age squa e) −0.001 ** 0.000 −0.001 0.000
Educa ion o head household ( e = p ima y)
Seconda y −0.124 0.231 −0.576 0.328
Highe −0.221 0.309 −0.825 0.384
Heal h co e age ( e = no) −0.362 0.435 −1.215 0.490
Residence ( e = Co onou)
U ban −0.523 ** 0.252 −1.017 −0.029
Ru al −0.731 *** 0.274 −1.268 −0.194
Cough las 3 mon hs ( e = no) 1.817 *** 0.414 1.004 2.629
High blood p essu e ( e = no) 2.729 *** 0.679 1.397 4.060
Mala ia las 3 mon hs ( e =no) 0.637 *** 0.269 0.110 1.164
Road acciden las 3 mon hs ( e = no) 3.292 *** 0.958 1.414 5.170
Cons an 3.579 *** 0.994 1.631 5.527
Random-e ec s pa ame e s
In e cep a household le el 1.383 0.405 0.779 2.456
Residual a iance 2.281 0.387 1.635 3.181
Mean dependen a 2.078
Numbe o obs 78.188
P ob > chi2 1761.402
No e: Any s a is ically signi ican es ima es a e deno ed wi h as e isks: * p< 0.10, ** p< 0.05, *** p< 0.01.
Economies 2025,13, 222 17 o 27
We now simula e he e ec s o COVID-19 on heal h expendi u e. The esul s p esen ed
in Figu e 5indica e ha households expe ienced an inc ease in heal h expendi u es a e
he COVID-19 shock, wi h he e ogeneous e ec s. Indeed, a dec ease in household income,
ce e is pa ibus, will lead o a dec ease in household heal h expendi u es. Be o e COVID-
19, he p e-COVID-19 shock cu e, as shown in Figu e 5, was abo e he pos -COVID-19
one. In addi ion, he esul s show ha he e ec o COVID-19 on heal h expendi u es
is he e ogeneous, meaning ha households wi h low heal h expendi u es we e he mos
a ec ed by he COVID-19 pandemic. The gap be ween he be o e-COVID-19 cu e and he
a e -COVID-19 cu es dec eases om low heal h expendi u es o high heal h expendi u es.

Figu e 5. Household heal h expendi u es6. Sou ce: au ho s.
These esul s can be explained by he ac ha households wi h low heal h expendi-
u es a e gene ally cha ac e ised by low income. This esul con i ms ou ea ly indings,
indica ing ha he low-income g oups o households we e he mos a ec ed by COVID-19.
Mo e impo an ly, ou summa y s a is ics epo ed ha less han 1% o Benin’s households
do no ha e heal h co e age. Indeed, hough COVID-19 dec eased bo h household income
and heal h expendi u e, he inc ease in heal h expendi u es indica es ha he decline in
income was g ea e han ha o heal h expendi u es.
Figu e 6depic s he e ec s o he COVID-19 c isis on households’ ca as ophic heal h
expendi u es. Two main conclusions can be d awn om hese esul s. Fi s , he COVID-19
c isis pushed a signi ican sha e o households in o ca as ophic heal h expendi u es, wi h a
g ea e e ec when he magni ude o he shock inc eased. No ably, he sha e o households
wi h ca as ophic heal h expendi u es a he 10% h eshold inc eased om 16% be o e he
COVID-19 pe iod o 20% a e a se e e shock. Second, hese indings a e obus when
mo ing om a 10% o a 40% h eshold o ca as ophic heal h expendi u es.
A gende analysis o he ca as ophic heal h expendi u es o he COVID-19 c isis
e ealed ha men we e he mos a ec ed (Figu e 7). Al hough he sha e o women wi h
ca as ophic heal h expendi u es was g ea e han ha o men be o e he pandemic, i
is wo h no ing ha he sha e o men wi h ca as ophic heal h expendi u es a e he
COVID-19 pandemic inc eased mo e han ha obse ed among women. Fu he mo e,
he egional analysis also suppo s he e idence o he e ogeneous e ec s o he COVID-19
pandemic on household heal h expendi u es. Mo e speci ically, he esul s om simula ions
indica ed ha u ban esiden s expe ienced he g ea es bu den in e ms o an inc ease in
Economies 2025,13, 222 18 o 27
heal h expendi u es, hough Co onou esiden s we e he mos ulne able be o e COVID-19
(Figu e 8).
Figu e 6. Sha e o households wi h CHE7. Sou ce: au ho s.
Figu e 7. Change in he sha e o households wi h CHE by gende (pe cen age poin s)
8
.
Sou ce: au ho s.
5.5. Discussion
Ou indings e ealed ha , ega dless o he conside ed scena io (low, mode a e, and
se e e COVID-19 shock), households in u ban a eas saw hei income decline as e han
hose in u al a eas. Equally, low-income households and men we e he mos ulne able o
COVID-19. This can be explained by he ac ha , in u ban a eas, he comme cial sec o ,
ep esen a i e o he mos employed sec ion o u ban esiden s, was s ongly a ec ed by
COVID-19. Indeed, he public heal h measu es unde aken by he go e nmen o con ain
he sp ead o he pandemic, such as limi a ions on public ga he ings, bo de closu es, and
he es ic ion imposed on public anspo a ion, d as ically limi ed households’ day- o-day
ac i i ies. Equally, low-income households a e gene ally hose wi h low skills, meaning ha
hey canno wo k emo ely as skilled wo ke s. These indings align wi h hose o Labo de
e al. (2021) and Andam e al. (2020), who ound ha he incomes o u ban households
we e he mos a ec ed by he COVID-19 pandemic compa ed o u al households, who
los an a e age o 18% o hei income.
Economies 2025,13, 222 19 o 27

Figu e 8. Change in CHE by place o esidence (pe cen age poin s)9. Sou ce: au ho s.
We ound ha he sha e o household heal h expendi u es be ween he be o e COVID-
19 and a e COVID-19 pe iods is inc easing wi h ega d o he magni ude o he shock.
This esul can be explained by he decline in household income on he one hand and he
inc ease in p e en i e and cu a i e heal h expendi u es on he o he . Thus, based on 10%
and 20% h esholds, he simula ions showed ha he sha e o households wi h ca as ophic
heal h expendi u es inc eased wi h he magni ude o COVID-19 shock. Mo e impo an ly,
he he e ogeneous e ec s on heal h expendi u es e ealed ha households wi h a highe
sha e o heal h expendi u es be o e he pandemic expe ienced a g ea e inc ease in heal h
expendi u es. Be o e he COVID-19 pandemic, he p opo ion o households in u al a eas
wi h ca as ophic heal h expendi u es was highe han ha o u ban households, including
Co onou. Simila ly, he p opo ion o households wi h ca as ophic heal h expendi u es
conside ing di e en shock scena ios (low, mode a e, and se e e) was high in u ban a eas,
especially in Co onou, compa ed o households in u al a eas. These esul s a e in line
wi h hose o Rajalakshmi e al. (2023), who showed ha du ing he COVID-19 pandemic,
ca as ophic household heal h expendi u e inc eased. We explain his esul by he ac ha
he es ablished co don sani ai e limi ed he mo emen o humans and goods be ween u al
and u ban a eas as much as possible, which made i possible o educe he isks o in ec ion
in u al a eas and he e o e limi medical expenses.
Ou indings e eal ha he nega i e impac o he COVID-19 pandemic on household
income was mo e p onounced among male-headed households, mo e educa ed ones,
young headed-households, and u ban esiden s, wi h a simul aneous ise in ca as ophic
heal h expendi u es. These esul s can be explained by he socioeconomic s uc u e o
Benin. Households headed by younge indi iduals appea mo e a ec ed by COVID-19-
ela ed income shocks because hey a e mo e likely o be engaged in in o mal, uns able, o
en y-le el jobs, such as pe y ade o app en iceships, ha we e se e ely dis up ed du ing
lockdowns. In con as , olde household heads may bene i om mo e s able occupa ions
o accumula ed esou ces. Male-headed households expe ienced g ea e income losses,
which e lec s hei highe concen a ion in sec o s hea ily impac ed by he c isis, such as
anspo a ion, cons uc ion, and la ge-scale ade. Female-headed households, al hough
o en mo e economically ulne able o e all, a e mo e likely o be in ol ed in small-scale,
communi y-based comme ce, which p o ed ela i ely esilien . In e es ingly, households
wi h mo e educa ed heads su e ed la ge income declines. This may be due o hei
highe dependence on o mal sec o employmen , which was mo e exposed o wage
cu s, layo s, and con ac suspensions du ing he pandemic. Finally, u ban households,
Economies 2025,13, 222 20 o 27
pa icula ly hose in Co onou, we e he mos a ec ed, due o hei g ea e exposu e o
mobili y es ic ions, ma ke dis up ions, and he in o mal u ban economy’s ulne abili y
o shocks. In con as , u al households, while gene ally poo e , end o ely mo e on
subsis ence ag icul u e and in o mal mu ual suppo sys ems, which o e ed a ela i e
bu e agains income losses. These indings highligh he di e en ia ed impac o COVID-
19 and sugges ha eco e y policies should be sensi i e o age, gende , educa ion le el,
and place o esidence.
Fu he mo e, he inc eased ca as ophic heal h expendi u e sugges s a heigh ened
ulne abili y o ou -o -pocke heal h cos s du ing he pandemic, possibly due o limi ed
access o subsidised ca e o inc eased eliance on p i a e heal h p o ide s in eme gency
con ex s. Fu he mo e, he disp opo iona e e ec on u ban households may e lec he
concen a ion o COVID-19 cases, he s ic e en o cemen o con ainmen measu es, and
he highe cos o li ing in u ban a eas compa ed o u al ones. U ban esiden s also end
o depend mo e on cash-based, non-ag icul u al li elihoods ha we e acu ely sensi i e o
lockdowns and economic dis up ions. While 39.5% o households al eady li e in po e y
(INSAE BENIN,2019), hese indings unde sco e he impo ance o ailo ing social p o ec-
ion esponses o accoun o gende dynamics and spa ial inequali ies, pa icula ly in he
design o pos -c isis eco e y and heal h inancing policies.
I is wo h no ing ha his s udy is subjec o se e al limi a ions, and as such, he
indings should be in e p e ed wi h cau ion. A key limi a ion conce ns he use o sel -
epo ed da a. The su ey da ase s ely on household- epo ed in o ma ion ela ed o
heal h and income, which may be subjec o epo ing biases, no ably ecall bias o social
desi abili y bias. In addi ion, due o he absence o ax- ela ed in o ma ion in he da ase s,
g oss income was used as a p oxy o household wel a e ins ead o disposable income. This
empi ical choice may lead o an unde es ima ion o he impac o COVID-19 on he a io o
heal h expendi u es o income, sugges ing ha pos -COVID-19 heal h- ela ed spending
may ha e been e en mo e ca as ophic han indica ed by he indings o his s udy.
6. Conclusions and Policy Implica ions
In his s udy, we in es iga ed he socioeconomic implica ions o he COVID-19 pan-
demic in Benin. To his end, a na ionally ep esen a i e da ase om he Benin household
li ing condi ions su ey and a apid su ey a e he COVID-19 ou b eak we e used o
empi ical es ima ions. We adop ed s a ic mic osimula ion modelling o es ima e he sho -
e m e ec o COVID-19 on household income and heal h expendi u es unde di e en
scena ios o COVID-19 shock. The esul s showed ha he COVID-19 pandemic had a
huge nega i e e ec on household income, on he one hand, due o he closu e o bo de s,
which limi ed ade be ween Benin and i s neighbou ing coun ies, and on he o he hand,
because o he sani a y co don ha limi ed ade be ween u ban owns and u al a eas.
Simila ly, COVID-19 inc eased household heal h expendi u es wi h he e ogeneous e ec s
ac oss income le els, place o esidence and gende . Due o he absence o ax in o ma ion
in ou da ase , we used g oss income ins ead o disposable income as a measu e o wel a e.
This empi ical s a egy may lead o he unde alua ion o he e ec s o COVID-19 on he
heal h expendi u es–income a io, meaning ha he pos -COVID-19 heal h expendi u es
may be mo e ca as ophic han he esul s o his s udy ha e shown. These indings ha e
implica ions o heal h inancing in Benin. Ta ge ed heal h co e age policies, along wi h
social policies aimed a educing employmen ulne abili y and a ge ed cash ans e s, a e
necessa y o add ess he ulne abili y o households o heal h c ises in Benin.
The go e nmen needs o a ge he poo es households led by men and u ban esi-
den s o heal h co e age p og ammes o add ess he ulne abili y o households o heal h
c ises, as highligh ed by he COVID-19 pandemic. Though Benin’s go e nmen , wi h he

Economies 2025,13, 222 21 o 27
suppo o o ganisa ions like he WHO, has been wo king owa ds expanding heal h co e -
age and implemen ing measu es o imp o e access o heal hca e o he mos ulne able
popula ions
10
, i is c ucial o ocus on bo h he supply and demand sides when o mula ing
heal h policies. Acco ding o he 2018 demog aphic and heal h su eys epo , only 1.2%
o he popula ion in Benin had access o heal h co e age, while 38.5% li ed below he
mone a y po e y line. These da a emphasise ha e en in he absence o heal h pandemics,
a signi ican po ion o he popula ion aces di icul ies in accessing heal h due o inancial
cons ain s. Uni ed Na ions o ganisa ions such as he Wo ld Heal h O ganisa ion (WHO),
ci il socie y, and local and o eign p i a e sec o s could play a co e ole in suppo ing
go e nmen s o design and mobilise sus ainable inancing.
The Benin go e nmen should design social policies aimed a educing employmen
ulne abili y among he popula ion, no ably o u ban esiden s and he poo es households
led by men. The p i a e sec o could be he co e pa ne in imp o ing he condi ions o
job ulne abili y. The COVID-19 pandemic has led o income loss, disp opo iona ely
a ec ing lowe -income quin iles. While he Benin Na ional Ins i u e o S a is ics and
Demog aphy epo ed an unemploymen a e o abou 2% in 2018, wi h a high a e o
ulne able employmen a 84.1%, i is wo h no ing ha he go e nmen add esses he
issue o unde -employmen , especially du ing pe iods o c ises. In addi ion o a ge ed
heal h co e age policies and educing employmen ulne abili y, i is essen ial o design
a ge ed cash ans e p og ams o educe he ulne abili y o he poo es households when
pandemics occu .
Au ho Con ibu ions: Concep ualiza ion, A.N.H. and N.B.; me hodology, A.N.H. and N.B.; so -
wa e, N.B.; alida ion, A.N.H.; o mal analysis, A.N.H. and N.B.; in es iga ion, A.N.H. and N.B.,
C.B.D.; esou ces, A.N.H. and N.B., C.B.D.; da a cu a ion, NB and C.B.D.; w i ing—o iginal d a
p epa a ion, A.N.H. and N.B.; w i ing— e iew and edi ing, A.N.H. and N.B., C.B.D.; isualiza ion,
N.B.; supe ision, A.N.H.; p ojec adminis a ion, A.N.H.; unding acquisi ion, A.N.H. and N.B.,
C.B.D. All au ho s ha e ead and ag eed o he published e sion o he manusc ip .
Funding: A ican Economic Resea ch Conso ium (AERC): RC225/6.
Ins i u ional Re iew Boa d S a emen : No applicable.
In o med Consen S a emen : No applicable.
Da a A ailabili y S a emen : The aw da a suppo ing he conclusions o his a icle will be made
a ailable by he au ho s on eques .
Con lic s o In e es : The au ho s decla e no con lic s o in e es .
Appendix A
Table A1. Sec o al job ulne abili y coe icien s.
Type o Ac i i y Pe cen
1. Ag icul u e, ishing, and o es y 0.00
2. Manu ac u ing and mining 16.95
3. Cons uc ion 18.64
4. Wholesale and e ail ade, epai o mo o ehicles, and
o he mo o s. 37.29
5. Ho els and es au an s 3.39
6. T anspo and communica ion 5.08
7. Educa ion and public adminis a ion 0.00
8. Heal h and social wo k ac i i ies 6.78
9. O he se ices 11.86
Economies 2025,13, 222 22 o 27
Table A2. Risk o job loss coe icien s by gende and esidence.
Gende Residence
Male Female Co onou U ban Ru al
Risk
coe icien s 1.29 0.71 1.04 1.00 0.96
Table A3. Heal h expendi u es be o e and a e COVID-19.
Household Heal h
Expendi u es Obs Mean S d. De . Min Max
Be o e COVID-19 899 2158.64 4655.426 0 50,250
Low shock 899 86,545.321 193,570.45 0 2,835,360
Mode a e shock 899 81,208.87 182,267.61 0 2,670,720.3
Se e e shock 899 73,968.176 169,792.54 0 2,506,080.3
Table A4. Summa y s a is ics o ca as ophic heal h expendi u es.
Th eshold Obs Mean S d. De . Min Max
Ca as ophic heal h expendi u es (Be o e COVID-19)
10% 899 0.165 0.371 0 1
20% 899 0.083 0.277 0 1
30% 899 0.051 0.22 0 1
40% 899 0.034 0.183 0 1
Ca as ophic heal h expendi u es (A e
COVID-19-low shock)
10% 899 0.175 0.38 0 1
20% 899 0.086 0.28 0 1
30% 899 0.055 0.227 0 1
40% 899 0.036 0.185 0 1
Ca as ophic heal h expendi u es (A e
COVID-19-ode a e shock)
10% 899 0.181 0.385 0 1
20% 899 0.092 0.29 0 1
30% 899 0.058 0.234 0 1
40% 899 0.038 0.191 0 1
Ca as ophic heal h expendi u es (A e
COVID-19-se e e shock)
10% 899 0.199 0.4 0 1
20% 899 0.102 0.303 0 1
30% 899 0.065 0.246 0 1
40% 899 0.042 0.201 0 1
Table A5. Summa y s a is ics o heal h expendi u e by gende .
Th eshold Scena io Man Woman
10%
Be o e COVID-19 0.136 0.195
Low 0.149 0.202
Mode a e 0.156 0.209
Se e e 0.181 0.218
Economies 2025,13, 222 23 o 27
Table A5. Con .
Th eshold Scena io Man Woman
20%
Be o e COVID-19 0.073 0.094
Low 0.076 0.096
Mode a e 0.084 0.101
Se e e 0.099 0.106
30%
Be o e COVID-19 0.043 0.06
Low 0.045 0.064
Mode a e 0.052 0.064
Se e e 0.056 0.073
40%
Be o e COVID-19 0.03 0.039
Low 0.032 0.039
Mode a e 0.037 0.039
Se e e 0.043 0.041
Table A6. Summa y s a is ics o heal h expendi u es by esidence.
Th eshold Co onou U bain Ru al
10%
Be o e
COVID-19 0.188 0.155 0.166
Low 0.194 0.17 0.172
Mode a e 0.208 0.177 0.175
Se e e 0.215 0.198 0.194
20%
Be o e
COVID-19 0.076 0.085 0.085
Low 0.076 0.087 0.087
Mode a e 0.097 0.095 0.087
Se e e 0.104 0.102 0.101
30%
Be o e
COVID-19 0.056 0.052 0.048
Low 0.056 0.058 0.051
Mode a e 0.056 0.06 0.056
Se e e 0.063 0.065 0.065
40%
Be o e
COVID-19 0.035 0.037 0.031
Low 0.035 0.04 0.031
Mode a e 0.042 0.04 0.034
Se e e 0.049 0.048 0.034
Economies 2025,13, 222 24 o 27

Figu e A1. Heal h expendi u es a e a low shock.

Figu e A2. Heal h expendi u es a e a mode a e shock.
No es
1
No e ha in he absence o in o ma ion on job loss by gende and esidence wi hin he sec o o ac i i ies, we assume independence
be ween he sec o o ac i i y, gende , and esidence. Wi h he absence o job s a us o indi iduals who ha e los hei job, we
assume a andom selec ion o job loss wi hin each sec o , each gende , and each esidence. Hence, we assume ha he pe cen age
o job loss equals he pe cen age o income loss.
2
In he absence o in o ma ion on he gende composi ion o he INSAE BENIN (2020) s udy, we assume ha his s udy is na ionally
ep esen a i e, so we used he gende composi ion o he Ha monised Su ey on Li ing Condi ions o Households in Benin
(INSAE BENIN,2019). These da a indica ed ha 78.55% and 21.45% o heads o households we e men and women, espec i ely.
3
Recall ha wi hin he same household, membe s can ha e di e en b anches o ac i i y, places o esidence, and gende s. To
ake in o conside a ion he composi ion o households, we calcula ed he income loss coe icien a he indi idual le el and hen
agg ega ed i a he household le el.
4The Chi-2 dis ance o he CDM es was 27.63, and he p- alue equaled 0.016.
5
Fo a be e isualisa ion o he plo , we limi ed he uppe alue o he dis ibu ion o income o 100,000 F CFA. Full desc ip i e
s a is ics a e epo ed in Table 4.
6
Fo a be e isualisa ion o he plo , we limi ed he uppe alue o he dis ibu ion o heal h expendi u es o 10000 F CFA. Plo s
o low and mode a e shocks a e epo ed in Figu es A1 and A2 in Appendix A. Full desc ip i e s a is ics a e epo ed in Table A3
in Appendix A.