How do income inequali y and iscal consolida ion impac on banking c ises? A
pos -Keynesian iew
Guille mo Peña*
*Assis an P o esso a he Depa men o Economic Analysis o he Uni e sidad de
Za agoza (Sa agossa, Spain), e-mail: [email p o ec ed].
ABSTRACT
This is he i s pape in es ima ing a popula ion-a e aged panel logi p obabili y model
o es he impo ance o he in e ac ion be ween de ici in he public budge ing and
income inequali y on banking c ises, o 36 de eloped coun ies om 1961-2011. New
empi ical e idence is shown on whe he ising inequali y is linked wi h inancial c ises,
co obo a ing heo e ical expec a ions o pos -Keynesian au ho s. Policy measu es a e
p o ided and es ed empi ically: whils in gene al e ms highe le els o income
inequali y could be associa ed wi h inancial c ises; coun ies wi h high le els could
educe he likelihood o a c isis be e in a con ex o iscal consolida ion. One eason
could be ha go e nmen s could use his public su plus o educing income inequali y,
which helps o educe de aul s and banking c ises. These esul s could be use ul o
academics, and policy-make s.
No o wo ds: 9783
Keywo ds: Fiscal Consolida ion, Income Inequali y, Banking C isis, Fiscal Policy
JEL Classi ica ion: I32, H62, G01, E62
Acknowledgemen s: I would like o hank he Edi o s and anonymous e e ees o hei
sugges ions ha ha e imp o ed he pape .
Re iew o Economic Analysis o hcoming 13(2021)
www.Ro EA.o g
1. In oduc ion
Many pape s ha e analyzed he pos -Keynesian iew ha conside s he in luence
o inequali y on inancial c ises (Raghu am Rajan, 2010; Michael Kumho , Romain
Ranciè e, and Pablo Winan , 2015; Ka olin Ki schenmann, Tuomas Malinen, and Hen i
Nybe g, 2016; in e alia). None heless, Sal a o e Mo elli, and An hony B. A kinson
(2015, p.31) s a e ha “ he o e all e idence is a om being conclusi e and he e a e
se e al easons o shed u he ligh on his impo an esea ch opic.” In ac , none o
hem ha e empi ically es ed he impac o he in e ac ion o income inequali y, measu ed
by he Gini coe icien , and public de ici s as a de e minan o banking c ises in a panel
logi model (Peña, 2016, 2017a). As a as known, his is he i s pape in which, applying
a panel logi model, income inequali y and iscal consolida ion a e in e ac ed o s udy his
issue.
The es o he pape is di ided in o six sec ions. Sec ion 2 p o ides he li e a u e
e iew, he de ini ion o banking c ises, he iew o o he schools abou i and he
de e minan s, consequences and policies o a oiding hese c ises. A e ha ing p o ided
o he angles, sec ion 3 de elops he pos -Keynesian hypo hesis o he pape in con as o
o he pe spec i es. Sec ion 4 explains da a and me hodology, using a mul i a ia e logi
panel da a model and a Sys em GMM o check i , whils he empi ical esul s a e p o ided
and discussed in Sec ion 5. The main esul o his sec ion is he co obo a ion o he pos -
Keynesian iew by a signi ican and obus posi i e impac o inequali y on he likelihood
o a banking c isis, join ly wi h he ac ha iscal consolida ion is also bene icial o
a oiding hese c ises. The e is also con ibu ion in policy measu es, s udying he impac
o he in e ac ion o income inequali y and iscal consolida ion on banking c ises. A
nega i e sign o he coe icien o his a iable is ound, which gi es a policy measu e:
es ablishing iscal consolida ion o educe income inequali y when a banking c isis s a s.
Finally, Sec ion 6 p o ides conclusions, inal ema ks and u he esea ch. Addi ionally,
he e is an Appendix wi h o he checks and es ima ions disca ded by non-nes es s.
2. Sys emic banking c ises
Fi s o all, i is equi ed o p o ide a de ini ion o sys emic banking c isis (Dab owski,
Beye s and Villie s, 2016). A banking c isis could be de ined as “ he occu ence o
simul aneous ailu es in he banking sec o ha signi ican ly impai s he capi al o he
banking sys em as a whole, which mos ly esul s in la ge economic e ec s and
go e nmen in e en ion” (Lainà, Nyholm, and Sa lin, 2015), o as “a si ua ion when
mo e han hal o he banking capi al o a leas a conside able sha e o he banking capi al
is los ” (Knu sen and Sjög en, 2009). None heless, he e we p o ide he de ini ion o
Lae en and Valencia (2020) o he upda e o he dependen a iable employed in his
pape . Acco ding o hem, “we de ine a banking c isis as an e en ha mee s wo
condi ions: 1) Signi ican signs o inancial dis ess in he banking sys em (as indica ed
by signi ican bank uns, losses in he banking sys em, and/o banking liquida ions). 2)
Signi ican banking policy in e en ion measu es in esponse o signi ican losses in he
baking sys em. We conside he i s yea ha bo h c i e ia a e me o be he yea when
he c isis became sys emic.”
A e his de ini ion, i is wo h o sepa a e he “banking c isis” om he “ inancial
c isis”. The o me is a kind o he la e . Financial c ises can be di ided in banking c ises,
cu ency c ises and so e eign c ises. The i s wo ones a e usually conside ed win c ises
because hey coincide (Kauko, 2014), as in he No dic (ea ly 1990) and Spanish (la e
Re iew o Economic Analysis o hcoming 13(2021)
www.Ro EA.o g
1970 and ea ly 1980) c ises (Reinha , 2002). None heless, as Bal eanu and E ce (2018)
s a e, he e a e also imes in which banking and so e eign c ises a e combined.
Acco ding o he ype o banking c isis, he e a e a leas wo ypes o banking c ises
ega ding he e idence (Knu sen and Sjög en, 2009): insol ency c ises as he c ises in
se e al de eloped coun ies du ing he 1980s and 1990s wi hou ele an ou b eaks o
bank panic o deposi o uns, wi h widesp ead c edi losses ha lead o high amoun s o
non-pe o ming loans, causing losses o capi al and insol ency. Second, a combina ion
o loan losses and deposi o uns, as he “Asian lu” in 1997 (Kauko, 2014) whe e ailing
mone a y policy was he main explana o y ac o o bank panics and he c isis.
Now some p o o ype coun ies and hei co esponden banking c ises a e going o be
e ised in o de o see he main aspec s ha lead hem o he b un o he banking c ises.
Fi s , i is wo h o highligh he case o he No dic coun ies, wi h he c ises o Sweden
(la e 1870s), No way ( i s hal o 1920) and Finland (1990s), wi h he excep ion o
Sweden. In all coun ies he banking c ises we e mainly d i en by de- egula ion and
libe aliza ion o he banking sys em in excess, encou aging he en ance o new banks
and an excessi e compe i ion ha pu down hei p o i abili y. Some au ho s sus ain ha
highly compe i i e ma ke s lead o a ise isking- aking beha io s (Keeley, 1990), and ha
he monopoly powe o banks is s abilizing (Noy, 2004). Beck e al. (2006) ob ain ha
mo e concen a ed banking sys ems a e less ulne able han a omis ic ones, a e
con olling wi h ins i u ional a iables such as inancial openness and economic eedom.
Fo his case o No dic coun ies, Knu sen and Sjög en (2009) con ibu e wi h an
explana ion, s ablishing ha all banking c ises a e p eceded by an ins i u ional clash as
he de- egula ion o hese coun ies, de ined as “a subs an ial shi o he domina ing
mac oeconomic policy a ached o he inancial sys em”. Excessi e egula ion could also
lead o c ises.
Ga in and Hausmann (1998) p o ide u he examples. Acco ding o hem, p e ious
coun ies also expe ienced an ad e se e ms-o - ade shock be o e o he bank c isis (wi h
he excep ion o Sweden), in addi ion o being he o igin o he cases o Spain (80s, 90s)
and mos La in Ame ican c ises. O he causes we e he exposi ion o ola ili y o he asse
p ices, as p eceden o he ailu e o many banks in A gen ine in he ea ly 1995 by he
Tequila e ec o being exposed o Mexican banks; a long pe iod o exchange- a e
o e alua ion and high in e es s as be o e he Chilean c isis o 1982-1983 o uns able
mac oeconomic and inancial policy ha unchained he 1980 inancial panic in A gen ina.
The e a e o he policies ha a e in ol ed in banking c ises, o ins ance iscal policy.
Rising public expendi u e while he cu ency is o e alued was he ge men o he Tequila
c isis a Mexico in 1995. Expansi e iscal and mone a y policy led o a banking c isis.
This kind o policies can also gene a e so e eign c ises. Bu ega ding banking c ises,
ola ile iscal policies can also hu he banking sys em, gene a ing des abilizing inc eases
in he in e es a es and, h ough he expec a ions o in la ion, also unchaining luc ua ions
in deposi demand, as Ga in and Hausmann (1998) s a e. In addi ion, ax paye s assume
esponsibili y when he c ash is la ge enough o equi e a bailou . This can also a ec o
income dis ibu ion, whe e sa e s can loss hei sa ings and also ha e o pay mo e axes,
in con as o high-income indi iduals ha easily elude he ax sys em. In addi ion, ax
elusion and a oidance could also de e io a e public accoun s and lead he public sec o o
demand mo e public deb , some imes p o ided by banks, and easing possible so e eign
c ises wi h he po en ial banking c isis. In La in Ame ican coun ies he e is plen y o
co up ion, ax a oidance and sca ce economic eedom, which encou ages banking
c ises. In gene al, aud is a oo o c ises since lending can be used o ci cum en
egula ion and he e can be ans e s o deposi s o hold companies ha a e uined, as
occu ed in he Tu kish banking c isis o 2000 (So al, İşcan and Hebb, 2006). Acco ding
Re iew o Economic Analysis o hcoming 13(2021)
www.Ro EA.o g
o Pe ei a e al. (2018), low le el o GDP and high deb in luence on banking c isis. Also
du ing c edi booms he e could be highe income inequali y because jus be o e he
ou b eak o a c isis he economy g ows a a speed highe han he a e age (Reinha ,
1999) and no always in he mos p o-equali y way. In ac , c edi booms p ecede o mos
o he inancial c ises as in A gen ina (1981), Chile (1981-82), U uguay (1982), Colombia
(1982-83), No way (1987), Sweden (1991), Finland (1991-92), Japan (1992) and Mexico
(1995), o ci ing some (Ga in and Hausmann, 1998).
The e a e o he c ises o in e es . Fo ins ance, low in e es a es lead o he Japanese
c isis in he 1990s, o he sub-p ime c isis and he Lehmann-B o he bank up cy, wi h
o igin in he USA, led o he Global Financial C isis (GFC) in he la e 2000s due o
inancial con agion and he consequen G ea Recession (GR) in he ea lies 2010s. Fell e
al. (2018) ha e highligh ed he impo ance on educing he le els o Non-Pe o ming
Loans (NPLs) in Eu opean coun ies ( aised a e he GFC and being cu en ly a an
a e age le el o 3% a Eu ope (EBA; 2020)) in o de o a oid he possible ad e se
impac s on inancial in e media ion. NPLs a e “loans whe e he bo owe - ei he a
company o a physical pe son - is no able o epay a bank loan1.”
The oo s and causes o banking c ises a e se e al. They can be di ided in o policy,
mac oeconomic, banking and ins i u ional o igins. Rega ding he de e minan s ela ed o
policy, Golds ein and Tu ne (1996) highligh he hea y in ol emen o he go e nmen
and he loose o con ols on connec ed lending, Ga in and Hausmann (1996) poin he
ola ile iscal policy as a sou ce o inco ec mone a y policy, as p e iously men ioned,
highligh ing he e ec s o money supply o he in e es a es (F ankel and Sa a elos,
2012). La ge iscal de ici s can p edic c ises while cen al bank independency p e en s
hem (Kauko, 2014). Lae en (2011) ocuses on he ole o go e nmen subsidized
housing policies as a gene a o o eal es a e booms o he po en ial p ocyclicali y o ai
alue accoun ing. Respec o he mac oeconomic ac o , i is wo h o highligh he
mac oeconomic ola ili y, ex e nal and domes ic (Golds ein and Tu ne , 1996), e ms o
ade o wo ld in e es a es (Ga in and Hausmann, 1996), G oss Domes ic P oduc
(GDP), ac o s ega ding cu en accoun , in la ion, capi al lows, ex e nal deb and
inancial openness (F ankel and Sa a elos, 2012).
Re e ing o he banking in luences, lending booms, asse p ice collapses and su ges
in capi al lows a e ele an (Golds ein and Tu ne , 1996), as well as ising bank liabili ies
wi h la ge misma ches in ma u i y o in cu ency (Golds ein and Tu ne , 1996), se ious
c edi c unches, bank le e age, illiquidi y, p esence o in o ma ion p oblems, in e ac ion
be ween bank sol ency and liquidi y, he ine icien paymen mechanism, p o ec ion o
deposi o s, wi h he explici go e nmen -p o ided deposi insu ance, low quali y and
iskiness o he asse s, he ole o deb o s and go e nmen in a bank up cy (Ga in and
Hausmann, 1996), equi y e u ns o deb composi ion (F ankel and Sa a elos, 2012), he
banking sec o - ele an s uc u al ac o s, as he oo-big- o- ail o excessi e banking
compe i ion (Kauko, 2014). Finally, ins i u ional causes could be, as Golds ein and
Tu ne (1996) asse s, he inadequa e p epa a ion o inancial libe aliza ion, he
weakness in he accoun ing, disclosu e and legal amewo k, exchange a e egime o he
dis o ed incen i es o bank owne s, manage s, bank deposi o s and o supe iso s.
Addi ional ins i u ional ac o s a e he absence o an app op ia e amewo k o bank
supe ision and egula ion as in mos cases o La in Ame ica, he limi s o banking
egula ion ega ding cos s, absence o he enough in o ma ion by he au ho i ies o he
need o a anspa en p uden ial egula ion (Ga in and Hausmann, 1996), he de egula ion
and aud as oo o se e al banking ailu es (Lae en, 2011).
1 h ps://ec.eu opa.eu/commission/p essco ne /de ail/bg/MEMO_18_310
Re iew o Economic Analysis o hcoming 13(2021)
www.Ro EA.o g
The main consequences o banking c ises a e p o ided by Lae en (2011). These c ises
usually ha e eal consequences, as dec easing ou pu and employmen , ising public deb ,
sha p dec ease on he con idence in he banking sys em as a whole, losses om o he
(domes ic o o eign) banks ha we e exposed o he bank ails, and losses in asse s. The e
is also an immedia e decline in he low o c edi o he eal economy, unchaining and
ampli ying he nega i e e ec s o he bo owe ’s weal h wi h p esence o asymme ic
in o ma ion. The e a e also income inequali y e ec s as ans e s om axpaye s o banks
when an insol en bank is ecapi alized o om sa e s o c edi o s when he e is in la ion
o cu ency dep ecia ion as a consequence.
Fo p e en ing o sol ing hese c ises, he e a e also ce ain policies. Add essing
eme ging liquidi y loans, liquidi y suppo , ex ensi e gua an ies on bank liabili ies, bank
holidays, egula o y capi al o bea ance, empo a y assump ion o execu i e powe s by
au ho i ies, imposi ion o closes o me ges wi h o he inancial ins i u ions, wo kou s o
dis essed asse s, es uc u e o deb o ecapi aliza ions and na ionaliza ions o banks a e
only some o he possible policies (Lae en, 2011). O he au ho s p o ide addi ional ones:
educing o li ing wi h unce ain y o ola ili y, adequa e mac oeconomic policies,
ese e and a iable capi al equi emen s, supe iso y ools (Golds ein and Tu ne , 1996
and Ra ho e, 2020) o imp o ing he se led Value a Risk (VaR) models used in banking
(E angelos and Sami as, 2020).
3. De elopmen o hypo heses
Se e al economis s conside ha income inequali y is co ela ed wi h inancial
c ises (Rajan, 2010; Robe B. Reich, 2010; Kumho , Ranciè e and Winan , 2015;
Kumho e al., 2012; James K. Galb ai h, 2012; Palley, Thomas I., 2012).2 The eason is
gi en by Till Van T eeck (2014, p.1):
As he bene i s o ising agg ega e income o e he pas decades we e con ined o
a a he small g oup o households a he op o he income dis ibu ion, he consump ion
o he lowe and middle income g oups was la gely inanced h ough ising c edi a he
han ising incomes. This p ocess was acili a ed by go e nmen ac ion.
These p e ious au ho s could be mo e o less ela ed o Pos -Keynesian school.
None heless, o he au ho s ind an insigni ican ela ionship. In ac , o he heo ies o
hinking schools a e no ocused on inequali y, some imes e en a oid in ol ing in his
opic. Rega ding hei iew abou banking c ises, o ins ance, he e is he heo y o he
e icien ma ke s, whe e ma ke s e icien ly e lec he ue p ices o he p oduc s,
showing all he in o ma ion. The mone a is app oach conside s ha banking c ises a e
due o money, while he “business cycle school” links he c ises o he business cycle.
Acco ding o Knu sen and Sjög en (2009), mains eam conside s p oblems ela ed o he
asymme y in in o ma ion combined wi h he p esence o mac oeconomic shocks, so
he e is a mix u e o he p e ious wo iews. Laidle (2009) con as s he mone a is
app oach ha conside s he excess o demand o money ha he cen al bank has o
elie e is a ea u e in banking c ises, while la e Wicksellian adi ion sus ains ha cen al
banke s ha e o a oid he in e empo al discoo dina ion in he alloca ion o esou ces. On
he o he hand, Bianco and Sa doni (2018), shows he iew o an addi ional school, he
New Keynesian, who ocus on he impe ec ions and ic ions in he eal sec o and
inancial ma ke s. In his case, income inequali y could be conside ed as a ic ion in hei
iew.
2 Fo a e iew o he li e a u e abou his link, see Rémi Bazillie and Jé ôme He icou (2017).
Re iew o Economic Analysis o hcoming 13(2021)
www.Ro EA.o g

O he au ho s conside banking c ises p o oke income inequali y, and no he opposi e.
Acco ding o A kinson and Mo elli (2011), nowadays he ela ionship may wo k in he
opposi e di ec ion. On he one hand, p essu es o iscal consolida ion may encou age a
cons an dec ease o he wel a e s a e. On he o he hand, o he au ho s he a oidance o
banking c ises may be necessa y o wa an he sus ainabili y o he social ins i u ions we
ha e de eloped o con ol inequali y, he wel a e s a e and he s abili y o he go e nance.
A kinson and Mo elli (2011) ind ha highe income inequali y seems no o p ecede
banking c ises, bu hey ind mo e e idence on he opposi e: income inequali y is
p eceded by banking c ises. Jespe Roine, Jonas Vlachos, and Daniel Waldens öm
(2009) use da a du ing he pe iod 1900-2000 o 16 coun ies and hey conclude ha a
inancial c isis would educe he op 1 pe cen income sha e by oughly 0.2 pe cen age
poin s o each yea o he c isis. The esul s o Michael D. Bo do, and Ch is ophe M.
Meissne (2012) sugges he e is no signi ican ela ionship be ween inequali y and c edi
booms. Using da a om 14 ad anced coun ies be ween 1920 and 2000, hey ob ain hese
a e no gene al ela ionships. They ind c edi booms heigh en he likelihood o a banking
c isis. Ne e heless, hey ind no e idence ha a ise in op income sha es leads o c edi
booms. Ins ead, he only wo obus de e minan s o c edi booms in hei da a se a e
economic expansions and low in e es a es. Hence, hey do no ind e idence o he
inequali y, c edi , c isis nexus, bu a adi ional boom-bus pa e n o educ ions in in e es
a es, excessi e g ow h, inc easing c edi , asse p ice booms and c ises.
Fi s , his pape will check whe he he e is e idence ha inancial c ises a e
p eceded by income inequali y, measu ed by he Gini index, and iscal consolida ion, in
a sepa a e way. Income inequali y is conside ed because, as Luca Agnello, and Rica do
M. Sousa (2012, p.1425) s a e, “inequali y inc eases be o e banking c isis episodes and
sha ply declines a e wa ds.” Being awa e o he de e minan s o inancial c ises, we
ollow he pos -Keynesian iew o conside ing inequali y as a de e minan o inancial
c ises and o mula e a heo e ical hypo hesis:
H: “Income inequali y and public de ici s inc ease he isk o a banking c isis. A
policy o educing banking c ises would be es ablishing public su plus o educing
income inequali y when he c isis a i es. So, he in e ac ion be ween iscal
consolida ion, and income inequali y lagged wo yea s o e lec ing he adap a ion o
income inequali y hanks o he su plus, would educe he likelihood o a banking c isis.”
This las iew would no be opposed o he New Keynesian iew o he p esence
ic ions, in his case he p oposed policy would seek o elimina ing a ic ion in he eal
sec o , he in e ac ion be ween so e eign de ici and income inequali y. P e ious- o-c isis
su pluses would o se a e -c isis de ici s, main aining he public budge ing balance in
long e m. F om he mone a is and mains eam angle, his in e ac ion can also be seen as
a mac oeconomic shock. Acco ding o he Wicksellian iew, his can be a discoo dina ion
in he alloca ion o he esou ces ha gene a es income inequali y and c ises; bu could
be sol ed by co ec ly employing a so e eign su plus.
In addi ion, he e a e, a leas , wo main easons o suppo ing his hypo hesis. Fi s ,
some au ho s ha e highligh ed he impo ance o he expansion o c edi on banking c ises
(Taylo , 2009; Peña, 2017b) o he impo ance o ax policy (Peña, 2020). None heless,
iscal consolida ion and c edi expansions a e co ela ed (so, in he p oposed econome ic
model, he la e a iable will no appea ), as Ağca and Igan (2019) show: highe iscal
consolida ion u ns in o a cos o c edi , o e all o small businesses. So, iscal
Re iew o Economic Analysis o hcoming 13(2021)
www.Ro EA.o g
consolida ion could be conside ed a ac o o educing banking c ises. On he o he hand,
public de ici s, bu also income inequali y, lead o highe ins abili y, due o empo a y
b eaking he balance budge ing ule. Second, iscal consolida ion has dis ibu ional
e ec s (so, income inequali y will be lagged wo pe iods in he model o clea ly a oid
mul ico ela ion, endogenei y and o show he highe income inequali y p e ious o he
iscal consolida ion) as Ball e al., (2013) show, he e o e, he in e ac ion be ween iscal
consolida ion and income inequali y lagged wo yea s would educe he isk o a banking
c isis.
Respec o he impac o income inequali y on banking c ises, join ly wi h he
ins abili y p o oked by public de ici s, is ele an o highligh ha he highe he income
inequali y in a coun y, he less middle income people and he mos lowe income people
would be, who could s a o all in c edi de aul and encou age o a banking c isis.
As Roy and Kemme (2012, p.274) sugges , “Consis en da a on Gini a ios o o he
simila measu es a e no eadily a ailable” o da ase s as he one o his pape . This may
be a eason o , as a as known, nobody has e e es ima ed a panel logi p obabili y model
explana o y o banking c ises using he Gini index and he public de ici in e ac ed.
4. Da a sample and me hodology
Ou dependen a iable, c isis, is a ailable in he Wo ld Bank and ep esen s a
dummy a iable ha is 1 i he coun y is expe iencing a banking c isis and 0 i no . We
es ima e he equa ions using a popula ion-a e aged panel logi p obabili y model, as
Büyükka abacak and Vale do (2010).
 
*
logi P 1|
i i i
c isis con ol con ol


(1)
The dependen a iable is c isis,
i
con ol
a e he con ol and a ge a iables,
*

he change in he logi o he p opo ion wi h
1c isis 
o a aise in
con ol
o a uni .
Panel da a is used, speci ically, an unbalanced panel om he yea 1961 o 2011
om 36 coun ies, all he EU (27) and some o he OECD coun ies acco ding o Table
1, which gi es some basic in o ma ion abou da a.
Table 1. Coun ies and yea s in he sample
Yea s: 52
Coun ies: 36
1961-2011
Aus alia, Aus ia, Belgium, Bulga ia, Canada, Chile, Czech Republic,
Denma k, Es onia, Finland, F ance, Ge many, G eece, Hunga y, I eland,
Iceland, Is ael, I aly, Japan, Ko ea, Luxembou g, La ia, Mexico, Li huania,
Ne he lands, New Zealand, No way, Poland, Po ugal, Slo ak Republic,
Slo enia, Spain, Sweden, Tu key, Uni ed Kingdom, Uni ed S a es.
Sou ce: By he au ho s
The a iable ha ep esen s income inequali y, measu ed by he Gini coe icien , is
lagged o wo pe iods, in o de o a oid ecip oci y and simul anei y p oblems
(Büyükka abacak and Vale , 2010). The eason is ha banking c ises also impac on
income inequali y (Agnello and Sousa, 2012), as well as iscal consolida ion (Ball e al.,
2013). The sou ce o he da a was he Wo ld Bank Da abase, excep gini2, which comes
om he da abases o Eu os a and OECD, and dis ance, om he GeoDis Da abase
(Thie y Maye and Soledad Zignago, 2011). Table 3 shows each a iable wi h i s
Re iew o Economic Analysis o hcoming 13(2021)
www.Ro EA.o g
espec i e sou ce, and he expec ed sign o he coe icien s o he con ol and a ge
a iables, as well as i s e e ences, based on he a gumen s o Sec ions 1 and 7. The e ec
o su plusgini2 on he dependen a iable will be disco e ed empi ically and discussed
heo e ically in Sec ion 6.
Table 2. Summa y s a is ics.
Va iable
Obse a ions
Mean
S anda d
De ia ion
Minimum
Maximum
c isis
1836
0.083
0.276
0
1
balance
1632
-0.348
6.548
-34.68
32.268
e ms
1527
7.04E+11
6.09E+12
-3.58E+13
6.97E+13
exch
1489
87.256
237.406
0.00001
1909.439
su plus
575
-1.546
4.351
-29.42
20.01
le ne
535
0.185
0.118
-1.609
0.503
dis
1,872
9.844
0.25
9.653
10.670
gini2
428
29.43
10.22
0.283
56.210
Co ela ions among independen a iables used in he model a e shown in Table 4.
Va iables wi h a high co ela ion a e no included in he models and hence, in he ma ix.
Some a iables we e omi ed due o a high co ela ion wi h o he a iables. A oiding
hese a iables, he highes one is he co ela ion be ween dis and gini2, wi h a alue o
0.3. This co ela ion is below 0.5, so i does no seem o be mul icollinea i y p oblems. In
spi e o ha , a VIF es has checked he p esence o his p oblem among hese a iables
and he esul s indica e hese a iables a e a om mul icollinea i y.
Table 3. Expec ed signs o he a iable coe icien s.
Va iable
Sou ce
Sign
Re e ences
balance
Wo ld Bank
+
Rose and Spiegel (2010, 2011) and
S ockhamme (2013)
e ms
Wo ld Bank
+
Ha dy and Paza başioğlu (1999)
exch
Wo ld Bank
-
Ha dy and Paza başioğlu (1999)
su plus
Wo ld Bank
Wo ld Bank
-
Demi guc-Kun and De agiache (1998)
le ne
Wo ld Bank
+
Beck e al. (2006)
dis
Maye and
Zignago (2011)
-
Ye and Han (2010)
gini2
OECD and
Eu os a da abases
+
S ockhamme (2013) and Wisman
(2013)
su plusgini2
su plus*gini2
-
By he au ho s
Sou ce: By he au ho s
The a iables a e desc ibed as ollows. The a iable c isis is he dummy a iable
ha akes he alue 1 i he e is a sys emic banking c isis o 0 o he wise, he a iable
e ms shows he e ms o ade o he coun y de ined as he capaci y o impo less expo s
o goods and se ices, exch e lec s he exchange ade o a coun y, he a iable su plus
is he cash su plus (i i is posi i e) o de ici (i i is nega i e) o he go e nmen o a
coun y, le ne is he deg ee o concen a ion o a banking sys em measu ed by he Le ne
index, dis is he loga i hm o he geog aphical dis ance o a coun y o he coun ies
F ance, he USA and Japan ha a e economic po encies, measu ing he a oidance o he
con agion e ec . Finally, gini2 is he deg ee o income inequali y measu ed by he Gini
Re iew o Economic Analysis o hcoming 13(2021)
www.Ro EA.o g
index and lagged wo yea s in o de o a oid possible co ela ion, simul aneously and
endogenei y p oblems. This a iable is made using he Eu os a sou ce o Eu opean
coun ies and he OECD sou ce o he es .
Table 4. Co ela ion ma ix o independen a iables
balance
e ms
exch
su plus
le ne
dis
gini2
balance
1
e ms
0.011
1
exch
0.132
0.101
1
su plus
0.23
0.107
0.058
1
le ne
-0.169
0.042
0.039
0.197
1
dis
-0.278
-0.044
0.144
-0.243
0.18
1
gini2
-0.011
-0.054
0.218
0.035
0.035
0.305
1
The es ima ion p ocedu e is as ollows. Fi s , a gene al model wi h he main
a iables ega ding policy, mac oeconomic and banking, and ins i u ional ac o s was
de eloped (including mo e han he a iables o Table 2, see he Appendix). None heless,
he e we e oo many a iables o achie ing con e gence in he model and enough
obus ness, so a i s di ision was needed o be made. Th ee kinds o models we e
de eloped, all o hem s a ing wi h he a ge a iables su plus, gini2 and hei
in e ac ion. In addi ion, o each o he h ee models we e chosen policy, mac o and
banking, and ins i u ional a iables, espec i ely, mainly unde s anding policy a iables
as hose ha he public sec o could manage in he sho un by de eloping di ec o
indi ec policies, and ins i u ional ac o s as hose o he long un. Mac oeconomic and
banking a iables a e hose ega ding he speci ic cha ac e is ics o he economy o he
coun y o he banks.
Once his is done o logi models, he nex s ep is o sequen ially elimina e om
he model he a iables wi h less signi ica i i y, un il all he coe icien s ha e a p- alue
o 0.15 o less, ob aining a educed logi model. A e ha , and in o de o a oid po en ial
endogenei y bias, o dynamize and o obus ly check he econome ic speci ica ion o he
p e ious models, Sys em GMM models a e es ima ed. A e his, he same sequence o
a iable-elimina ion is pe o med un il a inal model is ob ained. Nex , a logi model wi h
he same explana o y a iables as his inal model is es ima ed, and he e will be a es o
non-nes ed models by checking he AIC and BIC es s, which a e compa ed be ween hese
models and hei co esponden p e ious educed logi (ob ained by elimina ing non-
signi ican a iables s a ing om he logi model wi h all he a iables o he same ype).
The e o e, a “ inalis ” model is ob ained by choosing he one wi h he lowes AIC and
BIC o each logi models o he same ype o de e minan s o banking c ises.
Following, he e is an es ima ion o a logi model wi h all he a iables o he
educed logi models ( inalis s o no ) in o de o apply he same elimina ing-p ocedu e
and o ob ain a educed logi model based on an ini ial logi model wi h he ull signi ican
explana o y a iables o he o he models. The e is also a GMM Sys em es ima ion by
he same p ocedu e and gi ing he speci ica ion o an addi ional educed logi model. As
p e iously, he educed logi model wi h he lowes AIC and BIC indica o s is chosen as
inalis .
Finally, among he inalis s a de ini i e model will be chosen be ween he ou
inalis models (among he h ee inalis s o each kind o de e minan and he one pa ing
om he ull logi model wi h all he signi ican main a iables) acco ding o hei AIC
and BIC. To he non- educed o m o his de ini i e model (meaning he ini ial logi
Re iew o Economic Analysis o hcoming 13(2021)
www.Ro EA.o g
Table 9. Models wi h ins i u ional ac o s
c isis
(I)
(II)
(III)
(IV)
(V)
lag
0.611***
0.694***
0.000
0.000
lexpec
0.005
-0.037
-0.030**
0.709
0.497
0.064
depend
-0.012
0.022
0.574
0.817
ins q
0.2058
-2.715**
-2.201**
0.394
0.044
0.018
ulelaw
0.354
0.613***
0.430
4.192**
3.703**
0.314
0.002
0.716
0.029
0.041
co up con ol
-0.544***
-0.400**
-1.236
-1.907*
-1.746*
0.000
0.011
0.264
0.087
0.087
publicdeb
0.004
0.005
0.304
0.715
di ec ax
-0.006
0.013
0.590
0.582
su plus
0.072**
0.084**
-0.177
-0.255
-0.293***
0.025
0.012
0.627
0.461
0.000
gini2
0.004
-0.012
0.687
0.745
su plusgini2
-0.002*
-0.003**
-0.001
0.000
0.053
0.022
0.948
0.977
co e
0.981
1.824**
1.905***
0.269
0.010
0.000
No obse a ions
251
254
258
251
412
No g oups
33
34
34
33
35
AICa
271.542
252.6962
336.656
BICa
289.307
295.02
364.803
Sa gan (p- alue)
0.9593
0.999
A-B (p- alue o de
1/2)
0/0.885
0/0.981
No o ins umen s
88
85
R2
0.464
0.2492
Wald
1442.64
1673.97
15.87
71.61
104.56
Wald p- alue
0.000
0.000
0.003
0.000
0.000
Re iew o Economic Analysis o hcoming 13(2021)
www.Ro EA.o g

Table 10. Models wi h inal ac o s among o he models a e he elimina ing-p ocedu e
c isis
(I)
(II)
(III)
(IV)
(V)
lag
0.709***
0.684***
0.000
0.000
lexpec
-0.007
-0.318***
0.708
0.003
ins q
-0.1478346
4.628
-1.174***
0.588
0.201
0.000
ulelaw
0.611***
0.632***
3.508**
4.192779
0.036
0.000
0.025
0.150
co up con ol
-0.397***
-0.444***
-3.659**
-4.688***
0.002
0.000
0.011
0.000
gdppcg
-0.013***
-0.010**
-0.119
-0.206
-0.267**
0.003
0.010
0.238
0.175
0.001
lnc2
-0.003
0.412
0.951
0.116
in la ion
0.027*
0.018*
0.045
0.181
0.070
0.060
0.390
0.222
balance
0.011*
0.010*
0.102**
0.279***
0.063***
0.079
0.059
0.043
0.009
0.002
ex
-0.0002
-0.0005***
-0.011*
-0.001
0.362
0.000
0.054
0.269
su plus
0.094
0.111**
-0.148
0.544
0.105
0.019
0.611
0.242
gini2
0.013
0.112**
0.305
0.033
su plusgini2
-0.002
-0.003*
0.004
-0.020
-0.007***
0.216
0.057
0.645
0.165
0.000
co e
0.593
2.61***
0.962*
0.396
0.000
0.080
No
obse a ions
149
156
156
149
254
No g oups
24
25
25
24
33
AICa
119.107
102.660
223.948
BICa
146.556
144.716
245.172
Sa gan (p-
alue)
0.9113
0.9705
A-B (p- alue
o de 1/2)
0/ 0.6729
0/0.4223
No o
ins umen s
75
75
R2
0.335
0.3117
Wald
1.75E+06
10114.56
14.38
217.23
59.63
Wald p- alue
0.000
0.000
0.000
0.000
0.000
F om he p e ious models, i is wo h o men ion ha he A ellano-Bond (A-B) es has only
been possible o be es ima ed wi h he co esponden model in GMM in di e ences and he R2 is
ob ained om he co esponden s a ic pool O dina y Leas Squa es (OLS) model. In addi ion, i is
well known ha logi models a e one o he bes models in p edic ing c ises. Table 11 p esen s he
non-nes es o he es ima ed models, showing he AIC and BIC o he “ inalis ” models (i.e.: he logi
model o each kind o ac o s, including he inal o comple e models, wi h he lowes indica o s o
each ype o a iables), in o de o show why he chosen model and p esen ed in Sec ion 5 is he
de ini i e. As his Table shows, he Model (III) o Table 5 is chosen because o i s highes explana o y
powe among he o he models. Going u he , Model (V) o Table 5, based on he p e ious model bu
Re iew o Economic Analysis o hcoming 13(2021)
www.Ro EA.o g
wi h he inclusion o addi ional a iables, shows he lowes indica o among all he models (AIC=108,
BIC=140).
Table 10. Summa y o he explana o y powe o he models
Model
5(III)
8(IV)
9(IV)
Table
5
8
9
Fac o s
Policy
Mac o and
banks
Ins i u ional
AIC
119.574
192.535
252.696
BIC
144.519
228.295
295.020
No o explana o y a iables
8
8
11
Model
10(III)
5(III)
5(V)
Table
10
5
5
Fac o s
Mix u e
Chosen model
Augmen ed chosen model (de ini i e)
AIC
102.660
119.574
108.346
BIC
144.716
144.519
139.526
No o explana o y a iables
13
8
10
While Model 10(III) p esen ed lowe AIC han in Model 5(V), he las model was chosen
because o i s lowe BIC and less numbe o explana o y a iables han 10(III) and also han he es
o models.
Finally, in he de ini i e Model (V) o Table 5 he e is an addi ional es in o de o ensu e
ha he e is no e idence o endogenei y. The Du bin-Wu-Hausman (DWH) es 5 is pe o med o check
i . This es consis s on pe o ming an addi ional speci ica ion including he po en ial endogenous
a iables (in his case, o check i su plusgini2 and c isis p esen endogenous beha io ) and he
esiduals o he model, and eg ess hem o ano he a iable as dependen . I he coe icien o he
esidues is signi ican ly dis inc om ze o, hen he e is e idence o endogenei y. The DWH es gi es
he ollowing esul o he Model (V) o Table 5, a Chi o 0.59 and a p- alue o 0.4424, hen he
esiduals a e no signi ican ly dis inc om ze o, and hen his sugges s ha he e is no endogenei y
bias. Fu he mo e, he same Model (V) has been es ima ed bu wi h an ins umen al a iables p obi
model, which compa e he ull sample wi h he po en ial endogenous a iable ( he ins umen ed
a iable) wi h he educed sample wi h he ins umen s, and checks whe he he ins umen ed a iable
was eally endogenous. The lending in e es a e is used as ins umen al a iable because o i s
signi ica i e co ela ion wi h he ins umen ed a iable (su plusgini2), ob aining a Wald es o he
exogenei y o he ins umen ed a iable as null hypo hesis wi h a p- alue o highe han 0.6, accep ing
he null hypo hesis o he exogenei y o su plusgini2.
In addi ion, o he speci ica ions ha e been ied o use, bu due o he insigni ican ly change
o he esul s; hey ha e been a oided o expose. Some examples include models wi h he in e ac ion
be ween income inequali y and g ow h and ax a iables, o he elimina ion o he co e coun ies o
he use o coun ies o he Sou h o Eu ope (Spain, Po ugal, I aly and G eece).
5 h ps://www.s a a.com/suppo / aqs/s a is ics/du bin-wu-hausman- es /
Re iew o Economic Analysis o hcoming 13(2021)
www.Ro EA.o g
Re e ences
Ağca, Ş., & Igan, D. (2019), ‘Fiscal consolida ions and he cos o c edi ’, Jou nal o In e na ional
Economics
(h ps://www.sciencedi ec .com/science/a icle/abs/pii/S0022199619300534).
Agnello, Luca, and Rica do M. Sousa. (2012), ‘How do banking c ises impac on income inequali y?’
Applied Economics Le e s, 19(15), 1425-1429
(h ps://www. and online.com/doi/abs/10.1080/13504851.2011.631885).
A ellano, M. and Bo e , O. (1995), ‘Ano he Look a he Ins umen al Va iable Es ima ion o E o -
Componen s Models’, Jou nal o Econome ics, 68: 29–51.
A kinson, An hony B., and Sal a o e Mo elli. (2011), ‘Economic c ises and Inequali y.’ UNDP-
HDRO Occasional Pape s, (2011/6)
(h ps://pape s.ss n.com/sol3/pape s.c m?abs ac _id=2351471).
Ball, L. M., Fu ce i, D., Leigh, M. D., & Loungani, M. P. (2013), ‘The dis ibu ional e ec s o iscal
consolida ion’ (No. 13-151), In e na ional Mone a y Fund
(h ps://pd s.seman icschola .o g/aa 0/a872d71eb08c97c42b3b1e21d1d759c7ace .pd
).
Bal eanu, I., & E ce, A. (2018), ‘Linking bank c ises and so e eign de aul s: E idence om eme ging
ma ke s’, IMF Economic Re iew, 66(4), 617-664
(h ps://link.sp inge .com/a icle/10.1057/s41308-018-0066-4).
Ba ba, Aldo, and Massimo Pi e i. (2009), ‘Rising household deb : I s causes and mac oeconomic
implica ions—a long-pe iod analysis.’ Camb idge Jou nal o Economics, 33(1), 113-137
(h ps://academic.oup.com/cje/a icle-abs ac /33/1/113/1700026).
Bazillie , Rémi, and Jé ôme He icou , (2017), ‘The ci cula ela ionship be ween inequali y, le e age,
and inancial c ises’, Jou nal o Economic Su eys, 31(2), 463-496
(h ps://onlinelib a y.wiley.com/doi/abs/10.1111/joes.12150).
Beck, T., Demi guc¸-Kun , A., Le ine, R., (2006), ‘Bank concen a ion, compe i ion and c ises: i s
esul s’, J. Bank. Finance 30, 1581–1603.
Beck, Tho s en, Asli Demi güç-Kun , and Ross Le ine. (2006), ‘Bank concen a ion, compe i ion, and
c ises: Fi s esul s.’ Jou nal o Banking & Finance, 30(5), 1581-1603
(h ps://www.sciencedi ec .com/science/a icle/abs/pii/S0378426605001044).
Bianco, A., & Sa doni, C. (2018), ‘Banking heo ies and mac oeconomics’, Jou nal o Pos Keynesian
Economics, 41(2), 165-184.
Blundell, R. & Bond, S. (1998), ‘Ini ial Condi ions and Momen Res ic ions in Dynamic Panel Da a
Models’, Jou nal o Econome ics, 87, 115–43.
Bo do, Michael D., and Ch is ophe M. Meissne . (2012), ‘Does inequali y lead o a inancial c isis?’
Jou nal o In e na ional Money and Finance, 31(8), 2147-2161
(h ps://www.sciencedi ec .com/science/a icle/pii/S0261560612000976).
Bu man, Leona d E., e al. (2010), ‘Ca as ophic budge ailu e.’ Na ional Tax Jou nal, 63(3), 561-
584 (h ps://www. axpolicycen e .o g/si es/de aul / iles/al esco/publica ion-
pd s/1001564-Ca as ophic-Budge -Failu e.PDF).
Büyükka abacak, Be ak, and Ne en T. Vale . (2010), ‘The ole o household and business c edi in
banking c ises.’ Jou nal o Banking & Finance, 34(6), 1247-1256
(h ps://www.sciencedi ec .com/science/a icle/abs/pii/S0378426609003197).
Dab owski, J. J., Beye s, C., & de Villie s, J. P. (2016), ‘Sys emic banking c isis ea ly wa ning sys ems
using dynamic Bayesian ne wo ks’, Expe sys ems wi h applica ions, 62, 225-242.
Demi güç-Kun , Asli, and En ica De agiache. (1998), ‘The De e minan s o Banking C ises in
De eloping and De eloped Coun ies.’ IMF Economic Re iew, 45(1), 81-109
(h ps://link.sp inge .com/a icle/10.2307/3867330).
Eu opean Banking Au ho i y (2020), Risk Dashboa d Da a As O Q2 2020.
E angelos, V. and Sami as, A. (2020), ‘Value a Risk, Legisla i e F amewo k, C ises, and
P ocyclicali y: wha goes w ong?’, Re iew o Economic Analysis, o hcoming.
Fell, J. P., G odzicki, M., Me zle , J., & O'B ien, E. (2018). Non-pe o ming loans and eu o a ea bank
lending beha iou a e he c isis. Re is a de es abilidad inancie a. Nº 35 (no iemb e 2018),
p. 7-28
Re iew o Economic Analysis o hcoming 13(2021)
www.Ro EA.o g
(h ps:// eposi o io.bde.es/bi s eam/123456789/11241/1/Non_pe o ming_loans_es
panol.pd ).
Fe ei o, Jesús, Ca alina Gál ez, Ca men Gómez, and Ana González. (2016), ‘Bank escues and iscal
policy in he Eu opean Union du ing he g ea ecession.’ Panoeconomicus, 63(2), 211-230
(h p://panoeconomicus.o g/index.php/jo unal/a icle/ iew/384).
F ankel, J., & Sa a elos, G. (2012). Can leading indica o s assess coun y ulne abili y? E idence
om he 2008–09 global inancial c isis. Jou nal o In e na ional Economics, 87(2), 216-231.
Galb ai h, James K. (2012), ‘Inequali y and Ins abili y: A S udy o he Wo ld Economy Jus Be o e
he G ea C isis.’ Ox o d Uni e si y P ess.
Ga in, Michael, and Rica do Hausmann. (1996), ‘The Roo s o Banking C ises: The Mac oeconomic
Con ex .’ In Hausmann, Rica do, and Liliana Rojas-Suá ez (Ed.), Banking c ises in La in
Ame ica, In e -Ame ican De elopmen Bank, Washing on, D.C., 25-63.
Golds ein, M., & Tu ne , P. (1996), ‘Banking c ises in eme ging economies: o igins and policy
op ions’, T ade Cu encies and Finance, 301-363.
Ha dy, Daniel C., and Ceyla Paza başioğlu. (1999), ‘De e minan s and leading indica o s o banking
c ises: Fu he e idence.’ IMF Economic Re iew, 46(3), 247-258
(h ps://link.sp inge .com/a icle/10.2307/3867642).
Kauko, K. (2014). How o o esee banking c ises? A su ey o he empi ical li e a u e. Economic
Sys ems, 38(3), 289-308.
Keeley, M.C. (1990), ‘Deposi insu ance, isk and ma ke powe in banking’, Am. Econ. Re . 80,
1183–1200 .
Ki schenmann, Ka olin, Tuomas Malinen, and Hen i Nybe g. (2016), ‘The isk o inancial c ises: Is
he e a ole o income inequali y?’ Jou nal o In e na ional Money and Finance, 68, 161-
180 (h ps://www.sciencedi ec .com/science/a icle/pii/S0261560616300894).
Knu sen, S. and Sjög en, H. (2009), ‘Ins i u ional Clash and Financial F agili y. An E olu iona y
Model o Banking C ises’, MPRA Pape No. 13133, 5, Janua y 2009
(h p://mp a.ub.uni-muenchen.de/13133/).
Kumho , Michael, Clai e Leba z, Romain Rancie e, Alexande W. Rich e and Na haniel A.
Th ockmo on. (2012), ‘Income inequali y and cu en accoun imbalances’, IMF Wo king
Pape WP/12/8
(h ps://na h ockmo on.people.wm.edu/Pape s/KLRRT_Inequali y.pd ).
Kumho , Michael, Romain Ranciè e, and Pablo Winan . (2015), ‘Inequali y, le e age, and c ises.’
The Ame ican Economic Re iew, 105(3), 1217-1245
(h ps://www.aeaweb.o g/a icles?id=10.1257/ae .20110683).
Lae en, L. (2011), ‘Banking c ises: A e iew’, Annu. Re . Financ. Econ., 3(1), 17-40.
Lae en, L., & Valencia, F. (2020). Sys emic Banking C ises Da abase II. IMF Economic Re iew, 1-
55.
Laidle , D. (2009), ‘Financial s abili y, mone a ism and he Wicksell connec ion’, Re iew o Economic
Analysis, 1(1), 60-79
(h ps://openjou nals.uwa e loo.ca/index.php/ o ea/a icle/ iew/1479).
Lainà, P., Nyholm, J., & Sa lin, P. (2015), ‘Leading indica o s o sys emic banking c ises: Finland in
a panel o {EU} coun ies’, Re iew o Financial Economics, 24, 18–35.
(h p://www.sciencedi ec .com/science/a icle/pii/S1058330014000512).
Maye , Thie y, and Soledad Zignago. (2011), ‘No es on CEPII’s dis ances measu es: he GeoDis
Da abase’, CEPII Wo king Pape 2011-25
(h ps://pape s.ss n.com/sol3/pape s.c m?abs ac _id=1994531).
Mo elli, Sal a o e, and An hony B. A kinson. (2015), ‘Inequali y and c ises e isi ed’. Economia
Poli ica, 32(1), 31-51 (h ps://link.sp inge .com/a icle/10.1007/s40888-015-0006-y).
Neuhaus, John M., John D. Kalb leisch, and Wal e W. Hauck. (1991), ‘A compa ison o clus e -
speci ic and popula ion a e aged app oaches o analyzing co ela ed bina y da a.’
In e na ional S a is ical Re iew 59(1), 25–35 (h ps://www.js o .o g/s able/1403572).
Noy, I., (2004), ‘Financial libe aliza ion, p uden ial supe ision and he onse o banking c ises’,
Eme g. Ma k. Re . 5, 341–359.
Palley, Thomas I. (2012), ‘F om Financial C isis To S agna ion: The Des uc ion o Sha ed P ospe i y
and he Role o Economics.’ Camb idge Uni e si y P ess.
Re iew o Economic Analysis o hcoming 13(2021)
www.Ro EA.o g
Peña, Guille mo (2016). The de e minan s o banking c ises: Fu he e idence, MPRA WP 70093
(h ps://mp a.ub.uni-muenchen.de/70093/).
Peña, Guille mo (2017a): ‘Income inequali y, iscal consolida ion and banking c ises’. In he XXIV
Encuen o de Economía Pública. Uni e sidad de Cas illa-La Mancha, 2017. p. 58
(h ps://dialne .uni ioja.es/se le /a iculo?codigo=6121526).
Peña, Guille mo (2017b): ‘Money, lending and banking c ises’, Economic Pape s: A jou nal o
applied economics and policy, 36(4), 444-458
(h ps://onlinelib a y.wiley.com/doi/abs/10.1111/1759-3441.12195).
Peña, Guille mo (2020): ‘VAT expe ience as an in luence on banking c ises’, Na ional Accoun ing
Re iew, 2(1): 53-65
(h ps://www.aimsp ess.com/a icle/10.3934/NAR.2020003/Rela ed.h ml).
Pe ei a Ped o, C., Ramalho, J. J., and Vidigal da Sil a, J. (2018), ‘The main de e minan s o banking
c ises in OECD coun ies’, Re iew o Wo ld Economics, 154(1), 203-227
(h ps://link.sp inge .com/a icle/10.1007/s10290-017-0294-0).
Rajan, Raghu am. (2010), ‘Faul Lines.’ P ince on Uni e si y P ess, P ince on, NJ.
Ra ho e, A. S. (2020), ‘EBA's Capi al Exe cise and Technical E iciency o he Banks’, Re iew o
Economic Analysis, 12(1), o hcoming
(h ps://openjou nals.uwa e loo.ca/index.php/ o ea/a icle/ iew/1774).
Reich, Robe B. (2010), ‘A e shock: The Nex Economy and Ame ica's Fu u e.’ Al ed A. Knop ,
New Yo k.
Reinha , C. M. (2002), ‘De aul , cu ency c ises, and so e eign c edi a ings. he wo ld bank
economic e iew’, 16(2), 151-170.
Roine, Jespe , Jonas Vlachos, and Daniel Waldens öm (2009), ‘The long- un de e minan s o
inequali y: Wha can we lea n om op income da a?’ Jou nal o Public Economics, 93(7-8),
974-988
(h ps://www.sciencedi ec .com/science/a icle/abs/pii/S0047272709000383).
Rose, And ew K., and Ma k M. Spiegel (2010), ‘C oss‐Coun y Causes and Consequences o he 2008
C isis: In e na ional Linkages and Ame ican Exposu e.’ Paci ic Economic Re iew, 15(3),
340-363 (h ps://onlinelib a y.wiley.com/doi/abs/10.1111/j.1468-0106.2010.00507.x).
Rose, And ew K., and Ma k M. Spiegel (2011), ‘C oss-coun y causes and consequences o he c isis:
An upda e’, Eu opean Economic Re iew, 55(3), 309-324
(h ps://www.sciencedi ec .com/science/a icle/pii/S0014292110001285).
Roy, Sak inil, and Da id M. Kemme (2012), ‘Causes o banking c ises: De egula ion, c edi booms
and asse bubbles, hen and now.’ In e na ional Re iew o Economics & Finance, 24, 270-
294 (h ps://www.sciencedi ec .com/science/a icle/pii/S1059056012000299).
So al, H. B., İşcan, T. B., & Hebb, G. (2006). F aud, banking c isis, and egula o y en o cemen :
E idence om mic o-le el ansac ions da a. Eu opean Jou nal o Law and Economics, 21(2),
179-197.
S ockhamme , Engelbe (2013), ‘Rising inequali y as a cause o he p esen c isis.’ Camb idge
Jou nal o Economics, 39(3), 935-958 (h ps://academic.oup.com/cje/a icle-
abs ac /39/3/935/1713529).
Van T eeck, Till (2014), ‘Did inequali y cause he US inancial c isis?’ Jou nal o Economic Su eys,
28(3), 421-448 (h ps://onlinelib a y.wiley.com/doi/abs/10.1111/joes.12028).
Wisman, Jon D (2013), ‘Wage s agna ion, ising inequali y and he inancial c isis o 2008.’
Camb idge Jou nal o Economics, 37(4), 921-945 (h ps://academic.oup.com/cje/a icle-
abs ac /37/4/921/1712745).
Woold idge, Je ey M (2002), ‘Econome ic Analysis o C oss Sec ion and Panel Da a.’ The MIT
P ess, Camb idge.
Ye, Qing, and Liyan Han (2010), ‘The in e na ional p opaga ion o shocks in in e na ional equi y
ma ke s du ing he subp ime mo gage c isis,’ In e na ional Con e ence on Fu u e
In o ma ion Technology and Managemen Enginee ing (FITME), IEEE Oc obe (2), 468-471.
Zege , Sco L., Kung-Yee Liang, and Paul S. Albe (1988), ‘Models o longi udinal da a: a
gene alized es ima ing equa ion app oach.’ Biome ics, 44(4), 1049–1060
(h ps://www.js o .o g/s able/2531734?seq=1).
Re iew o Economic Analysis o hcoming 13(2021)
www.Ro EA.o g