Fab izi, En ico; Mussida, Chia a; Pa isi, Ma ia Lau a
A icle — Published Ve sion
Ma e ial and social dep i a ion among one-pe son
households: he ole o gende
Jou nal o Popula ion Economics
P o ided in Coope a ion wi h:
Sp inge Na u e
Sugges ed Ci a ion: Fab izi, En ico; Mussida, Chia a; Pa isi, Ma ia Lau a (2025) : Ma e ial and social
dep i a ion among one-pe son households: he ole o gende , Jou nal o Popula ion Economics,
ISSN 1432-1475, Sp inge , Be lin, Heidelbe g, Vol. 38, Iss. 1,
h ps://doi.o g/10.1007/s00148-025-01084-5
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/318552
S anda d-Nu zungsbedingungen:
Die Dokumen e au EconS o dü en zu eigenen wissenscha lichen
Zwecken und zum P i a geb auch gespeiche und kopie we den.
Sie dü en die Dokumen e nich ü ö en liche ode komme zielle
Zwecke e iel äl igen, ö en lich auss ellen, ö en lich zugänglich
machen, e eiben ode ande wei ig nu zen.
So e n die Ve asse die Dokumen e un e Open-Con en -Lizenzen
(insbesonde e CC-Lizenzen) zu Ve ügung ges ell haben soll en,
gel en abweichend on diesen Nu zungsbedingungen die in de do
genann en Lizenz gewäh en Nu zungs ech e.
Te ms o use:
Documen s in EconS o may be sa ed and copied o you pe sonal
and schola ly pu poses.
You a e no o copy documen s o public o comme cial pu poses, o
exhibi he documen s publicly, o make hem publicly a ailable on he
in e ne , o o dis ibu e o o he wise use he documen s in public.
I he documen s ha e been made a ailable unde an Open Con en
Licence (especially C ea i e Commons Licences), you may exe cise
u he usage igh s as speci ied in he indica ed licence.
h p://c ea i ecommons.o g/licenses/by/4.0/
Jou nal o Popula ion Economics (2025) 38:10
h ps://doi.o g/10.1007/s00148-025-01084-5
ORIGINAL PAPER
Ma e ial and social dep i a ion among one-pe son
households: he ole o gende
En ico Fab izi1·Chia a Mussida1,2 ·Ma ia Lau a Pa isi3
Recei ed: 17 Ma ch 2024 / Accep ed: 10 Janua y 2025
© The Au ho (s) 2025
Abs ac
We explo e whe he gende has a s a is ically signi ican impac on ma e ial and social
dep i a ion o single adul s a e accoun ing o o he cha ac e is ics. We use da a
om he 2022 Eu opean Union S a is ics on Income and Li ing Condi ions su ey
o six Eu opean coun ies. By assuming dep i a ion as an indi idual la en ai and
by ea ing di e en dep i a ion le els as anked ca ego ies, we es ima e a p opo -
ional odds model sepa a ely by coun y. Ou indings sugges a clea ole o gende ,
i.e., he isk o being ma e ially and socially dep i ed is ela i ely highe o women
e e ywhe e. The no el y is ha acing “unexpec ed expenses” is he wo s ouble
o women, clea ly coming om ela i e inancial and economic agili y. Indi idual
cha ac e is ics play a mo e impo an ole han mo e agg ega e indica o s a explaining
he isk o ma e ial and social dep i a ion. Finally, he es ima ed gende gap is obus
o a la ge se o changes in model speci ica ion and assump ions.
Keywo ds Gende gap ·Ma e ial and social dep i a ion ·P opo ional odds
eg ession model ·Financial agili y
Responsible edi o : Klaus F. Zimme mann
BMa ia Lau a Pa isi
[email p o ec ed]
En ico Fab izi
[email p o ec ed]
Chia a Mussida
[email p o ec ed]
1Depa men o Economic and Social Sciences, Uni e si á Ca olica del Sac o Cuo e,
Via Emilia Pa mense 84, 29122 Piacenza, I aly
2GLO, Essen, Ge many
3Depa men o Economics and Managemen , Uni e si y o B escia, Via San Faus ino 74/b,
25122 B escia, I aly
0123456789().: V,- ol 123
10 Page 2 o 25 E. Fab izi e al.
1 In oduc ion
Unlike analysis o he labo ma ke (Oli e i and Pe ongolo 2008; Cas ellano and
Rocca 2020), gende di e en ials in po e y and social exclusion a e di icul o assess.
Mos , i no all, measu es a e household-based, hus implici ly assuming equi y o
esou ce sha ing wi hin he household, unde a Becke ian, o uni a y, concep ion o
he household (Becke 1991). Despi e he limi ed in o ma ion usually a ailable om
la ge social su eys, a s eam o li e a u e has ied o shed ligh on wi hin-household
inequali ies (Benne 2013; Co si e al. 2016; Guio and Van den Bosch 2020; Ka a-
giannaki and Bu cha d 2020).
We in es iga e gende di e en ials in ma e ial and social dep i a ion by ocusing
on single adul households, wi h he e e ence adul aged be ween 18 and 64 (wo king-
age, non- e i ee). The eason o conside ing his household ype is wo old: i s , i
ci cum en s he p oblem o assessing in a-couple o wi hin-household inequali y;
second, single-pe son households a e on he ise (Ka agiannaki and Bu cha d 2020)
in Eu opean socie ies and a e pa icula ly exposed o he isk o po e y (Chzhen and
B adshaw 2012; T eano 2018). We es ic ou a en ion o wo king-age, non- e i ed
indi iduals as po e y and social exclusion o he elde ly dese e o be ea ed as
sepa a e p oblems.
Among he nume ous measu es o po e y, we ocus on ma e ial and social dep i-
a ion, which is ou inely moni o ed in he EU (and o he Eu opean) coun ies ia he
EU-SILC su eys. Ma e ial and social dep i a ion ocuses on he ex en o which he
esou ces a ailable o a household ma ch he ac ual needs o i s membe s (No en and
Guio 2019); his abili y no only e lec s he adequacy o income bu also addi ional
asse s ha can o canno be a ailable o he household, such as sa ings, gi s, in e -
household ans e s, o se ices use ul o inance he li ing s anda d (Is ael 2016). The
measu emen is based on a se o hi een i ems, co e ing se e al domains bo h a he
household and a he indi idual le el (Guio e al. 2016,2017) and lies on he idea o
en o ced lack: a li ing condi ion is labeled as in dep i a ion i i is en o ced by lack
o esou ces and no by choice.
Published a es o ma e ial and social dep i a ion a e based on a h eshold o 5
lacking i ems ou o 13 (dep i a ion is se e e when an indi idual lacks 7 i ems). Ou
analysis goes beyond ha . By assuming ma e ial and social dep i a ion as an indi idual
la en ai and by ea ing di e en dep i a ion le els as anked ca ego ies, we es ima e
a p opo ional odds model. Ou esea ch objec i e is o assess whe he single women
a e mo e exposed han single men o he isk o ma e ial and social dep i a ion. I his
gap is obse ed, can we explain i in e ms o obse able he e ogeneous cha ac e is ics
(indi idual cha ac e is ics, household cha ac e is ics, and mac o indica o s)? Wha is
he (da a-d i en) main explana ion o which he isk is no gende -neu al?
As he answe o hese ques ions may depend on he mac oeconomic and social
en i onmen , we conduc ou analysis sepa a ely o six EU coun ies: he i e mos
popula ed coun ies in he EU (Ge many, Spain, F ance, I aly, and Poland) plus Swe-
den, included as a ep esen a i e o No he n coun ies whose socie ies s and ou o
hei balance be ween gende oles and wel a e egimes. The coun ies o choice co -
espond o di e en wel a e egimes, labo ma ke s, and ins i u ions. A seconda y aim
123
Ma e ial and social dep i a ion among one-pe son households… Page 3 o 25 10
o ou esea ch is o assess whe he he e is a di e ence in he es ima ed gende gap
among hese coun ies.
The pape is o ganized as ollows. A e e iewing he ele an li e a u e in
Sec ion 2, we in oduce he model in Sec ion 3and he da a and me hodology in
Sec ion 4; Sec ion 5shows he able o esul s wi h a discussion; Sec ion 6 epo s
se e al analyses o obus ness; and Sec ion 7d aws concluding ema ks and policy
implica ions.
2 A e iew o he li e a u e
The issue o analyzing gende di e ences in po e y and social exclusion, as explained
in he In oduc ion o his pape , is complex. Fo ins ance, Co si e al. (2016) ecognize
ha o icial measu es o a - isk o po e y based on he assump ion o equal sha ing
o households’ esou ces incu a se ious isk o unde es ima ing he ue gende gap
in po e y. The exis ing household da a and coun y he e ogenei y, howe e , a e no
able o explain he a ailabili y and sha ing o esou ces wi hin he household, hus
iden i ying he ue di e ence in he isk o po e y o ma e ial dep i a ion be ween
men and women. None heless, a s eam o li e a u e has s a ed shedding ligh on
wi hin-household inequali ies (Benne 2013; Co si e al. 2016; Guio and Van den
Bosch 2020; Ka agiannaki and Bu cha d 2020).
One way o sol e o he iden i ica ion issue is wo king on sample selec ion, e.g.,
selec ing single heads o household o bo h sexes—i.e., no in a cohabi ing couple,1
and i necessa y single pa en s (Ch is ophe e al. 2002; Wiepking and Maas 2005).2
This ype o households is sp eading ou in Eu ope (see, o ins ance, Ka agiannaki and
Bu cha d 2020), changing he a e age amily s uc u e, and i is pa icula ly exposed
o he isk o social exclusion (Chzhen and B adshaw 2012; T eano 2018).
We lea n ano he way in Guio and Van den Bosch (2020), one o he ew a emp s
o es ima e he gende gap in ma e ial dep i a ion o ma ied and cohabi ing couples
(who may li e wi h o he adul s) in 23 Eu opean coun ies. The possibili y o iden i y
dep i a ion be ween emale and male indi iduals in he couple is gi en by he in e iew
mode.3In almos e e y coun y, women u n ou o expe ience an en o ced lack o
pocke money and leisu e, wi h some he e ogenei y ac oss i ems and coun ies.
A hi d loose a emp is ound in Aisa e al. (2019)’s con ibu ion on a sample
o wo king men and wo king women in 25 EU coun ies; in gene al, he li e a u e
ecognizes ha emale-headed households, especially i hey a e single pa en s, ace
an abo e-a e age isk o po e y and ma e ial dep i a ion (see, e.g., Chan 2003;
2004). Bá cena-Ma ín e al. (2014) sugges ha indi iduals who li e in households
whose e e ence pe son is a woman, o single pa en , ha e highe in ensi y o dep i a-
1Lone indi iduals li e on one’s own wi hou o he adul s in he household; single indi iduals may li e wi h
o he cohabi ing non-pa ne adul s (e.g., hei own pa en s) wi h whom hey may ha e (pa ially) common
inancial esou ces (P o enche and Ca l on 2018).
2We ollow pa o Wiepking and Maas (2005) in he spi i , by using a sample o single adul households.
3In his s udy, un in 2015, he i ems subjec o ma e ial dep i a ion a e 6: clo hes, a pai o new shoes, ge
oge he wi h iends, leisu e ime, pocke money, in e ne connec ion. See Sec ion 4and Table 1 o de ails
abou dep i a ion i ems.
123
10 Page 4 o 25 E. Fab izi e al.
ion. Papadopoulos and Tsakloglou (2016) analyze he long- e m ela ionship be ween
ma e ial dep i a ion and po e y in Eu ope be o e he c isis pe iod: women o med a
“medium- isk” g oup in ch onic ma e ial dep i a ion in all coun ies, being 1 o 1.5
imes a highe isk o accumula e ma e ial disad an age han he popula ion a e age.
Mussida e al. (2023) use EU-SILC da a o he pe iod 2014–2018 o ind ha he isk
o alling in o (low and high in ensi y o ) se e e ma e ial dep i a ion o emale heads
o household is abou 2.8 imes highe han hei male equi alen , ac oss he Spanish
communi ies.
Finally, in an e en mo e gene al se ing, we e iewed a icles which es ima e he
de e minan s o he p obabili y o alling in o ma e ial dep i a ion ac oss coun ies;
among he explana o y a iables we ind gende (No en and Guio 2020); age (see
Bá cena-Ma ín e al. 2014; Guio and Van den Bosch 2020; Dudek and Szczesny
2021); age and numbe o child en (P o enche and Ca l on 2018); amily’s heal h s a-
us (Bedük 2018 o bo h women and men); he disabili y o any amily membe (e.g.,
Guio and Van den Bosch 2020; Dudek and Szczesny 2021); educa ion (see all abo e-
ci ed pape s, including Ch is ophe e al. 2002); household’s s uc u e (Papadopoulos
and Tsakloglou 2016; Dudek and Szczesny 2021); he p esence and numbe o chil-
d en, especially o single-headed households, is a con e ing ac o (Tsakloglou and
Papadopoulos 2002; Boa ini and d’E cole 2006; Dewilde 2008); home enu e s a us
(Bá cena-Ma ín e al. 2014); coun y o bi h (Buse a e al. 2016); labo ma ke ea-
u es, such as wo k in ensi y (Lay e e al. 2001; Halle öd e al. 2006; de G aa -Zijl and
Nolan 2011;Figa i2012) o employmen ype (see, e.g., Bá cena-Ma ín e al. 2014);
and mac oeconomic ac o s, such as wel a e egimes and labo ma ke ins i u ions
(e.g., Nolan and Whelan 2010;Nelson2012; Alpe e al. 2020).
3 Model speci ica ion
In ou modelling exe cise, we a ge he dep i a ion sco e, de ined as he numbe o
i ems in dep i a ion in a mul i-i em scale. As an icipa ed, ou a ge popula ion is gi en
by singles aged be ween 18 and 64 yea s and no ye e i ed. Ou aim is o in es iga e
whe he gende has a s a is ically signi ican and ele an impac once o he ela ed
obse able cha ac e is ics a e accoun ed o .
In p e ious li e a u e, he dep i a ion sco e has been analyzed ei he using coun
(No en 2016; Bedük 2018) o p opo ional odds (Guio e al. 2012; No en and Guio
2020) eg ession me hods. In he i s case, he sco e is ea ed as an addi i e coun ing
a iable, while in he second, dep i a ion is ead as an indi idual la en ai and
di e en dep i a ion le els a e ea ed as anked ca ego ies ins ead o equidis an
ca ego ies.
In line wi h No en and Guio (2020), we assume a p opo ional odds model. Coun
eg ession models will also be conside ed, and hei abili y o i he da a is compa ed
in e ms o popula model selec ion c i e ia such as he Akaike in o ma ion c i e ion;
hei esul s a e compa ed o hose we ob ain wi h he p opo ional odds as a obus ness
check. Resul s ela ed o pa ial p opo ional odds models (see, o ins ance, Pe e son
and Ha ell J . 1990; Williams 2016) a e no epo ed, since he e idence o hei
supe io i is no suppo ed by B an es esul s (B an 1990) in all coun ies we
123
Ma e ial and social dep i a ion among one-pe son households… Page 5 o 25 10
conside , and would no compensa e he inc eased model complexi y. In any case,
esul s ob ained elaxing he odds p opo ionali y o he gende indica o and possibly
a ew o he selec ed a iables a e discussed below.
The eg ession esul s canno conside sampling weigh s and o he su ey design
aspec s, because published sampling weigh s accoun o he unequal inclusion p ob-
abili ies plus co ec ions aimed a a enua ing he e ec o non- esponse. Howe e ,
we use hese weigh s when compu ing es ima es o popula ion desc ip i e quan i ies
in Sec ion 4. In ha con ex , accoun ing o unequal inclusion p obabili ies is essen-
ial o ob ain consis en esul s. In he eg ession p oblem, he oppo uni y o using
weigh s depends on whe he hey a e in o ma i e, gi en he co a ia es. We es ed his
hypo hesis acco ding o a p ocedu e illus a ed in P e e mann and S e chko (1999)
and Long (2022), which led o he non- ejec ion o he null hypo hesis in all cases.
As a u he check, we also es ima e su ey-weigh ed o dinal eg ession models
using he me hodology illus a ed in Lumley (2011) (chap e 6), ob aining poin es i-
ma es la gely consis en wi h hose ob ained wi hou weigh s. A ull conside a ion o
he sampling design (s a i ica ion, clus e ing, a ying inclusion p obabili y, and pos -
s a i ica ion adjus men ) can ha e an impac on he es ima ion o he s anda d e o s.
Un o una ely, we do no ha e access o all ele an sampling design in o ma ion
(s a um and clus e iden i ie s) because o disclosu e cons ain s.
Le ηibe he la en unobse ed dep i a ion o indi idual iin he sample. We assume
ha he ma e ial and social dep i a ion scale di ides his con inuous a iable in o
K+1=14 in e als by means o (la en ) inc easing cons an s c1≤···≤ck≤···≤
CK, so ha he obse ed dep i a ion sco e Yiwi h suppo {0, ..., K=13}is such
ha
Yi=⎧
⎪
⎨
⎪
⎩
0η≤c1
kc
k≤ηi<ck+1(k=1,...,K−1)
Kηi≥cK
The p opo ional odds eg ession model is based on assuming he cumula i e logi
o be a linea unc ion o he eg esso s. In his case,
log P(Yi≥k)
P(Yi<k)=αk+xT
iβ(1)
Mo eo e , i is speci ic o his model he assump ion ha , while we ha e a dep i a-
ion le el-speci ic in e cep αk, he slopes a e assumed common. This pa simonious
speci ica ion en ails ha eg esso s impac he odds o mo ing om dep i a ion le el
k o k+1in hesameway, ega dlesso k. This somewha es ic i e assump ion
is implied by assuming ha eg esso s ha e an impac on he la en a iable ηionly
h ough i s loca ion (Fulle on and Ande son 2021).
4 Da a and samples
We use da a om he EU-SILC su ey, ha is based on a me hodology and de ini ions
ha a e s anda dized ac oss mos membe s o he Eu opean Union (Eu os a 2010). The
123
10 Page 6 o 25 E. Fab izi e al.
Table 1 I ems in he ma e ial and social dep i a ion indica o (EU-SILC)
# I ems Va iable code
1 Face unexpec ed expenses HS060
2 A o d one week annual holiday away om home HS040
3 A oid a ea s (in mo gage en , u ili y bills and/o hi e pu chase ins almen s) HS011 HS031
4 A o d a meal wi h mea , chicken, ish o ege a ian equi alen e e y second day HS050
5 A o d keeping hei home adequa ely wa m HH050
6 Ha e access o a ca / an o pe sonal use HS110
7 Replace wo n-ou u ni u e HD080
8 Replace wo n-ou clo hes wi h some new ones PD020
9 Ha e wo pai s o p ope ly i ing shoes PD030
10 Spend a small amoun o money each week on him/he sel (“pocke money”) PD070
11 Ha e egula leisu e ac i i ies PD060
12 Ge oge he wi h iends/ amily o a d ink/meal a leas once a mon h PD050
13 Ha e an in e ne connec ion PD080
opics co e ed by he su ey a e li ing condi ions, income, social exclusion, housing,
wo k, demog aphics, and educa ion o indi iduals. We selec c oss-sec ional da a o
six Eu opean coun ies in 2022, co esponding o he income yea 2021.4In de ail,
we se le on Ge many, Spain, F ance, I aly, and Poland, which a e he mos hea ily
popula ed coun ies in he EU, wi h he addi ion o Sweden. We decided o explo e he
phenomenon in hese coun ies as hey a e ep esen a i e o di e en wel a e egimes,
labo ma ke s, and ins i u ions. Nelson (2012), o ins ance, shows ha a meaning ul
pa o he c oss-coun y a ia ion in he le els o ma e ial dep i a ion comes om
di e en social assis ance and bene i s p o ided by go e nmen s. Acco ding o he
2030 Agenda o he Uni ed Na ions, as well as o he p e ious Eu ope 2020 s a egy
o he Eu opean Commission, measu ing non- inancial po e y, and dep i a ion in
pa icula , is e y impo an o moni o ing social exclusion (Guio e al. 2012). Ou
a iable o in e es is he e o e he ma e ial and social dep i a ion sco e (Guio e al.
2016). This sco e is based on 13 selec ed dep i a ions (i ems) de ined ei he a he
indi idual o he household le el. Indi idual i ems come om a ques ionnai e illed
ou by each adul in he household; ques ions on household i ems a e included in he
household ques ionnai e ( illed ou by a e e ence pe son o each household). Each
i em is based on he idea o an en o ced lack, so an indi idual/household is dep i ed
wi h espec o an i em i she/he canno a o d he speci ic good (and no o o he
easons based on p e e ences, see Guio and Van den Bosch 2020) o is no capable
o he speci ic social ac i i y/in e ac ion (Guio e al. 2016,2017). Acco ding o he
Eu os a guidelines, an indi idual is de ined as ma e ially and socially dep i ed i
she/he su e s om a lack o a leas 5 ou o 13 i ems. The ull lis o i ems is p o ided
in Table 1.
4COVID-19 is no an issue o ou analyses o a leas wo easons. Fi s , gi en ha we explo e dep i a ion
and no income, ou main e e ence yea is he su ey yea , i.e., 2022. Second, he negligible impac o he
pandemic on he dep i a ion a e was al eady eco e ed in 2022 (see Eu os a s a is ics).
123
Ma e ial and social dep i a ion among one-pe son households… Page 7 o 25 10
Ou sample includes single adul households, i.e., one-pe son households (single
wi h no child en), in which he adul is 18–64 yea s old and no e i ed. We use he same
age limi o e i emen in each coun y, ha is 64 yea s, i.e., he h eshold conside ed
by Eu os a in i s publica ions. In any case, we exclude e i ed wo ke s e en i aged
less han 64. This sol es he iden i ica ion issue discussed in Sec ion 2, because we a e
able o dis inguish be ween emale and male heads o household, hei esou ces, and
hei en o ced lacks. The selec ion lea es us wi h 19,886 obse a ions o one-pe son
households.
Table 2shows he (weigh ed) ma e ial and social dep i a ion a es es ima ed by
coun y and gende , as well as he gende gap, o ou a ge popula ion compu ed on
he 2022 EU-SILC sample, and o he gene al popula ion aged be ween 18 and 64
(Eu os a ). As o he gende gap, we calcula e - es s o i s s a is ical di e ence, and
when s a is ically signi ican a leas a 5% le el, i is in i alics. In Table 2, i is e iden
ha he gende di e ences a e s a is ically signi ican in he gene al popula ion, wi h
he only excep ion o I aly and Sweden. We ind a disad an age o women, i.e., a
nega i e gende gap in Spain, F ance, and Poland and a posi i e one in Ge many. Wi h
espec o he a ge popula ion, howe e , he gap is signi ican , and nega i e, only in
Spain (−8.2 p.p.).
We selec ed a sample (by coun y) o one-pe son households because his ype o
household is he only one ha enables o eally unde s and i he e is a “genuine” gende
gap in ma e ial and social dep i a ion. This indica o , in ac , is usually calcula ed
by conside ing i ems bo h a he household and a he indi idual le el ( o he lis ,
see Table 1). In he selec ed household ype, consis ing o one pe son only, he e
is no p oblem o sha ing he inabili y o a o d i ems wi h o he household membe s.
S a ing om his conside a ion, we now explo e he aw gende gap (and i s s a is ical
signi icance) in he dep i a ion i ems by coun y. The calcula ions a e epo ed in
Table 3. On he one hand, we see ha he i ems wi h a (signi ican ) gende gap o he
disad an age o women a e (i) acing unexpec ed expenses, (ii) a o ding o keep hei
home adequa ely wa m, (iii) eplacing wo n-ou u ni u e, (i ) pocke money, and ( )
leisu e. On he o he hand, “a ea s” is he i em wi h a ela i ely highe disad an age o
men. By looking a he coun y columns, Spain has he highes numbe o nega i e and
Table 2 Weigh ed ma e ial and
social dep i a ion p e alence (in
pe cen age) by gende in 2022
Ta ge households Gene al popula ion
MF Gap MF Gap
Ge many 17.8 17.1 0.7 16.9 14.1 2.8
Spain 15.8 24 −8.2 14.9 17.8 −2.9
F ance 17 19.2 −2.2 15.2 17.3 −2.1
I aly 11.9 13.6 −1.7 11.1 11.5 −0.4
Poland 12.3 11.7 0.6 11.7 13.8 −2.1
Sweden 7.8 7.7 0.1 6.8 6.6 0.2
No e: Gende gaps among a ge sample and gene al popula ion. I alics
indica e a s a is ical signi ican di e ence in (male– emale) means a
leas a 5% le el. Ta ge households a e one-pe son households, age
18–64. Sou ce: 2022 EU-SILC and Eu os a da a
123
10 Page 8 o 25 E. Fab izi e al.
Table 3 Gende gap in mean dep i a ion o he i ems by coun y, a ge sample
DE ES FR IT PL SE
1. Unexpec ed expenses (gap) −2.20 −6.11 −5.93 −2.25 −6.74 −3.34
%male 42.9 37.8 34.3 37.3 30.7 24.1
% emale 45.1 43.9 40.2 39.5 37.5 27.5
2. One week holidays away 3.33 −5.34 −0.73 −2.30 5.28 −1.99
%male 29.0 32.1 25.5 33.9 25.8 11.0
% emale 25.7 37.5 26.3 36.2 20.5 13.0
3. A ea s 1.80 −0.94 2.94 1.17 3.79 0.75
%male 8.0 13.6 12.4 6.3 13.7 9.7
% emale 6.2 14.6 9.5 5.2 9.9 9.0
4. Meal wi h mea / ish/ ege ables −1.19 −3.46 −4.10 0.87 3.05 −1.85
%male 15.2 6.1 12.9 10.9 9.8 2.8
% emale 16.4 9.6 17.0 10.0 6.8 4.7
5. Wa m home −0.39 −4.78 −6.14 −1.00 3.13 −2.94
%male 8.1 20.0 12.0 14.1 9.1 2.3
% emale 8.5 24.8 18.1 15.1 6.0 5.2
6. Access o a ca 1.35 −5.26 −0.09 −0.05 −1.51 −2.04
%male 15.5 7.1 7.8 4. 6 8.6 8.5
% emale 14.2 12.4 7.8 4.6 10.2 10.6
7. Wo n-ou u ni u e −0.34 −7.74 −4.78 −4.20 2.32 −0.10
%male 21.6 27.1 29.1 19.7 17.9 9.7
% emale 21.9 34.8 33.8 23.8 15.5 9.8
8. Wo n-ou clo hes 1.62 −4.85 −2.75 −2.19 2.30 −0.53
%male 11.9 9.6 11.7 7.2 9.6 5.3
% emale 10.3 14.5 14.5 9.4 7.3 5.8
9. Two pai s o shoes 2.35 −0.41 1.36 0.52 2.12 0.37
%male 5.5 3.6 5.3 4.6 3.3 0.9
% emale 3.2 4.0 4.0 4.1 1.1 0.5
10. Pocke money 0.94 −9.41 −5.87 −3.27 −2.28 −2.29
%male 13.6 13.2 10.8 5.9 9.9 8.8
% emale 12.7 22.6 16.7 9.2 12.2 11.0
11. Leisu e 0.59 −5.10 −4.45 −3.03 −3.89 −3.28
%male 17.2 13.9 16.8 10.8 10.2 7.0
% emale 16.6 19.0 21.3 13.9 14.1 10.3
12. Ge wi h iends/ amily 0.28 −2.74 −0.33 −1.15 1.50 −1.58
%male 10.0 7.4 6.9 5.4 8.8 1.3
% emale 9.7 10.1 7.2 6.6 7.4 2.9
123
Ma e ial and social dep i a ion among one-pe son households… Page 15 o 25 10
Table 6 con inued
DE ES FR IT PL SE
8|9 2.538∗∗∗ 2.640∗∗∗ 2.270∗∗∗ 3.664∗∗∗ 9.969∗∗∗ 3.588∗∗
(0.196) (0.264) (0.364) (0.221) (0.309) (1.410)
9|10 3.244∗∗∗ 3.248∗∗∗ 2.866∗∗∗ 4.175∗∗∗ 10.538∗∗∗ 4.917∗∗∗
(0.204) (0.273) (0.373) (0.230) (0.322) (1.493)
10|11 3.981∗∗∗ 3.931∗∗∗ 3.668∗∗∗ 4.598∗∗∗ 11.169∗∗∗ 17.341∗∗∗
(0.219) (0.292) (0.396) (0.241) (0.346) (1.493)
11|12 5.111∗∗∗ 5.281∗∗∗ 5.282∗∗∗ 5.869∗∗∗ 11.763∗∗∗ 17.834∗∗∗
(0.269) (0.383) (0.538) (0.309) (0.382) (1.493)
12|13 6.810∗∗∗ 6.681∗∗∗ 32.764∗∗∗ 8.136∗∗∗ 13.891∗∗∗ 18.607∗∗∗
(0.485) (0.630) (0.538) (0.737) (0.764) (1.493)
Obse a ions 7104 3230 2634 4270 1600 1048
No e: S anda d e o s in pa en heses. ∗p<0.1; ∗∗ p<0.05; ∗∗∗ p<0.01. Sou ce: au ho s’ es ima ions
based on EU-SILC 2022 da a
in o dep i a ion om ze o, while he gende di e ence a highe le els o dep i a ion
is smalle . In ac , when we use logis ic eg essions o ei he being ma e ially and
socially dep i ed o se e ely dep i ed (i.e., he o icial indica o s o he p obabili y o
being dep i ed i lacking 5 o 7 i ems, espec i ely), we obse e a dec ease in mag-
ni ude and signi icance o he gende e ec (con olling o explana o y a iables and
a cons an e m). Resul s a e epo ed in Table 3 in he Appendix and discussed in
Sec ion 6.6This conclusion is con i med by he applica ion o he pa ial p opo ional
odds model o ou da a, elaxing he p opo ionali y assump ion only o he gende
a iable. Fo all coun ies, he signs o he es ima ed gende coe icien a e posi i e,
bu hei magni ude dec eases as he h eshold inc eases (i.e., he e is a weake and
weake di e ence be ween men and women), and hey u n non-signi ican o high
dep i a ion le els. De ailed esul s abou hese models a e a ailable upon eques . This
e idence is in a o o he “ze o” h eshold p oposed by Bedük (2018).
A possible explana ion o hese esul s esides in he ela i e economic agili y
o women. Table 3in Sec ion 4highligh s a s ong disad an age o women in acing
unexpec ed expenses, e e ywhe e, which is he i em wi h he la ges p e alence o
dep i a ion o women in ou samples ac oss coun ies. Mo eo e , i we conside he
subse o indi iduals wi h exac ly one i em in dep i a ion, in mo e han 50% o he
cases, i has o do wi h he inabili y o ace unexpec ed expenses. Lack o pocke money
and leisu e ac i i ies play a simila ole, ha ecu s e y o en among he dep i a ion
o indi iduals wi h a ew (one o wo) i ems in dep i a ion. We no e ha he a ailabili y
o pocke money and leisu e (social dep i a ions) is ela ed o inancial dis ess, and
hey appea o go a a s ong disad an age o emale heads o household—excep in
Ge many (see also Guio and Van den Bosch 2020). Spanish and F ench women seem
6We also es ima ed p obi models on he wo o icial indica o s. Again, we no e a dec ease in he magni ude
and signi icance o he gende e ec s when mo ing om a h eshold o 5 i ems o 7 i ems. Fo he sake o
b e i y, we do no epo he esul s, which a e a ailable upon eques .
123
10 Page 16 o 25 E. Fab izi e al.
pa icula ly ulne able on hese dimensions, ollowed by I alian women (in line wi h
Mussida e al. 2023).
Financial agili y a ises o se e al easons. Indi iduals may ha e low inancial
li e acy, hey may o e es ima e o unde es ima e pe sonal compe encies, expe ience
unexpec ed la ge income declines, and su e om w ong inancial choices. A is ei
and Gallo (2022), o example, show ha he e is a gende gap in all hese dimensions,
de imen al o women. They a gue ha beha io al and psychological ai s play a
big ole in inancial decision-making; combined wi h social no ms, hey shape he
pe cep ion o gende oles and con ibu e o expose women mo e o inancial and
po e y isks (see also Shole a and Ha is 2019; Zhou and Gan 2023). Lusa di and
Mi chell (2017) show ha women ha e mo e di icul y coping wi h inancial oubles.
Mo eo e , he ac ha women ea n less in hei li e-cycle and ha e low ole ance
owa ds isk (Dawson 2023) dec eases hei sa ing beha io in he sho and in he
long un; his ac on a e age educes hei well-being ela i e o men (Fishe 2010).
In ou esul s, people’s condi ion abou pas ma i al s a us has a posi i e associa ion
wi h he isk o dep i a ion in all coun ies bu I aly and Poland. Ma i al s a us is a
dummy a iable equal o 1 i he indi idual in he pas had been ma ied/cohabi ing
o is widowed; i is ze o o ne e ma ied/cohabi ing people. Ou indings a e in line
wi h England (2002), who explo es he issue by gende . I aly and Poland ep esen an
excep ion in ou sample, no ma e wha “pas li e” indi iduals had, and i does no
a ec he p obabili y o ma e ial and social dep i a ion. The unclea ole o ma i al
dissolu ion on dep i a ion in I aly migh be pa ly due o he p esence o s ong am-
ily/in e gene a ional ies, ha is p oximi y and pa en al suppo (Dalla Zuanna 2001).
As o Poland, i is de ined as one o he EU coun ies wi h he lowes le el o amily
suppo (Szik a and Szelewa 2010).
Al hough ou indings con as wi h Bá cena-Ma ín e al. (2014) in he ac ha ,
o e all, indi idual cha ac e is ics play a mo e impo an ole wi h espec o mac o
indica o s a explaining he p obabili y o dep i a ion, some gende di e ences ac oss
coun ies may a ise because o exis ing gende no ms, a i udes owa ds gende oles,
wel a e egimes, and ans e p og ams—mo e o less a o able o women, and o he
ins i u ional di e ences, in he labo ma ke and ac i e labo ma ke policies. Acco d-
ing oOECD(2023), o example, in 2021, I aly had he highes employmen gende
gap among he six coun ies, and Sweden and F ance had he lowes . Nelson (2012)
inds ha abou 16% a ia ion o ma e ial dep i a ion can be explained by coun y
di e ences in he le els o social bene i s. Social assis ance is di ided in o a sys em o
con ibu o y bene i s (e.g., social insu ance, pa en al lea es, minimum wage schemes,
house bene i s, child bene i s, ax c edi s) and public se ices (e.g., p o ision o ca e
o dependen s). Social assis ance le els a e ai ly low in Eas e n Eu opean coun ies
(such as Poland) ela i ely o o he Eu opean egions; i is ela i ely high in Ge many,
I aly, and Sweden, al hough in he o me wo—plus Spain—la ge egional he e o-
genei y exis s in he implemen a ion o he bene i s. Fu he mo e, he I alian wel a e
model is ge ing a om he Eu opean social sys em, because i is based mo e hea ily
on amily ne wo k suppo (Addabbo e al. 2015).
Ma e ial and social dep i a ion has a non-linea ela ionship wi h age o Ge many,
Spain, F ance, and Sweden. The associa ion we ind is in line wi h he exis ing li e a u e
123
Ma e ial and social dep i a ion among one-pe son households… Page 17 o 25 10
(see, o ins ance, Bá cena-Ma ín and Mo o-Egido 2013, Guio and Van den Bosch
2020; Dudek and Szczesny 2021). I sugges s ha he you h a e pa icula ly exposed
o he isk o ma e ial and social dep i a ion (Whelan and Maî e 2010;Fab izie al.
2023). O e all, bo h labo and non-labo income componen s a e nega i ely associa ed
wi h he isk o ma e ial and social dep i a ion. Howe e , he e ec o he o me is
ela i ely s onge (in magni ude and signi icance, see Table 6) han he la e , which is
e en non-signi ican in I aly. No ably, hese indings should e lec he ela i e highe
impo ance o he labo income sha e o o al income in all coun ies. As a as I aly
is conce ned, he e is a signal o he p esence o non-labo income and/o ans e s,
which should no be e ec i e in educing he isk o dep i a ion (Bonanno e al.
2023). Likewise, ull wo k in ensi y is nega i ely associa ed wi h ma e ial and social
dep i a ion in all samples. Howe e , wo k in ensi y es ima es e eal ha when indi-
iduals wo k less han 20% o he wo kable mon hs in a yea , hey ha e e y low wo k
in ensi y and a e exposed o a highe isk o ma e ial and social dep i a ion, in e e y
coun y (a e con olling o hose who ne e wo ked and o income componen s o
he household income). People wi h a low wage need o wo k a ele an numbe o
hou s, ending up wi h s ong social and inancial cons ain s. Likewise, people wo k-
ing ew hou s/mon hs ace di icul ies o cope wi h expenses o conduc a decen way
o li e. Ou esul s show ha his may be ue in e e y coun y, independen ly on he
wage le el o he gende wage di e en ials (Lay e e al. 2001; Whelan e al. 2004;
Halle öd e al. 2006; de G aa -Zijl and Nolan 2011;Figa i2012). Job unce ain y (such
as ha spu ed by he COVID-19 c isis) and he expec a ions o u u e employmen
and income dec ease household inancial esou ces and inc ease he isk o ma e ial
dep i a ion (C e az 2015; F ied ich and Teichle 2024; Pé ez-Co al e al. 2024). As
ega ds social class, we obse e ha only pe aining o he p o essional occupa ions’
ca ego y is nega i ely associa ed wi h he isk o dep i a ion in he in es iga ed coun-
ies, wi h he excep ion o I aly. This inding is in line wi h Bedük (2018) who a gues
and inds a ole o social class in he isk o dep i a ion (when ze o i ems a e aken
as a h eshold). Non-Eu opean ci izens ha e a highe isk o dep i a ion in Ge many,
I aly, Poland, and sligh ly in F ance (as ega ds he disad an age o o eigne s, see, o
ins ance, Bá cena-Ma ín e al. 2014; Buse a e al. 2016). Te ia y educa ion is nega-
i ely associa ed wi h he p obabili y o being dep i ed in a leas one i em, wi h he
excep ion o Sweden (whe e he es ima e is nega i e and non-signi ican ). This esul
is in line wi h he e idence discussed in he li e a u e sec ion. The e is a e y la ge
consensus among schola s abou he p o ec i e ole o high (seconda y and mainly
e ia y) educa ion. Disabili y, ha is limi a ion in daily ac i i ies, is posi i ely asso-
cia ed wi h he isk o ma e ial dep i a ion in all coun ies. These indings a e s ic ly
linked o he indi ec (long- e m o pe manen ) impac o disabili y and ca ing ac i -
i ies on one’s own o o he household membe s’ labo ma ke pa icipa ion, as ound
by, among o he s, Fab izi and Mussida (2020). Finally, among he indi idual cha ac-
e is ics, ou igh owne ship p o ides a ela i ely (compa ed wi h enan paying o
en , ou base ca ego y) lowe isk o being ma e ially and socially dep i ed, hough
no in Poland.
Among he mac o-le el a iables, we do no ind a clea ole o he deg ee o
u baniza ion. The e is only a signi ican posi i e associa ion o bo h densely and
123
10 Page 18 o 25 E. Fab izi e al.
hinly popula ed a eas (wi h espec o ou base ca ego y, i.e., in e media e) in Poland
and F ance. Dep i a ion is no a mone a y indica o , and he p esence/absence o i ems
should be less s ongly associa ed wi h he deg ee o u baniza ion, since i is a mo e
complex phenomenon. Fo ins ance, social dep i a ion may a ise bo h in a densely
popula ed a ea and in a u al a ea, i people expe ience a educ ion o employmen ,
inancial cons ain s, o a scan y social en i onmen wi hou iends/ amily suppo .
The local—mac o- egional—unemploymen a e does no exe a clea ole on he
isk o ma e ial and social dep i a ion. When signi ican (in Spain and I aly), i has a
posi i e sign. Mo e checks on he ole o he unemploymen a e a e conduc ed in he
nex sec ion.
Finally, as shown in he bo om pa o Table 6, we see ha in e cep s a e in mos
cases equally spaced o close o his condi ion, which co obo a es he good i o ou
models. Ma ked a iable spacing only seldom happens in ou analysis, and when i
does, i is only o e y high le els o dep i a ion which a e in equen in he da a and
o which he abili y o he models o disc imina e is limi ed.
6 Robus ness checks
In his sec ion, we o e some obus ness checks o he indings o ou benchma k
model, in which we change me hods, speci ica ions, and/o samples. The addi ional
analyses can be summa ized as ollows: (1) sepa a ion o he ma e ial and social dep i-
a ion indica o in he “ma e ial” and “social” pa , o unde s and hei con ibu ion o
he o e all indica o ; (2) sepa a e isk eg essions o 13 dep i a ion i ems o inspec
which dep i a ion con ibu es mo e o he gende di e ence; (3) logi o he s anda d
ma e ial dep i a ion indica o and he s anda d se e e ma e ial indica o , i.e., he lack
o 5 o mo e i ems and 7 o mo e i ems, espec i ely, se by Eu os a ; (4) models wi h
in e ac ions be ween gende and educa ion (low/p ima y and high/ e ia y educa ional
a ainmen le el); models es ima ed on sepa a e sub-samples iden i ied by median age;
(5) ( h ee) di e en model speci ica ions on pooled da a. All he a emp s a e epo ed
in Tables 1 o 5 in he Appendix ile. Fo he sake o b e i y, in all hese ables, we
only epo he main coe icien s o in e es .
In he i s check, we sepa a e he “ma e ial and social dep i a ion” indica o in o i s
“ma e ial” and “social” pa . The o me includes nine i ems o he Eu opean s anda d
indica o o ma e ial dep i a ion (HS040, HS060, HS011, HS031, HS050, HH050,
HS110, HS100, HS080, HS070 in he EU-SILC code), while he la e includes he
“new” i ems, i.e., he i ems added o ob ain he “ma e ial and social dep i a ion”
indica o a e 2013 (PD050, PD060, PD070, HD080, PD020, PD030, PD080 in he
EU-SILC code). Table 1 in he Appendix epo s he es ima ed coe icien s o gende
o he ma e ial pa , he social pa , and he o e all ma e ial and social indica o
(which is ou benchma k, see Table 6abo e). The decomposi ion o he o al indica o
in he social and ma e ial componen s sugges s in e es ing e lec ions. Fi s , whe e he
gende was signi ican o he o e all ma e ial and social dep i a ion indica o , i.e., a
ela i ely highe isk o being dep i ed o single women han o men, hesepa a e
123
Ma e ial and social dep i a ion among one-pe son households… Page 19 o 25 10
social and ma e ial pa s a e bo h signi ican . Ge many p o ides a pa ial excep ion,
because gende is signi ican ly associa ed only wi h ma e ial dep i a ion. In I aly,
o which he coe icien associa ed wi h gende is no di e en om ze o o e all,
signi icance a ises o he social componen . The esponsible i ems o explaining he
disad an age o single women in hose componen s a e “unexpec ed expenses” o
Ge many and “leisu e” and “pocke money” o I aly (as s ylized in he desc ip i e
s a is ics, Table 3).
The second check deals wi h he ele ance o each i em. The gende impac is
ela i ely highe when indi iduals a e exposed o ew dep i a ion i ems, i.e., women a e
a maximum o hei disad an age when hey mo e om ze o o one dep i a ion i em.
This obse a ion equi es o un sepa a e eg essions—one o each dep i a ion i em.
The es ima ed gende coe icien s o each i em and coun y a e epo ed in Table 2
o he Appendix. On he one hand, we see ha women a e highly penalized when i
comes o “unexpec ed expenses” (posi i e and signi ican es ima e o all coun ies,
wi h he pa ial excep ion o I aly), “pocke money,” and “leisu e.” On he o he hand,
being emale educes he isk o ha ing “a ea s” and, hough o a lesse ex en , “ wo
pai s o i ing shoes.” In he o me case, i.e., a signi ican disad an age o emales
in “unexpec ed expenses,” “pocke money,” and “leisu e,” we ind a con i ma ion o
he hypo hesis o ela i e inancial agili y o women; in he la e case, we ind a
con i ma ion o he ela i e lowe ole ance o women owa ds isk, as discussed in
Sec ion 5 o he main esul s (Table 6). This exe cise, he e o e, e i ies ha ou main
indings in he benchma k model a e obus . In ac , he he e ogenei y ac oss coun ies
is con i med (I alian and Swedish women appea no o be di e en om hei male
coun e pa s in e ms o “unexpec ed expenses”).
To conduc ou analysis wi hin an o icial amewo k o ma e ial dep i a ion mea-
su es, in he hi d check, we es ima e he p obabili y o ma e ial dep i a ion and se e e
ma e ial dep i a ion ( ha is he lack o 5 i ems and 7 i ems, espec i ely) by using
simple logi models (as men ioned in Sec ion 5). Resul s o he es ima ed gende coe -
icien s a e epo ed in Table 3 in he Appendix. The e is e idence ha he coe icien
associa ed wi h gende in ma e ial dep i a ion logi is s ongly posi i ely signi ican
only o Spain and mode a ely signi ican o Poland. As a as se e e ma e ial dep i-
a ion, he coe icien is posi i e and s a is ically signi ican o Spain and nega i e
and signi ican o Ge many. The es o he es ima es a e non-signi ican . Two con-
clusions om his e idence eme ge: (i) he as majo i y o indi iduals in all coun ies
epo up o one o wo i ems o dep i a ion, wi h only a ew dep i ed wi h espec
o many i ems. Females a e signi ican ly mo e a isk when we conside no h esh-
old, while his e ec disappea s when dicho omizing he dep i a ion s a us based on
high h esholds; (ii) he p obabili y o se e e ma e ial dep i a ion e en e e ses a he
expense o men in Ge many, i.e., he p obabili y o be dep i ed o se en o mo e i ems
becomes highe o men han women (wi h a signi icance a 5% le el). De ailed esul s
abou hese models a e a ailable upon eques . This p o ides e idence in a o o he
“ze o” h eshold p oposed by Bedük (2018), and o he ac ha wha we p opose wi h
a p opo ional odds model b ings o ligh o he wise hidden/insigni ican ela ionships.
As an addi ional check, we also un p obi models and he indings a e in linewi h
123
10 Page 20 o 25 E. Fab izi e al.
hose jus discussed. Fo he sake o b e i y, we do no epo p obi es ima es, which
a e a ailable om he au ho s upon eques .
The ou h check explo es whe he educa ion and age in e ac ed wi h gende lead
o di e en indings o in e p e a ions. Speci ica ion (i) includes in e ac ions be ween
gende and educa ion classes (low/p ima y and high/ e ia y educa ional a ainmen
le el), and speci ica ion (ii) sepa a es samples below and abo e he coun y’s median
age, o es ima e he impac o gende and educa ion. Table 4 in he Appendix epo s
he es ima ed coe icien s o he main a iables o in e es . In model speci ica ion (i),
we obse e ha low-educa ed women a e no di e en om highly educa ed women,
in gene al. In Ge many and Poland, low-educa ed women a e a a bi less isk o
dep i a ion han low-educa ed men. In model speci ica ion (ii), on age-spli samples,
old women appea o be a a signi ican ly highe isk o dep i a ion han old men, excep
in Poland (agains he common in ui ion ha olde women, especially when educa ed,
ace equal isks as men), while young women appea o be a a signi ican ly highe
isk han young men in F ance, Spain, and Poland, e en i e ia y educa ion p o ec s
hem. We also calcula ed - es s o s a is ically signi ican di e ences be ween old
and young: only I aly, Poland, and Sweden ha e sligh ly di e en gende es ima es by
age. These analyses, he e o e, suppo he esul s o ou benchma k model.7
Finally, he las obus ness analysis conce ns pooling he da a. We es ima e h ee
model speci ica ions by pooling coun ies oge he o elabo a e mo e on coun y di -
e ences: (1) pooled da a wi h coun y dummies; (2) pooled da a wi h he in e ac ion
be ween gende and coun y dummies; (3) pooled da a wi h he in e ac ion be ween
gende and he local unemploymen a e. Table 5 in he Appendix epo s he esul s
o he h ee speci ica ions. Gende dispa i ies emain in place, a e con olling o
coun y dummies o he in e ac ions. In pa icula , he coe icien o he main gende
a iable is posi i e and signi ican , i.e., he e exis s a posi i e associa ion be ween
being emale and he isk o ma e ial and social dep i a ion. Howe e , we canno
specula e on coun ies’ di e ences o gende he e. I appea s ha coun y dummies
a e s a is ically signi ican (i.e., di e en om he e e ence ca ego y, Ge many, he
coun y wi h only one dep i a ion i em agains women, “unexpec ed expenses,” see
Table 3). Only Poland shows a ela i ely highe isk o dep i a ion (all o he coun ies
show nega i e signs, ha is a ela i ely lowe isk han Ge many). This should be due,
as we explain in Sec ion 5, o he ac ha he social assis ance le el is ai ly low in
Eas e n Eu opean coun ies ela i ely o o he Eu opean egions. I is ela i ely high
in Ge many, I aly, and Sweden, al hough o he o me wo—plus Spain—la ge he -
e ogenei y exis s in he implemen a ion o he bene i s among egions wi hin coun ies
(Nelson 2012).
Column 2 shows ha he gende gap pe se becomes non-signi ican when in e -
ac ions a e included, because o coun y he e ogenei y. A emale disad an age wi h
espec o Ge man emales can be de ec ed in Spain, F ance, and Sweden (no in
I aly and Poland). Coun y dummies—Ge many as he e e ence ca ego y—gi e e y
7As an addi ional exe cise, we es ima ed models sepa a ely by median age including he in e ac ions
be ween gende and educa ion. As o models (i) and (ii) abo e, we do no ind impo an di e ences ac oss
popula ion sub-g oups. Fo he sake o b e i y, we do no epo hese es ima es he e. Resul s a e a ailable
upon eques .
123
Ma e ial and social dep i a ion among one-pe son households… Page 21 o 25 10
simila esul s as in column 1, again wi h Poland showing a sligh ly highe isk o
dep i a ion.
Column 3 includes he in e ac ion be ween gende and he local unemploymen
a e. While he es ima ed gende coe icien loses signi icance, he in e ac ion wi h he
unemploymen a e is posi i e and signi ican . Coun y dummies gi e simila e idence
as in column 1. The e o e, coun y di e ences emain e en i we adjus o he di icul y
o ob aining jobs, i.e., he local unemploymen a es. The e is a gende gap, and, o
cou se, his is pa icula ly ue o unemployed single women.
O e all, hese addi ional analyses co obo a e he esul s o ou benchma k model,
which seems o be obus o al e na i e assump ions.
7 Conclusions
We in es iga e gende di e en ials in ma e ial and social dep i a ion by ocusing on
single adul heads o household, aged be ween 18 and 64, in six Eu opean coun-
ies. The sample selec ion ci cum en s he p oblem o assessing in a-couple o
wi hin-household inequali y o be e iden i y indi idual command o e esou ces.
Mo eo e , single-pe son households a e on he ise (Ka agiannaki and Bu cha d 2020)
in Eu opean socie ies and a e pa icula ly exposed o he isk o po e y (Chzhen and
B adshaw 2012; T eano 2018). We es ima e p opo ional odds models sepa a ely by
coun y, o cap u e he in ensi y o dep i a ion and he e ogenei y in he gende gap. Ou
indings sugges a clea ole o gende , i.e., he isk o (cumula i e) ma e ial and social
dep i a ion o singles is ela i ely highe o women han o men in all he explo ed
coun ies. The impac o gende is e iden especially a low le els o dep i a ion, as
women a e mo e likely o expe ience dep i a ion in he i ems o inancial dis ess,
such as acing unexpec ed expenses o lack o pocke money, when he dep i a ion
le el is low. The e ec ends o be less e iden i we wo k wi h h esholds like 5 o 7,
ha iden i y mo e se e ely bu clea ly a much smalle se o dep i ed indi iduals. I is
highes when women s ep in o dep i a ion om ze o. In o he wo ds, i is iskie o all
in o dep i a ion o a leas one i em o women wi h espec o men (he e, we do no
in end ha men a e no a isk hemsel es). A highe le els o dep i a ion, he gap is
mo e di icul o iden i y om a s a is ical poin o iew, because o he small numbe
o indi iduals wi h hose le els o dep i a ion. Mo eo e , while i is plausible ha
he COVID-19 pandemic has sp ead inancial conce ns among households, especially
i hei main ea ne s los pa o he /his income, he di icul y o ace “unexpec ed
expenses” and inancial cons ain s a e no exclusi ely ela ed o he pandemic pe iod.
Un o una ely, we do no ha e in o ma ion abou he ex en o unexpec ed expenses
o indi iduals and households (whe he la ge o small).
The o he indi idual cha ac e is ics play a ole in line wi h expec a ions. We ind a
ela i ely less impo an ole o mac o indica o s han indi idual ea u es. None he-
less, he answe o ou esea ch ques ion is ha he gende gap in ma e ial and social
dep i a ion emains in place e e ywhe e, a e con olling o indi idual, mac o a i-
able and a se o obus ness analyses. I we look a single i ems o dep i a ion, he
mos impo an one o women is “ acing unexpec ed expenses,” i.e., he i em showing
123
10 Page 22 o 25 E. Fab izi e al.
he ela i ely highe gap a he disad an age o women. The main explana ion o his
inding s ands in he ela i e inancial and economic agili y o women (e en i con-
olling o p o ec ing ac o s such as e ia y educa ion, ou igh home owne ship,
ull wo k in ensi y, and labo /non-labo income). O e all, ou indings o e impo -
an policy implica ions o educe he agili y o women. Wi h espec o inancial
agili y, in e en ions aimed a inc easing inancial li e acy would be desi able. As
o economic agili y, mo e gene al in e en ions should aim a imp o ing he labo
ma ke condi ions o women, especially educing he pay gap which is s ongly and
posi i ely associa ed wi h he isk o dep i a ion.
Supplemen a y In o ma ion The online e sion con ains supplemen a y ma e ial a ailable a h ps://doi.
o g/10.1007/s00148-025-01084-5.
Acknowledgemen s We hank edi o Klaus F. Zimme mann and wo anonymous e e ees o hei con-
s uc i e commen s. We also hank he pa icipan s a he GLO Be lin 2024 con e ence and he AISSEC
con e ence in Vicenza (I aly) o hei discussion o he main issues.
Au ho con ibu ion All au ho s con ibu ed o he s udy concep ion and design. Ma e ial p epa a ion was
pe o med by En ico Fab izi, Chia a Mussida, and Ma ia Lau a Pa isi, da a collec ion was pe o med by
Chia a Mussida, and da a analysis was pe o med by En ico Fab izi. The i s d a o he manusc ip was
w i en by Chia a Mussida and Ma ia Lau a Pa isi, and all au ho s commen ed on p e ious e sions o he
manusc ip . All au ho s ead and app o ed he inal manusc ip .
Funding Open access unding p o ided by Uni e si à degli S udi di B escia wi hin he CRUI-CARE Ag ee-
men .
Da a a ailabili y The da a ha suppo he indings o his s udy a e a ailable om Eu os a , bu es ic ions
apply o he a ailabili y o hese da a, which we e used unde license o he cu en s udy and so a e no
publicly a ailable. The da a a e, howe e , a ailable om he au ho s upon easonable eques and wi h he
pe mission o Eu os a .
Decla a ions
Con lic o in e es The au ho s decla e no compe ing in e es s.
Open Access This a icle is licensed unde a C ea i e Commons A ibu ion 4.0 In e na ional License, which
pe mi s use, sha ing, adap a ion, dis ibu ion and ep oduc ion in any medium o o ma , as long as you gi e
app op ia e c edi o he o iginal au ho (s) and he sou ce, p o ide a link o he C ea i e Commons licence,
and indica e i changes we e made. The images o o he hi d pa y ma e ial in his a icle a e included
in he a icle’s C ea i e Commons licence, unless indica ed o he wise in a c edi line o he ma e ial. I
ma e ial is no included in he a icle’s C ea i e Commons licence and you in ended use is no pe mi ed
by s a u o y egula ion o exceeds he pe mi ed use, you will need o ob ain pe mission di ec ly om he
copy igh holde . To iew a copy o his licence, isi h p://c ea i ecommons.o g/licenses/by/4.0/.
Re e ences
Addabbo T, Bas os A, Casaca SF, Du u y N, Ní Léime A (2015) Gende and labou in imes o aus e i y:
I eland, I aly and Po ugal in compa a i e pe spec i e. In Labou Re 154(4):449–473
Aisa R, La amona G, Pueyo F (2019) Po e y in Eu ope by gende : he ole o educa ion and labou s a us.
Economic Analysis and Policy 63:24–34
123
Ma e ial and social dep i a ion among one-pe son households… Page 23 o 25 10
Alpe K, Hube E, S ephens JD (2020) Po e y and social igh s among he wo king age popula ion in
pos -indus ial democ acies. Soc Fo ces 99(4):1710–1744
A is ei D, Gallo M (2022) Assessing gende gaps in inancial knowledge and sel -con idence: e idence
om in e na ional da a. Financ Res Le 46
A kinson AB, Guio A-C, Ma lie E (2017) Moni o ing social inclusion in Eu ope. Publica ions O ice o
he Eu opean Union Luxembou g
Becke GS (1991) A ea ise on he amily, Enla ged. Ha a d Uni e si y P ess, Camb idge
Bedük S (2018) Unde s anding ma e ial dep i a ion o 25 EU coun ies: isk and le el pe spec i es, and
dis inc i eness o ze os. Eu Sociol Re 34(2):121–137
Benne F (2013) Resea ching wi hin-household dis ibu ion: o e iew, de elopmen s, deba es, and me hod-
ological challenges. J Ma iage Fam 75(3):582–597
Boa ini R, d’E cole MM (2006) Measu es o ma e ial dep i a ion in OECD coun ies. OECD Social,
Employmen and Mig a ion Wo king Pape s 37, OECD Publishing
Bonanno G, Chies L, Pod ecca E (2023) The de e minan s o po e y exi s and en ies and he ole o social
bene i s: he I alian case. Economia Poli ica 40:533–552
B an R (1990) Assessing p opo ionali y in he p opo ional odds model o o dinal logis ic eg ession.
Biome ics, pages 1171–1178
Bá cena-Ma ín E, Lacomba B, Mo o-Egido AI, Pé ez-Mo eno S (2014) Coun y di e ences in ma e ial
dep i a ion in Eu ope. Re iew o Income and Weal h 60(4):802–820
Bá cena-Ma ín E, Mo o-Egido AI (2013) Gende and po e y isk in Eu ope. Fem Econ 19(2):69–99
Buse a A, Mili o AM, Oli e i AM (2016) The ma e ial dep i a ion o o eigne s: measu emen and de e -
minan s. In: Alle a G, Giommi A (eds) Topics in Theo e ical and Applied S a is ics. Cham. Sp inge
In e na ional Publishing, pp 181–191
Cas ellano R, Rocca A (2020) On he unexplained causes o he gende gap in he labou ma ke . In J Soc
Econ 47(7):933–949
Chan SH (2003) Female household headship and he eminisa ion o po e y: ac s, ic ions and o wa d
s a egies. LSE Resea ch Online, New Wo king Pape Se ies 9
Chan SH (2004) Dange ous equa ions? how emale-headed households became he poo es o he poo :
causes, consequences and cau ions. IDS Bull 35(4):19–26
Ch is ophe K, England P, Smeeding TM, Phillips KR (2002) The gende gap in po e y in mode n na ions:
single mo he hood, he ma ke , and he s a e. Sociol Pe spec 45(3):219–242
Chzhen Y, B adshaw J (2012) Lone pa en s, po e y and policy in he Eu opean Union. J Eu Soc Policy
22(5):487–506
Co si M, Bo i F, D’Ippoli i C (2016) The gende ed na u e o po e y in he EU: indi idualized e sus
collec i e po e y measu es. Fem Econ 22(4):82–100
C e az E (2015) Po e y and ma e ial dep i a ion among Eu opean wo ke s in imes o c isis. In J Soc
Wel 24(4):312–323
Dalla Zuanna G (2001) The banque o Aeolus: a amilis ic in e p e a ion o I aly’s lowes low e ili y.
Demog Res 15(4):131–162
Dawson C (2023) Gende di e ences in op imism, loss a e sion and a i udes owa ds isk. B J Psychol
114(4):928–944
de G aa -Zijl M, Nolan B (2011) Household joblessness and i s impac on po e y and dep i a ion in Eu ope.
J Eu Soc Policy 21(5):413–431
Dewilde C (2008) Indi idual and ins i u ional de e minan s o mul idimensional po e y: a Eu opean com-
pa ison. Social Indica o s Resea ch: An In e na ional and In e disciplina y Jou nal o Quali y-o -Li e
Measu emen 86(2):233–256
Dudek H, Szczesny W (2021) Mul idimensional ma e ial dep i a ion in Poland: a ocus on changes in
2015–2017. Quali y & Quan i y 55(2):741–763
England P (2002) Gende and access o money: wha do ends in ea nings and household po e y ell us?,
pages 131–153. S an o d Uni e si y P ess
Eu os a (2010) Desc ip ion o a ge a iables: c oss-sec ional and longi udinal, eu-silc 065/2010. Eu os a ,
Luxembou g
Fab izi E, Mussida C (2020) Assessing po e y pe sis ence in households wi h child en. The Jou nal o
Economic Inequali y 18(4):551–569
Fab izi E, Mussida C, Pa isi ML (2023) Compa ing ma e ial and social dep i a ion indica o s: iden i ica ion
o dep i ed popula ions. Soc Indic Res 165(3):999–1020
123
10 Page 24 o 25 E. Fab izi e al.
Feng CX (2021) A compa ison o ze o-in la ed and hu dle models o modeling ze o-in la ed coun da a.
Jou nal o s a is ical dis ibu ions and applica ions 8(1):8
Figa i F (2012) C oss-na ional di e ences in de e minan s o mul iple dep i a ion in Eu ope. The Jou nal
o Economic Inequali y 10(3):397–418
Fishe P (2010) Gende di e ences in pe sonal sa ing beha io s. Jou nal o Financial Counseling and
Planning 21
F ied ich M, Teichle N (2024) Do empo a y employees expe ience inc eased ma e ial dep i a ion? E i-
dence om Ge man panel da a. Jou nal o Eu opean Social Policy 0(0):09589287241300011
Fulle on AS, Ande son KF (2021) O de ed eg ession models: a u o ial. P e en ion Science, pages 1–13
Guio A-C, Go don D, Ma lie E (2012) Measu ing ma e ial dep i a ion in he EU: indica o s o he whole
popula ion and child-speci ic indica o s. Eu os a Me hodologies and wo king pape s, Publica ions
o ice o he Eu opean Union, Luxembou g
Guio A-C, Go don D, Naje a H, Poma i M (2017) Re ising he EU ma e ial dep i a ion a iables (anal-
ysis o he inal 2014 EU-SILC da a). Final epo o he Eu os a G an ’Ac ion Plan o EU-SILC
imp o emen s’
Guio A-C, Ma lie E, Go don D, Fahmy E, Nandy S, Poma i M (2016) Imp o ing he measu emen o
ma e ial dep i a ion a he Eu opean Union le el. J Eu Soc Policy 26(3):219–333
Guio A-C, Van den Bosch K (2020) Dep i a ion o women and men li ing in a couple: sha ing o unequal
di ision? Re iew o Income and Weal h 66(4):958–984
Halle öd B, La sson D, Go don D, Ri akallio V-M (2006) Rela i e dep i a ion: a compa a i e analysis o
B i ain, Finland and Sweden. J Eu Soc Policy 16(4):328–345
Is ael S (2016) Mo e han cash: socie al in luences on he isk o ma e ial dep i a ion. Soc Indic Res
129(2):619–637
Ka agiannaki E, Bu cha d T (2020) In a-household inequali y and adul ma e ial dep i a ion in Eu ope.
Cen e o Analysis o Social Exclusion, London School o Economics
Lay e R, Whelan CT, Maî e B, Nolan B (2001) Explaining le els o dep i a ion in he Eu opean Union.
Ac a Sociologica 44(2):105–121
Long J (2022) j ools: analysis and p esen a ion o social scien i ic da a. 2020. R package e sion 2(0)
Lumley T (2011) Complex su eys: a guide o analysis using R. John Wiley & Sons
Lusa di A, Mi chell OS (2017) Olde women’s labo ma ke a achmen , e i emen planning, and household
deb . In: Women wo king longe : Inc eased employmen a olde ages, pages 185–215. Uni e si y o
Chicago P ess
Mu els R, Foua ge D (2004) The ole o Eu opean wel a e s a es in explaining esou ces dep i a ion. Social
Indica o s Resea ch: An In e na ional and In e disciplina y Jou nal o Quali y-o -Li e Measu emen
68(3):299–330
Mussida C, Pa isi ML, Pon a ollo N (2023) Se e i y o ma e ial dep i a ion in Spanish egions and he ole
o he Eu opean S uc u al Funds. Socioecon Plann Sci 88:101651
Nelson K (2012) Coun e ac ing ma e ial dep i a ion: he ole o social assis ance in Eu ope. J Eu Soc
Policy 22(2):148–163
Nolan B, Whelan CT (2010) Using non-mone a y dep i a ion indica o s o analyze po e y and social
exclusion: lessons om Eu ope? J Policy Anal Manage 29(2):305–325
No en G (2016) How po e y indica o s con ound po e y educ ion e alua ions: he a ge ing pe o mance
o income ans e s in Eu ope. Soc Indic Res 127:1039–1056
No en G, Guio A-C (2019) The impac o social ans e s on income po e y and ma e ial dep i a ion. In:
Decen Incomes o All. Imp o ing Policies in Eu ope, pages 85–107. Ox o d Uni e si y P ess
No en G, Guio A-C (2020) A he ma gin : by how much do social ans e s educe ma e ial dep i a ion in
Eu ope ? : 2020 edi ion. Publica ions O ice o he Eu opean Commission and Eu os a
OECD (2023) Economic Policy Re o ms 2023. OECD Publishing, Pa is
Oli e i C, Pe ongolo B (2008) Unequal pay o unequal employmen ? A c oss-coun y analysis o gende
gaps. J Law Econ 26(4):621–654
Papadopoulos F, Tsakloglou P (2016) Ch onic ma e ial dep i a ion and long- e m po e y in Eu ope in he
p e-c isis pe iod. IZA Discussion Pape s 9751, Ins i u e o he S udy o Labo (IZA), Bonn
Pé ez-Co al AL, Bas os A, Casaca SF (2024) Employmen insecu i y and ma e ial dep i a ion in amilies
wi h child en in he pos -g ea ecession pe iod: an analysis o Spain and Po ugal. J Fam Econ Issues
45(2):444–457
Pe e son B, Ha ell FE J (1990) Pa ial p opo ional odds models o o dinal esponse a iables. J Roy S a
Soc: Se C (Appl S a ) 39(2):205–217
123
