A ne ić, Josip; Kike ec, Ma ej; Skok, B animi
A icle
T ends and d i e s o housing a o dabili y in he EU:
Insigh s om panel da a analysis
C oa ian Re iew o Economic, Business and Social S a is ics (CREBSS)
P o ided in Coope a ion wi h:
C oa ian S a is ical Associa ion (CSA), Zag eb
Sugges ed Ci a ion: A ne ić, Josip; Kike ec, Ma ej; Skok, B animi (2024) : T ends and d i e s o
housing a o dabili y in he EU: Insigh s om panel da a analysis, C oa ian Re iew o Economic,
Business and Social S a is ics (CREBSS), ISSN 2459-5616, C oa ian S a is ical Associa ion (CSA),
Zag eb, Vol. 10, Iss. 2, pp. 49-62,
h ps://doi.o g/10.62366/c ebss.2024.2.004
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/323440
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-nc-nd/4.0/
C oa ian Re iew o Economic, Business and Social S a is ics 49
CREBSS 10(2):49–62
T ends and d i e s o housing a o dabili y in he EU: Insigh s om panel
da a analysis
Josip A ne i´
c1, Ma ej Kike ec 1and B animi Skoko 2,*
1Uni e si y o Zag eb, Facul y o Economics and Business, C oa ia
2Uni e si y o Mos a , Facul y o Economics, Mos a , Bosnia and
iiiHe zego ina Q
ARTICLE TYPE
O iginal scien i ic pape
ARTICLE INFO
Recei ed: Sep embe 12, 2024
Accep ed: No embe 12, 2024
DOI: 10.62366/c ebss.2024.2.004
JEL: C23, O18, R21
SUMMARY
Housing a o dabili y is a c ucial issue ha a ec s bo h indi idual
and socie al well–being. A o dable housing ensu es ha households
can mee hei basic li ing needs wi hou expe iencing undue inan-
cial s ess. I in luences labo mobili y, consume spending, economic
g ow h, and esilience. Typically, housing a o dabili y is measu ed by
he p opo ion o household income spen on housing cos s, including
en o mo gage paymen s, u ili ies, and main enance. Howe e , no
single me ic is uni e sally accep ed (cos – o–income a io, esidual
income app oach o subjec i e measu es assessing households’ pe -
cep ions o hei housing a o dabili y). These di e se indica o s e-
lec he complexi y o housing a o dabili y and highligh he need o
comp ehensi e analysis using mul iple me ics, which is he pu pose
o his pape . Panel analysis o he socio–economic and demog aphic
demand and supply d i e s o housing a o dabili y is essen ial o
de eloping e ec i e policies ha ensu e all ci izens ha e access o ad-
equa e and a o dable housing, as many Eu opean Union coun ies
ha e aced a housing a o dabili y c isis cha ac e ized by ising hous-
ing p ices, housing cos s and insu icien housing uni s supply.
KEYWORDS
EU membe s, housing a o dabili y, housing cos s, panel da a
1. In oduc ion
Housing a o dabili y pe ains o he inancial capaci y o households o a o d housing cos s
ela i e o hei income. This concep is conce ned wi h he o e all economic bu den o hous-
ing expenses on households. In con as , a o dable housing e e s o speci ic housing uni s
ha a e p iced a le els deemed a o dable o low o mode a e income households, o en
p o ided h ough public policy ini ia i es, subsidies, o egula ions (Kike ec,2024). This e-
sea ch ocuses exclusi ely on housing a o dabili y, examining he ex en o which households
can a o d o pu chase o en housing. E ec i e housing policies a e essen ial o add essing
housing a o dabili y issues. These policies can include measu es o inc ease he supply o
∗Co esponding au ho
©2024 Copy igh o his a icle is e ained by he au ho (s)
This is an open access a icle unde he CC BY–NC–ND 4.0 license
50 A ne i´c, Kike ec & Skoko
a o dable housing, p o ide inancial assis ance o low income households, and implemen
egula ions o s abilize housing ma ke s wi h espec o housing p ices. Addi ionally, demo-
g aphic policies ha add ess popula ion g ow h, mig a ion, and household o ma ion can
signi ican ly impac housing a o dabili y.
In ecen decades, mos EU coun ies ha e aced a housing a o dabili y c isis, cha ac-
e ized by ising housing cos s and an insu icien supply. Housing p ices and en al a es
ha e inc eased signi ican ly in many Eu opean u ban cen e s due o high demand, limi ed
supply, and specula i e in es men s in eal es a e. Meanwhile, wages and sala ies ha e s ag-
na ed, making i inc easingly di icul o bo h low and middle income households o a o d
adequa e housing (Kike ec,2024). As a esul , many households ha e been o ced in o dis-
placemen , gen i ica ion, and long commu es o hose unable o a o d housing nea hei
wo kplaces. Fo he same eason, a shi owa d en ing a he han owning has been ob-
se ed in many EU coun ies, including Ge many, Sweden, he Ne he lands, and Denma k,
e en hough wo- hi ds o he EU popula ion s ill li e in households owning hei housing
uni (EUROSTAT,2023).
P e ious s udies illus a e dis inc indings based on he a ying measu emen s o hous-
ing a o dabili y, di e en explana o y a iables, and di e se me hodological app oaches.
Some s udies ely on concep ual analysis (Anacke ,2019) o me a–analysis (Lee e al.,2022)
while o he s employ econome ic echniques such as ixed-e ec s modeling (Fili´c,2022) and
spa ial eg ession (Ismail and Wilhelmsson,2024).
This pape con ibu es o he ongoing s udies on housing a o dabili y by add essing
h ee key esea ch ques ions. Fi s , i explo es he heo e ically speci ied d i e s ha shape
housing a o dabili y in he EU. By e iewing he exis ing li e a u e and empi ical indings,
he pape iden i ies he mos impo an ac o s ha in luence housing a o dabili y. Second,
he pape in es iga es whe he hese d i e s con inue o ha e a signi ican impac on housing
a o dabili y when di e en measu es o a o dabili y a e employed wi hin a panel da a anal-
ysis. Acco ding o his objec i e, h ee housing a o dabili y p oxy measu es a e u ilized as
dependen a iables: (a) he sha e o housing cos s o low income households, (b) he hous-
ing cos o e bu den a e o low income households, and (c) he housing cos o e bu den
a e o all households. Th ough obus es ing and model compa ison, he analysis e eals
ha some ac o s e ain hei signi icance ac oss a ious a o dabili y measu es. S anda d
panel da a me hodology is applied wi hin he pooled model, indi idual–speci ic ixed e -
ec s model (FE indi idual), indi idual and ime–speci ic ixed e ec s model (FE wo–ways),
indi idual-speci ic andom e ec s model (RE indi idual), and indi idual and ime–speci ic
andom e ec s model (RE wo–ways). Acco dingly, app op ia e goodness–o – i compa ison
as well as diagnos ic checking was conduc ed.
Finally, he pape examines he implica ions o he indings o imp o ing housing a -
o dabili y policies o he EU membe s a es by analyzing demand d i e s, such as mig a ion,
employmen , u baniza ion, and household size and supply d i e s o housing a o dabili y,
including housing p ices, cons uc ion cos s, and building pe mi s. The empi ical esul s o e
aluable insigh s o he de elopmen o mo e e ec i e, e idence–based housing policies ha
ensu e access o adequa e and a o dable housing o all ci izens, wi h he po en ial o mi -
iga e he a o dabili y challenges aced by bo h low income and middle income households
(Kike ec,2024). These insigh s can enhance he knowledge o policymake s and s akeholde s,
helping o shape in e en ions ha add ess he housing a o dabili y c isis and con ibu e o
u ban planning and u ban de elopmen .
T ends and d i e s o housing a o dabili y in he EU: Insigh s om panel da a analysis 51
2. Theo e ical concep and p e ious s udies e iew
Housing a o dabili y is a complex concep ha is de ined and measu ed in a ious ways in
li e a u e and public policies (Bogdon and Can,1997). Essen ially, i e e s o he abili y o
households o a o d adequa e housing a accep able cos s (S one,2006). Howe e , he e a e
di e en in e p e a ions o his basic concep ega ding wha is conside ed adequa e and a -
o dable housing. The cos – o–income a io is he mos commonly used measu e o housing
a o dabili y in li e a u e (Bogdon and Can,1997). I e e s o he sha e o household income
ha is spen on housing, whe he i is en al cos s, mo gage paymen s, o o he housing
expenses such as u ili ies and main enance (S one,2006). Al hough he e is no uni e sally
accep ed consensus, mos au ho s conside a ios below 30% o indica e a o dable housing,
while a ios abo e 50% poin o excessi e housing cos s and una o dabili y (Jewkes and Del-
gadillo,2010). Eu os a egula ly publishes s a is ics on he sha e o EU households wi h
housing cos s abo e 40% as a measu e o "o e bu den" (EUROSTAT,2023). The main ad an-
age o his measu e is i s simplici y o calcula ion and in e p e a ion. Despi e he c i icism
ha neglec s he absolu e le el o housing cos s and o al household income, e.g. a 40% a io
may ep esen d ama ically di e en absolu e cos s and ma e ial condi ions o a poo and
a weal hy household Lux and Sunega (2020), emains he dominan housing a o dabili y
measu e in academic and policy ci cles (Hsieh and Mo e i,2019).
Unlike he cos – o–income a io which is based on ela i e igu es, absolu e amoun s
and an assessmen o he ma e ial s anda d a household can a o d a e used in he esidual
income app oach, aking in o accoun eal housing cos s. Acco ding o his concep , hous-
ing is a o dable i a e paying all housing cos s ( en , mo gage ins allmen s, u ili ies) he
household is le wi h enough unds o co e o he basic li ing expenses and main ain a
minimally accep able s anda d o li ing (S one,2006). Howe e , i also equi es de e mining
ha "minimum s anda d", which ca ies no ma i e challenges and compa abili y di icul ies.
This measu e also has i s c i ics, bu he ac is ha i mo e ealis ically e lec s households’
inancial si ua ions (Chaplin,1994;Chaplin and F eeman,1999).
Subjec i e measu es o housing a o dabili y a e based on pe cep ions, expe iences and
assessmen s o households hemsel es ega ding he a o dabili y o housing cos s and sa -
is ac ion wi h li ing condi ions, collec ed h ough su eys, e.g. EU–SILC (Eu opean Union
S a is ics on Income and Li ing Condi ions). Subjec i e measu es commonly include: as-
sessmen s o he a o dabili y o cu en housing cos s, pe cei ed inancial s ain o housing
cos s, sa is ac ion wi h apa men size, quali y and ameni ies and sense o housing secu i y
(Heylen,2021). Subjec i e measu es complemen "ha d" s a is ics and p o ide insigh in o
he a o dabili y expe ience om he ci izens’ pe spec i e (Kike ec,2024).
Housing a o dabili y can also be iewed h ough he p ism o access o mo gage lend-
ing, i.e. he abili y o households o ake ou housing loans o pu chase eal es a e (Le man
and Reede ,1987). Since mos households inance he pu chase o an apa men o house
h ough bo owing, lending e ms and c edi wo hiness c ucially a ec he a o dabili y o
homeowne ship. The e o e, a o dabili y measu es based on he sha e o numbe o house-
holds mee ing he condi ions o ob aining mo gage loans wi h " easonable" in e es a es
and epaymen e ms can be ound in he li e a u e (Bogdon and Can,1997). Howe e , " ea-
sonable" lending condi ions a e ela i e and ha dly compa able be ween coun ies. Di e en
de ini ions and measu es o housing a o dabili y lead o di e en assessmen s and policies.
The e o e, i is impo an o analyze hem c i ically.
52 A ne i´c, Kike ec & Skoko
Economic ac o s a e c ucial o de e mining housing a o dabili y and he h ee main
economic ac o s a e eal es a e p ices, household incomes and c edi wo hiness (Hsieh and
Mo e i,2019). Real es a e p ices, whe he o pu chasing o en ing housing, ha e a di ec
impac on a o dabili y. When p ices ise as e han household income g ow h, a o dabil-
i y dec eases. A signi ican d op in eal es a e p ices can empo a ily inc ease a o dabili y,
howe e , i is unsus ainable in he long un wi hou p ice and income s abiliza ion (Kike ec,
2024). The second impo an ac o is household income. Highe a e age incomes allow
la ge housing cos ou lays wi hou comp omising basic li ing needs (Chaplin and F eeman,
1999;Lux and Sunega,2020). Lowe incomes cons ain households’ abili y o a o d housing
cos s. Households in he lowe income deciles a e especially ulne able (Ya es,2008). The
hi d key ac o is access o mo gage lending (Le man and Reede ,1987). Income and c ed-
i wo hiness de e mine households’ abili y o abso b hese cos s (Hancock,1993). The e o e,
i is impe a i e o obse e hem in an in eg a ed manne when designing public policies o
imp o e housing a o dabili y.
Housing a o dabili y is impac ed by a ious demog aphic ac o s pe aining o he size,
composi ion, and li ecycle s age o households. Key de e minan s include household o -
ma ion a es, popula ion g ow h, mig a ion lows, ends in household size and ype, and
age dis ibu ion dynamics (Kike ec,2024). Rising le els o household o ma ion, due o
young adul s mo ing ou o amily homes o pa ne ship b eakdowns, gene a e subs an-
ial demand o a o dable s a e homes (Ya es and Milligan,2007). High popula ion g ow h
a es h ough na u al inc ease o immig a ion also eed in o g ea e housing needs ac oss
all segmen s (Mye s and Ryu,2008;Mye s and Pi kin,2009). Wi hin many coun ies, ends
o declining household sizes, aging popula ions, and g ow h in single–pe son households
u he impac a o dabili y p essu es and policy esponses equi ed. Rapid g ow h in he
numbe o households, whe he due o young adul s se ing up homes o immig an lows,
educes a o dable housing a ailabili y i cons uc ion lags behind. Many coun ies ha e wi -
nessed homeowne ship a es declining among young coho s o e ecen decades, linked o
housing becoming less a o dable o i s - ime buye s on a e age incomes (Cigdem and Whe-
lan,2017). G ea e p i a e en al demand simila ly squeezes a o dabili y o lowe -income
households seeking o en (Ya es and Milligan,2007). S ong popula ion expansion h ough
ele a ed bi hs, ex ended longe i y, o immig a ion he e o e isks ampli ying cons ain s
ac oss mul iple enu e op ions o disad an aged g oups. Ongoing social shi s owa ds
smalle households on a e age, h ough lowe e ili y a es, pa ne ship b eakdowns, aging,
and inc eased li espans spen li ing alone, al e agg ega e housing needs. A la ge num-
be o smalle households inc eases popula ion–adjus ed esiden ial demand and po en ially
hinde s pe –capi a a o dabili y (Mye s and Ryu,2008;Mye s and Pi kin,2009). Housing a -
o dabili y ba ie s a y ac oss age g oups and a e o en mos p oblema ic o hose en e ing
employmen o e i ing. High en s and house p ices hinde labo ma ke lexibili y among
young wo ke s when mo ing jobs in ol es una o dable eloca ion cos s. A la e li e s ages,
declining incomes o e i ees heigh en a o dabili y s esses. Spa ial misma ches be ween he
geog aphical sp ead o housing e sus employmen oppo uni ies also dampen a o dabili y,
especially o younge and lowe income households (Ong e al.,2013).
Housing a o dabili y is also shaped by a ious ins i u ional o ces, pa icula ly ega d-
ing housing supply esponses, subsidy p og ams, and egula o y policies pu sued by go -
e nmen s (Kike ec,2024). When app op ia ely calib a ed, housing policies can imp o e a -
o dabili y ac oss owne ship, p i a e en al, and social en al ma ke segmen s.
T ends and d i e s o housing a o dabili y in he EU: Insigh s om panel da a analysis 53
Howe e , inadequa ely add essing housing and planning sys em cons ain s isks com-
pounding a o dabili y p essu es o e ime. Boos ing he housing supply h ough upzoning
land, unding social housing p ojec s, and add essing cons uc ion sec o ba ie s can mi -
iga e moun ing a o dabili y issues in g owing ci ies (Gu an and Phibbs,2013;Gu an e
al.,2018). Insu icien ma ke - a e housing de elopmen o accommoda e household g ow h
and e ol ing loca ional p e e ences lessens a o dabili y by in ensi ying bidding compe i-
ion o a ailable p ope ies (Glaese and Gyou ko,2018). Cons ain s such as zoning e-
s ic ions, in as uc u e unding gaps, cons uc ion cos s, and ledgling build- o- en sec o s
commonly hinde supply esponses ac oss coun ies and equi e policy e o s a ge ing iden-
i ied blockages (Ba ke ,2004). Go e nmen –p o ided en assis ance paymen s and home
pu chase g an s help ecipien households be e a o d hei exis ing housing. Howe e ,
such demand–side subsidies isk being capi alized in o highe en s and p ices i no cou-
pled wi h ac ions coun e ing supply cons ain s (Fen on,2010). Mo e cons uc ion–linked
subsidies can a oid in la iona y e ec s while aiding ma ginal occupan s (Whi ehead,2007).
S amp du y educ ions and sha ed equi y schemes suppo ing p ospec i e buye s also assis
a o dabili y a a coho le el alongside economy-wide impac s om s imula ing ansac ion
ac i i y (Helde man and Mulde ,2007). Planning egula ions undamen ally shape housing
ma ke ope a ions and p icing signals guiding cons uc ion. U ban con ainmen bounda ies,
densi y con ols, app o al lags, ca pa king manda es, and building code obliga ions a i-
ously in luence de elopmen easibili y and a o dabili y ou comes (Gu an e al.,2018). Re-
o ms s eamlining app o als, allowing g ea e densi ica ion, educing manda o y de elope
con ibu ions and easing codes p o ide scope o imp o e a o dabili y whe e esponsibly im-
plemen ed. Though egula ions aim o enhance ameni y and sus ainabili y, an o e egula ed
sys em hampe s esponsi eness and a o dabili y.
Using a s a is ical model ha cap u es p icing dynamics, Blackwell e al. (2023) ha e
ound ha de egula ion and ma ke –d i en compe i ion ha e no signi ican ly imp o ed a -
o dabili y, especially in high–demand u ban a eas. Dubois and Ni akoski (2023) con ibu ed
o he EU con ex by examining housing a o dabili y ac oss Eu ope based on Eu o ound’s
su ey da a, which includes bo h objec i e indica o s (e.g. cos - o-income a ios) and subjec-
i e measu es (e.g. pe cei ed inancial s ain). The me hodology e eals how inadequa e and
una o dable housing disp opo iona ely a ec s low income households, wi h u ban a eas ex-
pe iencing g ea e a o dabili y challenges due o demand and supply misma ches. Simila ly,
in he EU con ex , Fili´c (2022) employs a panel da a app oach o analyze housing a o dabil-
i y, ocusing on a ious socio–economic and demog aphic d i e s. Explo ing housing ma ke
ola ili y, Engs ed e al. (2016) used ime–se ies da a om he OECD o in es iga e he p es-
ence o specula i e bubbles in housing p ices. They employed coin eg a ion es s and ound
ha house p ices in ad anced economies exhibi ola ile ends d i en by specula ion, which
can lead o a o dabili y c ises as p ices ou pace income g ow h, which highligh s he need
o inancial egula ions o s abilize housing p ices.
3. Resea ch me hodology and empi ical esul s
In his esea ch, he p ice– o–income a io is no used due o i s nume ous d awbacks, as
explained in he p e ious sec ion o he pape , despi e being commonly used in exis ing
empi ical s udies. The e is a lack o pape s add essing housing a o dabili y indica o s o he
han he p ice– o–income a io, and his esea ch aims o ill ha gap (Kike ec,2024).
54 A ne i´c, Kike ec & Skoko
The h ee po en ial indica o s o housing a o dabili y ( he sha e o housing cos s o low
income households, he housing cos o e bu den a e o low income households, and he
housing cos o e bu den a e o all households) exhibi ex emely high posi i e co ela ions
(0.91, 0.92, and 0.93), which jus i ies he eason o al e na e wi h hese p oxy measu es as
dependen a iables (Kike ec,2024). Fu he mo e, his esea ch p o ides a comp ehensi e
panel da a analysis wi h de ailed explana ions o all diagnos ic checks in he pos –es ima ion
phase, an aspec o en igno ed in exis ing s udies employing simila me hodology. Iden i y-
ing he bes i ing panel model is no s aigh o wa d, no is de e mining which a iables
a e mos ele an o educing housing o e bu den o imp o ing a o dabili y.
Table 1.
Desc ip i e s a is ics o h ee po en ial a o dabili y p oxies o e yea s ac oss EU membe s
Sha e o housing cos o Housing cos Housing o e bu den a e
low income households o e bu den a e o low income households
yea mean min max mean min max mean min max
2010 20.30 10.70 33.20 9.04 3.10 21.90 33.68 10.90 71.10
2011 20.36 11.20 32.30 9.51 3.00 24.20 34.59 10.50 78.80
2012 21.24 11.00 37.00 10.30 2.60 33.10 37.07 11.90 90.50
2013 21.32 10.40 39.90 10.49 2.50 36.90 36.99 11.20 93.10
2014 21.15 8.70 42.50 10.55 1.60 44.90 36.86 5.80 93.30
2015 20.75 7.50 42.20 10.17 1.10 45.50 35.61 4.80 94.00
2016 20.15 7.80 41.90 9.67 1.40 40.50 35.36 5.70 91.90
2017 19.63 6.90 41.10 9.24 1.40 39.60 34.94 5.60 89.70
2018 19.12 7.80 40.90 8.60 1.70 39.50 32.92 5.60 90.70
2019 18.54 8.20 38.90 8.25 2.30 36.20 31.98 9.20 88.20
2020 17.59 9.00 36.90 7.24 1.90 33.30 29.09 7.50 83.40
2021 17.57 9.00 34.20 7.15 2.40 28.80 28.67 8.80 76.70
2022 18.37 8.80 34.20 7.89 2.50 26.70 30.98 10.90 84.50
Sou ce: au ho ’s calcula ion in RS udio using da a p o ided by EUROSTAT
(a) Sha e o housing cos s o
low income households
(b) Housing cos o e bu den
a e ( h eshold 40%)
(c) Housing cos o e bu den
a e o low income households
Figu e 1.
A o dabili y p oxy measu es ac oss EU membe s
Be o e conduc ing panel da a analysis, desc ip i e s a is ics o h ee po en ial a o dabil-
i y p oxies o e yea s ac oss EU membe s a e epo ed in Table 1. Minimum and maximum
demons a e ha a o dabili y g ea ly a ies ac oss coun ies, e.g. 84.50% o he G eek pop-
ula ion in 2022 li ed in low income households (below 60% o median equi alised income),
whe e housing cos s ep esen mo e han 40% o disposable income), while he mean in-
dica es upwa d ending o housing cos o e bu den a e. Likewise, i can be concluded
T ends and d i e s o housing a o dabili y in he EU: Insigh s om panel da a analysis 55
ha housing is mo e a o dable in I eland, Cyp us, Mal a, and Li huania due o hei lowe
housing cos s and o e bu den a es (Figu e 1).
Table 2.
Lis o demand and supply d i e s o housing a o dabili y
Va iable Desc ip ion Measu emen uni
PRICE Housing p ice Index (2015=100)
CONSTRUCTION Cons uc ion p oduce p ice Index (2015=100)
SIZE A e age numbe o pe sons pe household Pe sons
PERMITS Building pe mi s Index (2015=100)
URBAN Popula ion li ing in u ban a eas % o popula ion
MIGRATION Ne –mig a ion (immig a ion – emig a ion) % o popula ion
OWNERSHIP Popula ion li ing in owning dwellings % o popula ion
EMPLOYMENT To al employmen a e wi hin age 15–64 % o labo o ce
No e: da a a e p o ided by EUROSTAT public sou ces
Among explana o y a iables in Table 2 (demand and supply d i e s o housing a o d-
abili y) he highes and posi i e co ela ion (0.85) is obse ed be ween house p ice index wi h
espec o pu chasing exis ing o newly buil dwellings and cons uc ion p oduce p ice index
o new esiden ial buildings, which was expec ed. Fo he same eason bo h a iables will
be omi ed om panel analysis due o mul icollinea i y issue and because hese p ices a e
al eady embedded, al hough indi ec ly, in housing a o dabili y indica o s h ough mo gage
o en al paymen s (Kike ec,2024). Va iable "size" which measu es he a e age household
size, will be also omi ed as i is almos ime–in a ian . The e o e, i e a iables will be used:
building pe mi s, deg ee o u baniza ion, ne –mig a ion, owne ship and employmen a e,
while h ee housing a o dabili y p oxies will be swapped in he new panel model speci i-
ca ion (Kike ec,2024). Fo each dependen a iable 5 s a ic panel models a e es ima ed: (1)
pooled model, (2) FE indi idual, (3) FE wo–ways, (4) RE indi idual, and (5) RE wo–ways.
The i s pa o Table 3 p esen s pa ame e es ima es wi h s anda d e o s in pa en hesis, he
second pa p o ides commonly used goodness–o – i measu es (coe icien o de e mina-
ion, adjus ed coe icien o de e mina ion, Akaike In o ma ion C i e ion, Bayes In o ma ion
C i e ion and Roo Mean Squa ed E o ), while he hi d pa exhibi panel diagnos ic es s
(F–s a is ic, B eusch–Pagan s a is ic, Hausman s a is ic, Woold idge and Pesa an CD s a is ic).
When he sha e o housing cos s in disposable income o low income households is
conside ed as he dependen a iable (HA p oxy), a wo–ways ixed e ec s model is mos ap-
p op ia e, indica ing ha building pe mi s, employmen and owne ship educe housing cos s
(and hence imp o es housing a o dabili y), while deg ee o u baniza ion inc eases housing
cos and consequen ly diminishes housing a o dabili y (Kike ec,2024). Fo example, a 1%
inc ease o employmen imp o es housing a o dabili y on a e age by 0.212%. Likewise, a
1% inc ease o building pe mi s imp o es housing a o dabili y on a e age by 0.013%, as-
suming all o he a iables a e cons an . Con a y, housing a o dabili y wo sens by 0.038%
i popula ion li ing in u ban a eas inc eases by 1%. Al hough nega i e, ne –mig a ion is
no s a is ically signi ican a iable. I should be no ed ha only building pe mi s a e aken
in o logs as a iable which is no exp essed in pe cen ages, while o he a iables a e. A bes
i model FE wo–ways is alida ed h ough diagnos ic checking when compa ing o o he
models (Table 3). In ha con ex , he F s a is ic was i s applied o es he signi icance o
indi idual–speci ic e ec s, as well as bo h indi idual and ime e ec s, in a ixed–e ec s (FE)
56 A ne i´c, Kike ec & Skoko
model o de e mine whe he hese e ec s imp o e he model’s i compa ed o a pooled panel
model (a model wi hou indi idual e ec s). Unde he null hypo hesis, i is assumed ha all
indi idual e ec s a e equal o ze o. I he null is ejec ed, hen he FE model is p e e able o
he pooled model. The esul s sugges ha bo h FE indi idual and FE wo–ways p o ide a
be e i compa ed o he pooled model.
Table 3.
Panel models esul s wi h housing cos s sha e o low income households as HA p oxy
Va iable Pooled FE indi idual RE indi idual FE wo–ways RE wo–ways
In e cep 51.184∗∗∗ 57.424∗∗∗ 57.024∗∗∗
(5.796) (5.353) (5.389)
URBAN −0.034 0.028 0.024 0.038∗0.025
(0.025) (0.018) (0.018) (0.018) (0.017)
Log(PERMITS) −0.999 −1.738∗∗∗ −1.734∗∗∗ −1.268∗∗ −1.683∗∗∗
(0.674) (0.397) (0.389) (0.431) (0.392)
MIGRATION −2.361∗∗∗ 0.034 −0.021 −0.255 −0.050
(0.445) (0.220) (0.219) (0.232) (0.219)
EMPLOYMENT −0.145∗∗ −0.272∗∗∗ −0.265∗∗∗ −0.212∗∗∗ −0.264∗∗∗
(0.052) (0.046) (0.044) (0.063) (0.045)
OWNERSHIP −0.184∗∗∗ −0.188∗∗ −0.176∗∗ −0.197∗∗ −0.175∗∗
(0.031) (0.072) (0.058) (0.074) (0.059)
Obse a ions 351 351 351 351 351
R20.200 0.397 0.378 0.189 0.358
R2adj. 0.188 0.338 0.369 0.075 0.349
AIC 2174.3 1408.6 1439.8 1376.5 1435.1
BIC 2201.3 1431.8 1466.9 1399.7 1462.1
RMSE 5.25 1.77 1.84 1.69 1.83
F s a is ic 95.788∗∗∗ 69.894∗∗∗
BP s a is ic 1431.3∗∗∗ 1432.4∗∗∗
Hausman s a is ic 12.34∗∗ 116.29∗∗∗
Woold idge s a is ic 97.472∗∗∗ 112.07∗∗∗ 107.11∗∗∗ 112.33∗∗∗
Pesa an CD s a is ic 4.753∗∗∗ 4.905∗∗∗ −0.149 3.959∗∗∗
No e: signi icance le els a e indica ed as ∗p<0.1, ∗∗ p<0.05, ∗∗∗ p<0.01, while s anda d e o s a e in pa en hesis
The ea e , he B eusch–Pagan s a is ic was used o check he assump ion o cons an
a iance. Speci ically, i he null hypo hesis o ze o a iance in indi idual e ec s is ejec ed,
his implies ha a andom e ec s (RE) panel model is mo e sui able han he pooled model.
Acco dingly, wo B eusch–Pagan s a is ics we e conduc ed: one o he RE indi idual model
and he o he o he RE wo-ways model. In bo h RE models he null hypo hesis was ejec ed,
indica ing ha a andom panel model is mo e adequa e han a pooled model.
The Hausman s a is ic helps in deciding be ween ixed e ec s (FE) and andom e ec s
(RE) models by es ing he null hypo hesis o no co ela ion be ween he indi idual e ec s
(also known as unobse ed indi idual he e ogenei y) and he explana o y a iables, which
is he key assump ion o he RE model. I no di e ence is ound be ween he andom e ec s
and ixed e ec s es ima es, i indica es ha he RE model is consis en and e icien . Howe e ,
ejec ion o he null hypo hesis ypically a o s he FE model, as i sugges s ha he FE model
p o ides unbiased es ima es. Acco dingly, wo Hausman es s we e conduc ed, indica ing
ha bo h ixed e ec s models a e mo e sui able.
