Le Khuong Ninh
A icle
Sel -selec ion ou o o mal c edi ma ke s: E idence om
u al Vie nam
Asian Jou nal o Economics and Banking (AJEB)
P o ided in Coope a ion wi h:
Ho Chi Minh Uni e si y o Banking (HUB), Ho Chi Minh Ci y
Sugges ed Ci a ion: Le Khuong Ninh (2025) : Sel -selec ion ou o o mal c edi ma ke s: E idence
om u al Vie nam, Asian Jou nal o Economics and Banking (AJEB), ISSN 2633-7991, Eme ald,
Leeds, Vol. 9, Iss. 1, pp. 105-125,
h ps://doi.o g/10.1108/AJEB-02-2023-0011
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/334139
S anda d-Nu zungsbedingungen:
Die Dokumen e au EconS o dü en zu eigenen wissenscha lichen
Zwecken und zum P i a geb auch gespeiche und kopie we den.
Sie dü en die Dokumen e nich ü ö en liche ode komme zielle
Zwecke e iel äl igen, ö en lich auss ellen, ö en lich zugänglich
machen, e eiben ode ande wei ig nu zen.
So e n die Ve asse die Dokumen e un e Open-Con en -Lizenzen
(insbesonde e CC-Lizenzen) zu Ve ügung ges ell haben soll en,
gel en abweichend on diesen Nu zungsbedingungen die in de do
genann en Lizenz gewäh en Nu zungs ech e.
Te ms o use:
Documen s in EconS o may be sa ed and copied o you pe sonal
and schola ly pu poses.
You a e no o copy documen s o public o comme cial pu poses, o
exhibi he documen s publicly, o make hem publicly a ailable on he
in e ne , o o dis ibu e o o he wise use he documen s in public.
I he documen s ha e been made a ailable unde an Open Con en
Licence (especially C ea i e Commons Licences), you may exe cise
u he usage igh s as speci ied in he indica ed licence.
h ps://c ea i ecommons.o g/licenses/by/4.0/
Sel -selec ion ou o o mal c edi
ma ke s: e idence om
u al Vie nam
Le Khuong Ninh
Can Tho Uni e si y, Can Tho, Vie nam
Abs ac
Pu pose –This pape examines why a me s sel -selec ou o o mal c edi ma ke s e en hough hey need
ex e nal unds.
Design/me hodology/app oach –We use p obi and Bayesian p obi es ima o s o de ec he de e minan s
o sel -selec ion beha io based on a p ima y da ase o 2,212 ice a me s in Vie nam. A e ha , we use he
mul inomial p obi (MNP) and Bayesian MNP es ima o s o e eal he impac o ele an ac o s on he decision
o sel -selec o a me s belonging o each sel -selec ion ca ego y.
Findings –The p obi and Bayesian p obi es ima o s show ha he decision o sel -selec depends on
household head age, income pe capi a, a m size, whe he o no o ha e ela i es o iends wo king o banks,
he numbe o p e ious bo owings, isks ela ed o na u al disas e s, diseases, and ice p ice, and he numbe o
banks wi h which he a me has ela ionships. The MNP and Bayesian MNP es ima o s gi e u he insigh s
in o he decision o a me s o sel -selec in ha de e minan s o he sel -selec ion beha io depend on he easons
o sel -selec . In conc e e, a m size and he numbe o p e ious bo owings mi iga e he sel -selec ion o a me s
who did no apply o loans due o ha ing access o o he p e e ed sou ces o c edi . The sel -selec ion o a me s
no applying o loans because o un a o able loan e ms is condi ional on household head age, a ming
expe ience, income, a m size, he numbe o p e ious bo owings, na u al disas e isk, and he numbe o banks
he a me has ela ionships wi h. Se e al ac o s, including educa ion, income, he dis ance o he nea es bank,
whe he o no ha ing ela i es o iends wo king o banks, he numbe o p e ious bo owings, isks, and he
numbe o banks he a me has ela ionships wi h, a ec he sel -selec ion o a me s no applying o loans
because o high bo owing cos s. The sel -selec ion o a me s no applying o loans because o complex
applica ion p ocedu es depends on income and he numbe o p e ious bo owings. Finally, he household
head’s age, gende , expe ience, income, a m size, he amoun o ade c edi g an ed, he numbe o p e ious
bo owings, na u al disas e isk, and he numbe o banks he a me has ela ionships wi h a e he
de e minan s o he sel -selec ion o a me s no applying o loans because o a ea no being able o epay.
P ac ical implica ions –This pape ills he knowledge gap by in es iga ing why a me s sel -selec ou o
o mal c edi ma ke s. I p o ides e idence o how he a me s’subjec i e pe cep ions o u al c edi ma ke s
con ibu e o hei sel -selec ion.
O iginali y/ alue –This pape shows ha demand-side cons ain s a e also i al o a me s’access o bank
c edi . Imp o ing c edi access ia easing supply-side cons ain s may no inc ease c edi up ake wi hou
add essing demand-side ac o s. Gi en ha inding, i ecommends policies o imp o e access o bank c edi o
a me s ega ding he demand side.
Keywo ds Bayes, C edi , Mul inomial p obi , Subjec i e pe cep ion, Sel -selec ion, Vie nam
Pape ype Resea ch pape
1. In oduc ion
Access o o mal c edi helps a me s imp o e p oduc i i y, p oduc quali y, and income
(Chandio e al., 2017;Hu chins, 2023;Ki os and Meshesha, 2022). The e o e, denied access, o
Asian Jou nal o
Economics and
Banking
105
JEL Classi ica ion —D10, D82, G21, G51
© Le Khuong Ninh. Published in Asian Jou nal o Economics and Banking. Published by Eme ald
Publishing Limi ed. This a icle is published unde he C ea i e Commons A ibu ion (CC BY 4.0)
licence. Anyone may ep oduce, dis ibu e, ansla e and c ea e de i a i e wo ks o his a icle ( o bo h
comme cial and non-comme cial pu poses), subjec o ull a ibu ion o he o iginal publica ion and
au ho s. The ull e ms o his licence may be seen a h p://c ea i ecommons.o g/licences/by/4.0/
legalcode
The cu en issue and ull ex a chi e o his jou nal is a ailable on Eme ald Insigh a :
h ps://www.eme ald.com/insigh /2615-9821.h m
Recei ed 2 Feb ua y 2023
Re ised 27 May 2023
14 Ap il 2024
9 May 2024
31 May 2024
20 June 2024
Accep ed 24 June 2024
Asian Jou nal o Economics and
Banking
Vol. 9 No. 1, 2025
pp. 105-125
Eme ald Publishing Limi ed
e-ISSN: 2633-7991
p-ISSN: 2615-9821
DOI 10.1108/AJEB-02-2023-0011
Downloaded om h p://www.eme ald.com/ajeb/a icle-pd /9/1/105/9699955/ajeb-02-2023-0011.pd by ZBW Ge man Na ional Lib a y o Economics use on 16 Decembe 2025
c edi a ioning, imposes cons ain s and induces hem o e ea o low-expec ed e u n
ac i i ies (Kuhn and Bobojono , 2023;Migheli, 2024). S udies ha e asce ained ha supply-
side ac o s s emming om in o ma ion asymme y, leading o ad e se selec ion and mo al
haza d, a e causes o c edi a ioning agains a me s, especially smallholde s (Ba slund and
Ta p, 2008;Do and Baue , 2016;Du hues e al., 2004;Pham and Izumida, 2002;Phan e al.,
2013;T an e al., 2018). Alle ia ing supply-side cons ain s is hus ecommended o ge o e
he p oblem. Howe e , ha app oach may no necessa ily inc ease c edi up ake wi hou
add essing demand-side ac o s, which also play a i al ole in de e mining a me access o
c edi . Fa me s may decide no o apply o o mal c edi (o sel -selec ) due o hei
pe cep ions o his c edi sou ce and isk p e e ences.
Sel -selec ed a me s canno be igno ed in analyzing c edi a ioning because hey a e
nume ous (B iggeman e al., 2009;F eel e al., 2012;Le enson and Willa d, 2000;Ogane, 2023).
Failing o conside hem can bias he es ima es since hei sel -selec ion induces c edi
ins i u ions o apply sc eening ules di e en om wha would p e ail i hey did no sel -
selec . Such a law would dis o he empi ical analysis o he e iciency o u al inancial
ma ke s and esul in disappoin ing policies o de elop hem. Bu , e idence on why a me s
sel -selec ou o o mal c edi ma ke s is somehow scan . The p esen pape aims o ill he
knowledge gap by in es iga ing wha d i es a me s who need ex e nal unds o sel -selec .
To do ha , we use a p ima y da a se om 2,212 ice a ming households in he Mekong Ri e
Del a (MRD) in Vie nam – he egion ha accoun s o mo e han 50% o he coun y’s ice
ou pu . I s con ibu ion o he exis ing li e a u e is o p o ide e idence o how subjec i e
pe cep ions o a me s p opel hei sel -selec ion, dis inguishing i om p e ious pape s ha
mainly ocused on i ms (e.g. Beck e al., 2018;Bi can and de Haas, 2019). Mo eo e , i s inding
implies ha sel -selec ion would ha e consequences o c edi access denial, o one should
conside sel -selec ion beha io when s udying c edi a ioning agains a me s and
p oposing policies o de elop u al c edi ma ke s.
This pape p oceeds as ollows. Sec ion 1 o he in oduc ion is ollowed by Sec ion 2,
which e iews he backg ound li e a u e. Sec ion 3 p esen s he se ing in which he a me s
ope a e. The empi ical model and he me hodology appea in Sec ion 4.Sec ion 5 discusses
sampling and he da a se used in he pape . Sec ion 6 p esen s he esul s, and Sec ion 7
concludes and ende s policy ecommenda ions.
2. Backg ound li e a u e
In isky se ings, agen s may no know he p obabili y o unce ain happenings and hus
make decisions based on subjec i e pe cep ions ha may no necessa ily co espond o he
u h. Tha holds o a me s ega ding hei loan applica ions. Some imes, hey know he
bank’s policy and decide o apply o loans. This ac ion occu s in se ings whe e bo owe s
equen ly in e ac wi h banks. In o he cases, hey make decisions based on impe ec
in o ma ion, e.g. expe ience o pas obse a ions, ins ead o he will-be decision o he bank
owa d clien s, which is known o success ul applican s only. Then, hey consciously sel -
selec ou o o mal c edi ma ke s (i.e. demand-side cons ain ) due o hei mispe cep ion o
bo owing cos s, sc eening ules, o in e es a es, condi ional on hei isk p e e ences.
Besides he supply-side cons ain , demand-side ac o s could a ec he unc ioning o u al
c edi ma ke s, he adop ion o new p oduc ion echnologies, and a me li elihoods (Balana
and Oyeyemi, 2022;Pe ick, 2004). In he case o unc edi wo hy bo owe s, sel -selec ion is
no p oblema ic since i adds o he e icien unc ioning o he c edi ma ke s. On he
con a y, he beha io o c edi wo hy bo owe s leads o sub-op imal le els o in es men ,
supp essing hei p oduc ion and income. The e o e, i he ex en o sel -selec ion is
subs an ial, add essing ha issue p o ides mo e app op ia e policy implica ions han he
adi ional supply-side mechanism (F eel e al., 2012).
AJEB
9,1
106
Downloaded om h p://www.eme ald.com/ajeb/a icle-pd /9/1/105/9699955/ajeb-02-2023-0011.pd by ZBW Ge man Na ional Lib a y o Economics use on 16 Decembe 2025
Ne e heless, he subjec i e pe cep ions o bo owe s could be w ong since hey would ge
he loans i applying (Fe ando and Mullie , 2022). Then, hey alsely sel -selec because o
hei isk p e e ences, loss a e sion, and in o ma ion wedges be ween hemsel es and he
banks. Gi en in o ma ion wedges, hey decide based on hei expe ience o obse a ions o
banks’disc imina ion agains disad an aged bo owe s wi h ejec ed applica ions. The e a e
wo ypes o disc imina ion, i.e. s a is ical and as e-based (De And
es e al., 2021;Han, 2004).
S a is ical disc imina ion occu s due o in o ma ion asymme y as banks ejec applican s
based on obse ed cha ac e is ics. They use a se o loan applican s’a ibu es (e.g. asse s,
income, age, o ela ionship lending) o p edic hei c edi wo hiness and decide whe he o
no o accep hei loan applica ions. Tas e-based disc imina ion emana es om animus o
p ejudices owa ds applican s acco ding o speci ic ea u es, e.g. ace, gende , o e hnici y
(Gu yan and Cha les, 2013).
Le enson and Willa d (2000) and F eel e al. (2012) epo ed ha sel -selec ed bo owe s
we e wice as ejec ed. B iggeman e al. (2009) e ealed he sha e o sel -selec ed a me s
being 16% o hei sample. The e o e, a good unde s anding o sel -selec ion and i s
consequences could be i al o designing app op ia e policies o enhance he access o c edi
o economic en i ies. Fo banks, i is only p o i able o lend o c edi wo hy bo owe s, so hey
ha e o lea n bo owe cha ac e is ics h ough sc eening bu possibly e . Sc eening e o s
occu when banks deny c edi wo hy applican s and accep unc edi wo hy ones. Bo owe
cha ac e is ics and bank sc eening quali y ma e o sel -selec ing because o he sunk cos s
in he applica ion p ocess (Fe ando and Mulie , 2022). Bo owe s who need c edi ade o
he cos s and bene i s o applying o a loan, condi ional on hei subjec i e pe cep ions. As
a gued by Kahneman and T e sky (1979),Kahneman e al. (1991), and Kahneman and
T e sky (1992), ou comes pe cei ed as gains a e alued acco ding o a conca e unc ion,
whe eas ou comes pe cei ed as losses a e alued acco ding o a con ex unc ion, and he loss
unc ion is much s eepe han he gain unc ion. The es ima es o his di e en ial alua ion
a e in he ange o 2.25/1.00 (i.e. a loss o a alue o x gene a es a deg ee o disu ili y ha is
mo e han wice he u ili y gene a ed by a gain o he same alue). The e o e, bo owe s end
no o apply o c edi when acing high bo owing cos s (in e es paymen s, oppo uni y
cos s, and applica ion cos s), low expec ed e u ns om p oduc ion ha de e mine hei
abili y o hono he deb s, and a high ex-an e likelihood o ejec ion. Applica ion cos s, which
a e among he mos i al elemen s ha d i e a me sel -selec ion, ake di e se o ms,
including inancial (i.e. he cos o accoun an s), in-kind ( ime spen ), and psychological (denial
discom o ).
S udies o en ocused on de eloped coun ies whe e he p oblem o sel -selec ion is less
pa amoun han in de eloping ones wi h unde de eloped inancial sys ems, a high deg ee o
in o ma ion asymme y, and subs an ial bo owing cos s (Popo and Ongena, 2011). In he
la e coun ies, bo owe s ha e di icul ies signaling hei unobse able quali y o banks.
Mo eo e , hey a e no always willing o elease in o ma ion since hey ind i ime-
consuming (cos ly) o wan o hide away a ibu es and ac ions. The unobse able quali y o
he bo owe , which ende s banks’sc eening e o s and subs an ial applica ion cos s, is also
a de e minan o sel -selec ion (Gamma e al., 2017;Kon and S o ey, 2003). Sel -selec ed
bo owe s gi e up seeking loans because hey expec ei he high cos s o ge ing loans o low
c edi limi s main ained by banks o a oid de aul s om bo owe s (Chak a a y and Xiang,
2013). Those sel -selec ed bo owe s may be w ong in hei an icipa ions because hey may
ge loans i applying. Ye , mos a m-le el s udies in he ex an li e a u e do no con ol o
a me sel -selec ion, he eby ende ing a po en ial sou ce o bias in es ima es. The p esen
pape conside s he choice o loan applica ion as sel -de e mined (i.e. bo owe s hemsel es
choose whe he o apply o no acco ding o hei subjec i e pe cep ions), which helps be e
explain a me access o c edi , in addi ion o he c edi a ioning om he supply side o en
analyzed.
Asian Jou nal o
Economics and
Banking
107
Downloaded om h p://www.eme ald.com/ajeb/a icle-pd /9/1/105/9699955/ajeb-02-2023-0011.pd by ZBW Ge man Na ional Lib a y o Economics use on 16 Decembe 2025
3. Se ing
Vie nam’s ice sec o has played a i al ole in helping ensu e na ional ood secu i y and
enabled he coun y o become a leading ice expo e since he 1980s (Ba e al., 2019;Dao e al.,
2023;Pham and Izumida, 2002). Tha achie emen was mainly a ibu able o he economic
e o ms (Doi moi) and he e icien cos compe i i eness s a egy. Howe e , he e e se side o
he medal is ha Vie nam is o en known o being an expo e o ice o in e io quali y
(Cao and Le, 2020), a bad epu a ion ha pe sis s and leads o a cos -p ice squeeze due o low
p ices and high p oduc ion cos s. Smallholde s who accoun o app oxima ely 70% o
Vie namese ice a me s and can no longe abso b high p oduc ion cos s p ocu e in e io
inpu s, p oduce low-quali y ice, and use ou da ed ha es ing and pos -ha es echniques
and machine y. They hus c ea e iny added alue and ace declining p o i abili y. One way
o b eak he icious ci cle o he cos -p ice squeeze and gain a b igh e image o Vie namese
ice is o imp o e access o c edi o i s a me s, who ha e su e ed a subs an ial ex en o
c edi a ioning (Ba slund and Ta p, 2008;Nguyen e al., 2022).
Se e al s udies ha e in es iga ed he incidence o c edi a ioning o Vie namese a me s
(Ba slund and Ta p, 2008;Do and Baue , 2016;Du hues e al., 2004;Nguyen e al., 2022;Phan
e al., 2013;T an e al., 2018). They ace c edi a ioning due o in o ma ion asymme y, lack o
colla e al, isks o many kinds (e.g. disease, ice and inpu p ice ola ili ies, d ough , and
lood), and high ansac ion cos s. As a esul , hey ind i ha d o pu chase inpu s and apply
da ed p oduc ion echnologies, hus limi ing he op imum p oduc ion choice and p oduc
quali y. They end o s ay small acco dingly, ailing o exploi economies o scale o imp o e
e iciency and income. Those indings seemingly ha e meaning ul policy implica ions o
u al de elopmen and a ming household li elihood (Pham and Izumida, 2002;Nguyen e al.,
2022). Access o c edi a ec s a ming household wel a e ia a leas wo channels. Fi s , by
alle ia ing capi al cons ain s on households, access o c edi imp o es hei abili y o
p ocu e necessa y inpu s, educes he oppo uni y cos s o capi al-in ensi e asse s,
encou ages labo -sa ing echnology, and inc eases land p oduc i i y. Second, households
wi h access o c edi may be mo e willing o apply p omising bu isky echnologies and will
be e s ay away om isk-mi iga ing bu undesi able li elihood s a egies (Ba slund and
Ta p, 2008).
When s udying c edi a ioning agains a me s o p opose solu ions and policies, one
may ace po en ial selec ion bias because assigning hem o c edi -cons ained and no c edi -
cons ained g oups may no be andom. Ins ead, so ing hem in o a ious g oups o c edi
a ioning depends on hei loan-applying beha io s de e mined by hei subjec i e
pe cep ions o he c edi sou ce and o he ele an ac o s. No ully conside ing his
aspec , p e ious s udies ha e jus di e en ia ed a me s who ob ained c edi a e mul iple
a emp s om hose who did no . Howe e , some a me s decided no o apply o loans o
sel -selec (B iggeman e al., 2009). The solu ions and policies p oposed may no wo k i
excluding sel -selec ed bo owe s, a p oblem o many s udies on c edi access o Vie namese
a me s, which he p esen pape add esses.
4. Empi ical model and me hodology
We i s iden i y he de e minan s o he sel -selec ion o a me s by es ima ing an empi ical
model using a p obi es ima o . The model eads as ollows:
Sel Selec i¼
α
iþβiZiþ
ε
i;(1)
whe e Sel Selec
i
akes a alue o 1 i a me iis sel -selec ed and 0 o he wise, and Ziis a ec o
o explana o y a iables, including ea u es o household heads (educa ion a ainmen , age,
gende , and a ming expe ience) and o households ( he numbe o amily labo e s, pe capi a
income, a m size, he amoun o ade c edi g an ed, dis ance o he nea es bank, whe he o
AJEB
9,1
108
Downloaded om h p://www.eme ald.com/ajeb/a icle-pd /9/1/105/9699955/ajeb-02-2023-0011.pd by ZBW Ge man Na ional Lib a y o Economics use on 16 Decembe 2025
no o ha e ela i es and iends wo king o banks, he numbe o p e ious bo owings,
isks, and he numbe o banks he household has ela ionships wi h).
We hen use a mul inomial p obi (MNP) eg ession o model he di e en easons o he
sel -selec ion o he a me s. The mul inomial speci ica ion allows us o join ly es ima e all he
easons o sel -selec ion while accoun ing o po en ial co ela ion be ween di e en easons.
The mul inomial model is as ollows:
P ðReasonSel Selec i¼jjApplyi¼0Þ¼ΨðγiþκjZiÞ;(2)
whe e Ψð$Þis a mul inomial unc ion. ReasonSel Selec
i
is a ca ego ical a iable aking alue
o 1 (j51) i he eason is “ha ing access o o he p e e ed sou ces o c edi ”;2(j52) i he
eason is “un a o able loan e ms”;3(j53) i he eason is “high bo owing cos s”;4(j54) i
he eason is “complex loan applica ion p ocedu es”; and 5 (j55) i he eason is “ ea o an
inabili y o epay he loan”.Ziis a ec o o explana o y a iables, and κjis he ec o o
coe icien s o be es ima ed.
MNP model is commonly used when he dependen a iable is ca ego ical and akes mo e
han wo ca ego ies. In such a si ua ion, he dependen a iable yis an uno de ed ca ego ical
a iable, and an indi idual may selec one o he al e na i es, o all unde one o he
ca ego ies. The choices o ca ego ies can be coded as j50, 1, ...,m, whe e mis he numbe o
a ailable choices o ca ego ies. In ou analysis, we le y
i
be he ca ego ical a iable ha akes
alues j ¼0;1;2, ..., 5 ha ep esen s he i h household’s decision o sel -selec . De ine y*
ij as
he unobse ed p opensi y o he i h a me o sel -selec o eason j:
y*
ij ¼x=
iβþ
ε
ij
The obse ed ca ego y is he one wi h he highes p opensi y. The MNP model ha he i h
household alls in he j h decision o sel -selec can hus be modeled as ollows:
Pij ¼Pðyi¼jÞ¼Py*
ij >y*
ik¼wx=
jβ;∀j≠k;
whe e P
ij
ep esen s he p obabili y ha he i h indi idual alls in o he j h ca ego y, x=
jis a
ec o o eg esso s, βis he pa ame e s o be es ima ed, and wis a p obi unc ional
e alua o .
5. Sampling and da a
The da a used in his pape come om a su ey o 2,212 ice a ming households in he MRD
in 2018, a yea be o e he b eakdown o COVID-19. We use di ec su ey in o ma ion o
cons uc he dependen a iable (i.e. he sel -selec ion ou o o mal c edi ma ke s) and he
explana o y a iables in Models (1) and (2). To c ea e he sample, we apply he s a i ied
mul is age clus e sampling me hod. The MRD consis s o 12 p o inces and one ci y. We
s a i y his egion by p o ince (ci y). Fi s , we chose he dis ic wi h he la ges land a ea
de o ed o ice a ming wi hin each locali y (p o ince o ci y). F om ha , we selec ed he
illage wi h hela ges ice land a ea. Then, we andomly selec ed 250 ice a ming
households om he selec ed illage, using he lis s acqui ed om he people’s commi ee, o
in e iew. We use a ques ionnai e o conduc ace- o- ace in e iews wi h household heads.
Howe e , due o di icul ies eaching some household heads, being e used by hem, and
missing and implausible in o ma ion, we could c ea e a da ase o 2,212 households.
The MRD, which makes up mo e han 50% o Vie nam’s ice ou pu , is a egion endowed
wi h a o able ag o-ecological po en ial, especially in ice a ming, and has nume ous e ile
a eas na u ally i iga ed, helping p oduce up o h ee c ops pe yea (Kompas e al., 2012).
Al hough his egion does no ep esen he en i e coun y and he la ge egion, i s
Asian Jou nal o
Economics and
Banking
109
Downloaded om h p://www.eme ald.com/ajeb/a icle-pd /9/1/105/9699955/ajeb-02-2023-0011.pd by ZBW Ge man Na ional Lib a y o Economics use on 16 Decembe 2025
impo ance in ice p oduc ion, ad anced ice a ming echniques, ib an ice ma ke , and
ich and di e se na u al esou ces p o ide an excellen case o iden i ying he easons
behind he sel -selec ion o a me s.
6. Resul s
6.1 Desc ip i e s a is ics
Table 1 shows ha he educa ion o he household heads is no ably low, wi h a mean o
6.42 yea s o schooling. Tha ac is unde s andable since mos household heads a e ela i ely
old (wi h a mean age o 52.10) and g ew up in he cen al planning e a when he educa ion
sys em was imma u e. Such a low le el o educa ion may induce hem o sel -selec because i
migh be ha d o g asp he loan applica ion o ms (Le, 2021), among o he s. Abou 92% o he
household heads a e male and a e mo e ac i e in ice a ming, na u al esou ce ex ac ion
( ishing, hun ing, collec ing, o logging), and social and business ac i i ies (e.g. waged
labo ing). The household heads ha e ela i ely long expe ience in ice a ming (29.92 yea s
on a e age) because hey ha e almos esided he e since bi h.
The a e age household has 3.24 wo king-age membe s (Table 1), implying a labo
abundance gi en he small a m size (see below). Howe e , because young and be e -
educa ed people end o mig a e, lea ing child en, ma ied women, and aged people behind,
he e is a lack o u al labo in he MRD. The annual pe capi a income is VND 41.13 million on
a e age, wi h a s anda d de ia ion o VND 38.81 million. The mean a m size is 17.35 cong (i.e.
1,000 m2) wi h a s anda d de ia ion o 15.26 cong (Table 1), meaning ha he sample co e s a
wide ange o a m sizes (i.e. be ween 1 and 130 cong). In he MRD, land accumula ion has
occu ed as well-o a me s pu chased, while less p oduc i e ones wi h high le els o land
endowmen and ewe asse s end o sell i . Some a me s sold hei land when mig a ing,
empo a ily o pe manen ly, o ake up oppo uni ies elsewhe e. This ac ion esul s in a bi
g ea e land-use e iciency bu causes he eme gence o a s a um o well-o a me s and
smallholde and landless ones (T an, 2018).
Table 1 also di ulges ha he a me s ha e ecei ed a lo o ade c edi om inpu
supplie s (VND 44.81 million on a e age), which bene i s bo h pa ies (Am ago and Mensah,
2022;Bu ka and Ellingsen, 2004). Ne e heless, his c edi a ies among hem because he
supplie s g an i based on buye epu a ion, epaymen abili y, in ima e ela ionships, and
C i e ia Mean S.D. Min Max
Educa ion o household head 6.42 3.16 0 15
Age o household head 52.10 11.70 24 93
Gende o household head 0.92 0.28 0 1
Fa ming expe ience (yea s) 29.92 12.20 0 70
Numbe o household labo e s 3.24 1.13 0 8
Pe capi a income (VND million) 41.13 38.81 0 471
Fa m size (1,000 m
2
) 17.35 15.26 0 130
T ade c edi (VND million) 44.81 60.98 0 993
Dis ance o he nea es bank (km) 6.46 5.12 0 40
Rela i es o iends wo king o banks 0.17 0.37 0 1
Numbe o p e ious bo owings 1.34 2.34 0 19
Na u al disas e - ela ed isk 0.16 0.37 0 1
Disease- ela ed isk 0.22 0.42 0 1
Ou pu p ice- ela ed isk 0.41 0.42 0 1
Numbe o banks wi h which he a e age a me has ela ionships 0.77 0.49 0 5
Sou ce(s): The au ho
Table 1.
Desc ip i e s a is ics
AJEB
9,1
110
Downloaded om h p://www.eme ald.com/ajeb/a icle-pd /9/1/105/9699955/ajeb-02-2023-0011.pd by ZBW Ge man Na ional Lib a y o Economics use on 16 Decembe 2025
aluable asse s (Cao and Le, 2020). A i s simples de ini ion, ade c edi allows i s ecipien
o delay paymen o he p oduc . In o he wo ds, i is a join commodi y and inancial
ansac ion in which a selle sells a good and simul aneously ex ends c edi o he pu chase
o he buye . T ade c edi is hus a subs i u e o bank c edi (Am ago and Mensah, 2022;
Mu o and Pe uzzi, 2022). Gi en he ecen u al in as uc u e de elopmen s as an e o o
he go e nmen , he mean dis ance om a a me ’s esidence o he nea es bank is
6.46 kilome e s, which is sho and en ails an ad an age o he a me s i wishing o
app oach banks o loans. Howe e , a ela i ely la ge de ia ion o 5.12 kilome e s implies
ha he dis ance is di e se since he a me s eside spa sely and ew comme cial banks ha e
b anches in u al a eas (Migheli, 2024). Fa me s close o a bank may enjoy lowe ansac ion
and oppo uni y cos s in bo owing and be e c edi - ela ed in o ma ion and ha e access o a
la ge pool o c edi oppo uni ies.
Abou 17% o he su eyed households ha e ela i es o iends wo king o banks.
Unde inc easing compe i i e p essu e, banks in Vie nam began o shi ocus o an e ec i e
sales o ce o ind clien s o pa onize p oduc s and e ain hem. This ma ke ing s a egy
eso s o pe sonal selling as a panacea o pe suade clien s o buy p oduc s. In his espec ,
in ima e kinship and iendship ies b ing a mu ual bene i since hey help he bank o ice s
imp o e hei wo k pe o mance, and hei ela i es and iends ge be e access o c edi .
Rela i es and iends wo king o banks also ac as in e media ies ha connec he a me s
wi h p ospec i e banks, hus educing he in o ma ion asymme y ha ha ms bank p o i s
and incen i izing he o me o apply o loans. The a e age numbe o p e ious bo owings
is abou 1.34, bu he s anda d de ia ion (2.34) indica es ha his igu e la gely a ies ac oss
he a me s. Rice p ice- ela ed isk, which esul s om he unde de eloped ice ma ke ing
channel and he s a e’s dis o ed in e en ions in he ice ma ke , seems pe manen (Ful on
and Reynold, 2015), isking he a me ’s income. Na u al disas e - ela ed isks and diseases-
ela ed isks appea unimpo an o a me s. Finally, he a e age a me has ela ionships
wi h only 0.77 banks because a ew banks ha e ope a ed in he u al MRD due mainly o high
ansac ion cos s and de aul isk.
6.2 Fa me sel -selec ion
To ob ain in o ma ion abou he sel -sec ion o a me s, we asked hem i hey had e e
applied o a bank loan when wan ing one. Any esponden answe ing “no”was kindly
eques ed o choose one ou o he ollowing easons o ha decision: ha ing access o o he
p e e ed sou ces o unds (Reason 1), un a o able loan e ms (Reason 2), high bo owing
cos s (Reason 3), complex loan applica ion p ocedu es (Reason 4), and ea o an inabili y o
epay he loan (Reason 5). We ega d sel -selec ed bo owe s as hose who need ex e nal
inance bu choose no o apply o a bank loan o any o he abo emen ioned easons.
The e a e 974 sel -selec ed bo owe s ou o 2,212 su eyed households (44%), whe eas
304 a me s applied bu did no ge h ough (13.74%). This ou come means ha he demand
side is mo e cons ained han he supply side, ein o cing he a ionale o his s udy. Table 3
documen s he equency o he easons o sel -selec ing ou o o mal c edi ma ke s. This
able shows ha Reason 5 ( ea o an inabili y o epay he loan o deb a e sion) leads he
easons o a me s o sel -selec . Deb a e sion (i.e. an unwillingness o en e a deb con ac )
dis o s in es men and inancing decisions (Eckel e al., 2007;Nguyen e al., 2021) and is
somehow ela ed o hei isk-a e se a i ude. Indeed, isk-a e se a me s a e less willing o
ake on ac i i ies ha may b ing good ou comes bu ca y isks o ailu e (Ullah e al., 2015).
The isk a e sion s ems om hei pe cep ion o calami ous isk sou ces, including loods,
hea y ains, pes s, diseases, d ough , and saline in usion (Le and T uong, 2019;Nguyen
e al., 2022). Vola ili ies in ice p ices also make hem isk-a e se conce ning income ha
cons i u es hei abili y o epay deb s (Ful on and Reynolds, 2015). Facing mul iple isks,
Asian Jou nal o
Economics and
Banking
111
Downloaded om h p://www.eme ald.com/ajeb/a icle-pd /9/1/105/9699955/ajeb-02-2023-0011.pd by ZBW Ge man Na ional Lib a y o Economics use on 16 Decembe 2025
a me s sel -selec om o mal c edi ma ke s because hey pe cei e hei deb epaymen
capaci y as so low ha hey should no apply. Following Reason 5 is Reason 1 (ha ing access
o o he p e e ed sou ces o unds), which accoun s o 30.18% o he sel -selec ed a me s.
This inding implies ha a me s use o he c edi sou ces (especially ade c edi due o i s
popula i y) as a subs i u e o bank c edi (Am ago and Mensah, 2022).
When he lende canno obse e o condi ion loan o e s on he e ogeneous a me
cha ac e is ics, s ic bo owing equi emen s migh sc een ou unc edi wo hy bo owe s,
who a e mo e likely o de aul (Sengup a, 2014). As an unin en ional mis ake, hey may also
sc een ou c edi -cons ained a me s wi h high alua ions who would epay hei loans. Due
o in o ma ion opaqueness, ansac ion cos , and weak en o cemen , banks o en main ain
de ailed con ac s o shi con ac ual isks o bo owe s and ha e hem epay he loan. Such
a policy appea s p oblema ic o ill-educa ed a me s who ind i ha d o ul ill he loan
applica ion o ms and hose who eside a a om banks because o subs an ial a el cos s. In
ou case, complex loan applica ion p ocedu es esul ing om hose s ic bo owing
equi emen s (Reason 4) induce 205 a me s (21.05%) o sel -selec .
Acco ding o Table 2, 120 a me s (12.32%) sel -selec ed because o high bo owing cos s
(Reason 3). Indeed, ansac ion cos s we e unduly signi ican o u al c edi ansac ions.
Since c edi con ac s a e no ins an aneous like spo ma ke ansac ions, he ansac ing
pa ies ace ad e se selec ion and mo al haza ds, leading o ansac ion cos s (S igli z and
Weiss, 1981;Gin
e, 2011). T ansac ion cos s in c edi deli e y can be non inancial ones
incu ed by lende s and bo owe s be o e, du ing, and a e disbu semen . Fo bo owe s,
ansac ion cos s include anspo a ion cos s o and om he bank and oppo uni y cos s (i.e.
ime spen a eling, going h ough leng hy bu eauc a ic p ocedu es, and wai ing o
accep ance by he bank). Gi en a low income, many a me s ca e abou he cos sa ings
associa ed wi h he ease and quickness o c edi deli e y, which explains why, in many cases,
hey p e e in o mal c edi sou ces (Migheli, 2024).
The las eason o a me s o sel -selec (i.e. un a o able loan e ms, which include he
loan’s epaymen pe iod, he in e es a e and ees associa ed wi h he loan, penal y ees
cha ged o bo owe s, and any o he special condi ions ha may apply) appea s i ial since
i a ec s only 5.58% o he sel -selec ed a me s. This inding is unde s andable, as nume ous
a me s ha e had o bo ow om in o mal lende s and inpu ade s a high e ec i e in e es
a es due o he unde de elopmen o he u al c edi ma ke s. The e o e, a ew deem he loan
e ms main ained by banks un a o able, al hough hey do ca e ha he ime o selling ice o
ea n cash o epay he loan may no ma ch i s epaymen pe iod.
6.3 Sel -selec ed e sus non-sel -selec ed a me s
Table 3 shows he - es on he di e ence be ween non-sel -selec ed and sel -selec ed a me s.
The e is a signi ican di e ence in a ming expe ience be ween hese wo ca ego ies o
a me s as he non-sel -selec ed appea mo e expe ienced han hei coun e pa s. The
Reasons Numbe o obse a ions Pe cen age o o al
1 Ha ing access o o he p e e ed sou ces o unds 294 30.18
2 Un a o able loan e ms 57 5.85
3 High bo owing cos s 120 12.32
4 Complex loan applica ion p ocedu es 205 21.05
5 Fea o an inabili y o epay he loan 298 30.60
To al 974 100.00
Sou ce(s): The au ho
Table 2.
Reasons o sel -
selec ing
AJEB
9,1
112
Downloaded om h p://www.eme ald.com/ajeb/a icle-pd /9/1/105/9699955/ajeb-02-2023-0011.pd by ZBW Ge man Na ional Lib a y o Economics use on 16 Decembe 2025
Finally, he mul iple-bank ela ionship gi es incen i es o apply o a loan since hey can
selec he bank ha can i hei demand he mos . The mul iple-bank ela ionship also helps
a me s educe he “liquidi y isk” ha o ces hem o liquida e p ema u ely due o und
sho ages (De agiache e al., 2000).
Like some o he g oups o a me s, he income o a me s who pe cei e he applica ion
p ocedu es as complex likely induces hem o sel -selec (column (5) o Table 6). This inding
a ises because well-o a me s may ha e a high oppo uni y cos o ime and a s ong
mindse o e iciency, as ea lie a gued. The numbe o p e ious bo owings mo i a es hose
a me s o apply o a loan o no o sel -selec . This upsho makes sense since a lending
ela ionship may help hem ul ill he p ocedu es as e and mo e smoo hly (Mu o and
Pe uzzi, 2022). Finally, he numbe o banks ha hose a me s ha e c edi ela ionships
educes he deg ee o hei sel -selec ion o he eason iden i ied p e iously.
Column (6) e eals ha age makes a me s who sel -selec because o “ ea o an inabili y
o epay he loan”mo e sel -selec ed because aging dep i es hem o sel -con idence in he
capaci y o make money o epay deb s. This column also shows ha male a me s end o be
mo e sel -selec ed since hey a e he ones who bea he expense bu den o he amily. This
e ec may be e iden in low-income, smallholding households. Expe ience in ice a ming
inspi es hese a me s o apply o a loan because, as explained ea lie , he expe ience may
make hem less isk-a e se and hei p oduc ion p o i able. Fa me s wi h highe incomes and
la ge a m sizes end no o sel -selec because hey a e weal hie and be e able o hono he
deb . T ade c edi en e s posi i ely and signi ican ly, implying ha a me s who ge ade
c edi a e less likely o depend on bank loans ha a e p obably subs i u es o ade c edi
and ice e sa. The numbe o p e ious bo owings s ipula es a me applica ion o a new
loan because o he e ec o ela ionship lending (Mu o and Pe uzzi, 2022). Disease isk
in ensi ies sel -selec ion because o he ea o disease des oying he c op, exace ba ing hei
abili y o epay deb s. Again, he numbe o banks alle ia es hei incen i e o sel -selec o
opening up chances o hem o selec he bank o e ing a o able loan e ms.
6.4.3 Robus ness checks. Robus ness checks a e a s anda d ea u e o empi ical wo k.
A e es ablishing hei main esul s, esea che s may ask i al e na i e models can explain
hei indings and p o ide addi ional insigh s in o he p oblem. As you may see, we ely on
p- alues o es ing he e ec s o ac o s on he a me ’s decision o sel -selec (Table 5) and
easons o hei sel -selec ing (Table 6) using he p obi and he MNP es ima o s,
espec i ely. Using p- alues o hypo hesis es ing has been he no m in he economics
li e a u e. Howe e , p- alues e lec only he p obabili y o he es ima ed e ec , assuming he
null hypo hesis is ue. p- alues a e calcula ed om an in ini e numbe o eplica ions ha
ne e eally happened. They migh o e s a e he e idence agains he null (Kass and Ra e y,
1995;Louis, 2005) and do no e lec he size o he impac o he independen a iables on he
dependen one. p- alues a e also sensi i e o sample size and he da a used, i.e. he so-called
sampling-based concep . To a oid hose limi a ions o p- alues, one can eso o he
Bayesian app oach.
Gi en i s ad an ages, Bayes pos e io s ( ac o s) can be a mo e a ac i e al e na i e o
hypo hesis es ing han p- alues (Assa and Tsionas, 2018). Bayesian modeling can
inco po a e p io knowledge in o he model. In o he wo ds, Bayesian modeling is an
app oach o da a analysis based on Bayes’ heo em, which upda es a ailable knowledge
abou pa ame e s in a s a is ical model wi h he in o ma ion in obse ed da a. The
backg ound knowledge is exp essed as a p io dis ibu ion and combined wi h obse a ional
da a ia a likelihood unc ion o e eal he pos e io dis ibu ion (Van de Schoo e al., 2021).
Then, he pos e io helps o make p edic ions abou u u e e en s. Since he Bayes pos e io s
( ac o s) come om he Bayesian app oach, which elies solely on he obse ed sample o
p o ide di ec p obabili y s a emen s abou he pa ame e s o in e es , i is mo e sui ed o
hypo hesis es ing. In addi ion, he Bayesian app oach expands he ange o es able
Asian Jou nal o
Economics and
Banking
119
Downloaded om h p://www.eme ald.com/ajeb/a icle-pd /9/1/105/9699955/ajeb-02-2023-0011.pd by ZBW Ge man Na ional Lib a y o Economics use on 16 Decembe 2025
hypo heses and in e p e s esul s in in ui i e ways ha do no ely on null hypo hesis
signi icance es ing like p- alues. F om his pe spec i e, Bayes pos e io s ( ac o s) can be
conside ed as al e na i es o p- alues (o signi icance p obabili ies) o es ing hypo heses
and o quan i ying he deg ee o which obse ed da a suppo o con lic wi h a hypo hesis
(La ine and Sche ish, 1999). The Bayesian app oach enables us o make di ec p obabili y
s a emen s abou he pa ame e o in e es . The e o e, we pe o m obus ness checks o he
indings in Table 5 using he Bayesian p obi es ima o and Table 6 using he Bayesian MNP
es ima o , wi h non-in o ma i e p io s selec ed. We selec non-in o ma i e p io s since we
ha e no p io in o ma ion abou he pa ame e s and wan p io s wi h minimal in luence on
he pos e io s.
Columns 4 and 5 o Table 5 epo he pos e io means compu ed by he Bayesian p obi
es ima o , and Table 7 epo s he pos e io means by he Bayesian MNP es ima o . Fo all
he a iables in Table 5, he pos e io means a e accep able (Assa and Tsionas, 2018) and
simila o he p obi es ima es. Fo all he a iables ha ha e s a is ically signi ican
coe icien s iden i ied by he MNP es ima o (Table 6), he pos e io means (Table 7) a e
simila o he MNP es ima es. Fo only h ee a iables ha do no ha e s a is ically
signi ican coe icien s (i.e. educa ion in column 6, amilylabo in column 4 and column 5, and
p ice isk in column 1 o Table 7), he pos e io means di e om he MNP es ima es since he
impac s o hose independen a iables on he dependen one a e weak o ambiguous. All
hese indings help con i m he obus ness o ou esul s.
7. Conclusion and policy ecommenda ions
Fa me sel -selec ion ou o o mal c edi ma ke s is phenomenal in u al Vie nam, bu
p e ious s udies ha e no documen ed i . The ailu e o accoun o his ac biases he esul s
Dependen a iable: Sel Selec –1 i sel -selec ing and 0 o he wise
Va iables
Reasons o sel -selec ing
Ha ing access o
o he p e e ed
sou ces o unds
Un a o able
loan e ms
High
bo owing
cos s
Complex loan
applica ion
p ocedu es
Fea o an
inabili y o
epay he loan
(1) (2) (3) (4) (5) (6)
C0.1586 1.6422 0.4624 1.1772 0.8059
Educa ion 0.0224 0.0271 0.0540 0.0172 0.0072
Age 0.0047 0.0138 0.0079 0.0055 0.0160
Gende 0.1733 0.0667 0.2309 0.0319 0.3226
Expe ience 0.0068 0.0144 0.0006 0.0043 0.0134
FamilyLabo 0.0790 0.0358 0.0139 0.0132 0.0251
Income 0.0023 0.0049 0.0042 0.0043 0.0068
Fa mSize 0.0227 0.0265 0.0142 0.0055 0.0290
T adeC edi 0.0018 0.0003 0.0020 0.0016 0.0023
Dis ance 0.0074 0.0003 0.0182 0.0026 0.0155
Rela i es 0.1633 0.1111 0.3692 0.1343 0.1621
NumBo 0.4153 0.2286 0.3022 0.3085 0.2905
Disas e Risk 0.0512 0.6499 0.6840 0.2036 0.0814
DiseaseRisk 0.0513 0.1167 0.5250 0.0626 0.3496
P iceRisk 0.0323 0.0374 0.3988 0.0786 0.0728
Numbe Bank 0.9144 0.6213 0.5051 0.1119 0.6818
N2,212
MCMC i e a ions 12,000
MCMC sample size 10,000
Accep ance a e 0.2178
Sou ce(s): The au ho
Table 7.
Bayesian MNP
es ima es o easons o
sel -selec ing
AJEB
9,1
120
Downloaded om h p://www.eme ald.com/ajeb/a icle-pd /9/1/105/9699955/ajeb-02-2023-0011.pd by ZBW Ge man Na ional Lib a y o Economics use on 16 Decembe 2025
since hei sel -selec ion may induce banks o apply di e en sc eening ules om hose ha
would p e ail i hey did no sel -selec . I also leads o useless sugges ions o imp o e lending
policies in pa icula and mone a y policies in gene al. This p oblem mo i a es us o conduc
his pape o ill he gap in unde s anding his issue using a p ima y da ase o 2,212 ice
a me s in he MRD o Vie nam.
The p esen pape uses he p obi and Bayesian p obi es ima o s o s udy he
de e minan s o a me s’sel -selec ion beha io . The esul s show ha olde a me s a e
inclined o sel -selec , as does he pe -capi a income. Fa m size is posi i ely ela ed o he
decision o apply o a loan because a la ge a m en ails a high lump sum cos o p oduc ion.
Fa me s wi h la ge a ms ha e sizable colla e al, making hem con iden abou ge ing he
loan. Fa me s wi hou ela i es o iends wo king o banks a e mo e inclined o sel -selec
due o a lack o in o ma ion. In o he wo ds, he ela i es o iends wo king o banks ac as
in e media ies be ween banks and a me s o ge mu ual bene i s. The numbe o bo owings
en e s posi i ely and signi ican ly, ein o cing he a gumen s by ela ionship lending. Risks
s emming om na u al disas e s ( loods, d ough s, and sal in usion), diseases, and ice p ice
ola ili y make he a me s less sel -selec ed. The numbe o banks wi h which a me s ha e
bo owing ela ionships has a signi ican ly nega i e coe icien , meaning ha a me s who
ha e es ablished ela ionships wi h mo e banks end no o sel -selec .
The esul s o he MNP and Bayesian MNP es ima o s e eal di e en de e minan s o he
sel -selec ion o a me s who do so o a ious easons. This inding would in e ha a me s a e
loss a e se, i.e. a cogni i e bias, meaning ha hey weigh a loss o esou ces mo e hea ily han a
gain o hose. Loss a e sion helps explain he widesp ead a me isk a e sion because esou ce
losses could esul in po e y, e en s a a ion, and a e hus a mo e impo an conside a ion han
gaining an ex a bi o hem. As one may see, he de e minan s o sel -selec ion a y acco ding o
why a me s do so. Howe e , he esul s show consis en e ec s o pe -capi a income, he
numbe o p e ious bo owings, and he numbe o banks a me s ha e ela ionships on he sel -
selec ion. This inding has no ed policy implica ions ha banks and policymake s should
conside he demand-side c edi cons ain s o induce a me s o apply o bank loans. Policies
ega ding educing isks o a me s, like c op insu ance, should be in place o sa egua d banks
om de aul s since isks make a me s less sel -selec ed. Wi h adequa e insu ance co e age,
a me s may unde ake somewha iskie bu mo e ewa ding a ming ac i i ies and apply o
bank loans. Al hough we can iden i y he de e minan s o a me s’sel -selec ion ou o o mal
c edi ma ke s, gi en he da a se , we canno es whe he sel -selec ion has eal e ec s on
a me s’p oduc ion and income. This issue is pe haps a opic o u he s udies.
Re e ences
Am ago, E.C. and Mensah, N.O. (2022), “T ade c edi om ag ochemical endo s as an al e na i e
sou ce o inance o cabbage p oduce s in he Bono Eas Region o Ghana”,Ag icul u al
Finance Re iew, Vol. 83 No. 1, pp. 43-82, doi: 10.1108/a -11-2021-0155.
Asadullah, M.N. and Rahman, S. (2009), “Fa m p oduc i i y and e iciency in u al Bangladesh: he
ole o educa ion e isi ed”,Applied Economics, Vol. 41 No. 1, pp. 17-33, doi: 10.1080/
00036840601019125.
As aw, A. and Admassie, A. (2004), “The ole o educa ion on he adop ion o chemical e ilize unde
di e en socioeconomic en i onmen s in E hiopia”,Ag icul u al Economics, Vol. 30 No. 3,
pp. 215-228, doi: 10.1016/j.agecon.2002.12.002.
Assa , A.G. and Tsionas, M. (2018), “Bayes ac o s s P- alues”,Tou ism Managemen , Vol. 67,
pp. 17-31, doi: 10.1016/j. ou man.2017.11.011.
Ba, H.A., de Mey, Y., Tho on, S. and Demon , M. (2019), “Inclusi eness o con ac a ming along he
e ical coo dina ion con inuum: e idence om he Vie namese ice sec o ”,Land Use Policy,
Vol. 87, pp. 1-14, doi: 10.1016/j.landusepol.2019.104050.
Asian Jou nal o
Economics and
Banking
121
Downloaded om h p://www.eme ald.com/ajeb/a icle-pd /9/1/105/9699955/ajeb-02-2023-0011.pd by ZBW Ge man Na ional Lib a y o Economics use on 16 Decembe 2025
Balana, B.B. and Oyeyemi, M.A. (2022), “Ag icul u al c edi cons ain s in smallholde a ming in
de eloping coun ies: e idence om Nige ia”,Wo ld De elopmen Sus ainabili y, Vol. 1, pp. 1-12,
doi: 10.1016/j.wds.2022.100012.
Ba slund, M. and Ta p, F. (2008), “Fo mal and in o mal u al c edi in ou p o inces o Vie nam”,
Jou nal o De elopmen S udies, Vol. 44 No. 4, pp. 485-503, doi: 10.1080/00220380801980798.
Beck, T., Deg yse, H., de Haas, R. and an Ho en, N. (2018), “When a m’s leng h is oo a : ela ionship
banking o e he business cycle”,Jou nal o Financial Economics, Vol. 127 No. 1, pp. 174-196,
doi: 10.1016/j.j ineco.2017.11.007.
Bha a h, S., Dahiya, S., Saunde , A. and S ini asan, A. (2011), “Lending ela ionships and loan con ac
e ms”,Re iew o Financial S udies, Vol. 24 No. 4, pp. 1141-1203, doi: 10.1093/ s/hhp064.
Bi can, C. and de Haas, R. (2019), “The limi s o lending: banks and echnology adop ion ac oss
Russia”,Russian Re iew o Financial S udies, Vol. 33 No. 2, pp. 536-609, Fo hcoming, doi: 10.
1093/ s/hhz060.
Boo , A.W.A. (2000), “Rela ionship banking: wha do we know?”,Jou nal o Financial In e media ion,
Vol. 9 No. 1, pp. 7-25, doi: 10.1006/j in.2000.0282.
B iggeman, B.C., Towe, C.A. and Mo eha , M.J. (2009), “C edi cons ain s: hei exis ence, de e minan s,
and implica ions o U.S. a m and non a m sole p op ie o ship”,Ame ican Jou nal o Ag icul u al
Economics, Vol. 91 No. 1, pp. 275-289, doi: 10.1111/j.1467-8276.2008.01173.x.
Bu ka , M. and Ellingsen, T. (2004), “In-kind inance: a heo y o ade c edi ”,Ame ican Economic
Re iew, Vol. 94 No. 3, pp. 569-590, doi: 10.1257/0002828041464579.
Cao, V.H. and Le, K.N. (2020), “Impac o c edi a ioning on capi al alloca ed o inpu s used by ice
a me s in he Mekong Ri e Del a, Vie nam”,Jou nal o Economics and De elopmen , Vol. 22
No. 1, pp. 47-60, doi: 10.1108/jed-11-2019-0067.
Ca le o, C., Sa as ano, S. and Zezza, A. (2013), “Fac o a i ac : he impac o measu emen e o s on
he a m size –p oduc i i y ela ionship”,Jou nal o De elopmen Economics, Vol. 103,
pp. 254-261, doi: 10.1016/j.jde eco.2013.03.004.
Chak aba i, A. (2021), “S a us o women: a compa a i e s udy o emale and male household heads in
India”,China Popula ion and De elopmen S udies, Vol. 4, pp. 405-438, doi: 10.1007/s42379-020-
00065-3.
Chak a a y, S. and Xiang, M. (2013), “In e na ional e idence on discou aged small business”,Jou nal
o Empi ical Finance, Vol. 20, pp. 63-82, doi: 10.1016/j.jemp in.2012.09.001.
Chandio, A.A., Jiang, Y., Wei, F., Rehman, A. and Liu, D. (2017), “Fa me s’access o c edi : does
colla e al ma e o cash low ma e ? –E idence om Sindh, Pakis an”,Cogen Economic and
Finance, Vol. 5, pp. 1-13.
Dao, L.T.A., Nguyen, T.A. and Chandio, A.A. (2023), “Clima e change and i s impac s on Vie nam
ag icul u e: a mac oeconomics pe spec i e”,Ecological In o ma ics, Vol. 74, 101960, doi: 10.
1016/j.ecoin .2022.101960.
De And
es, P., Gimeno, R. and de Cabo, R.M. (2021), “The gende gap in bank c edi access”,Jou nal o
Co po a e Finance, Vol. 71, 101782, doi: 10.1016/j.jco p in.2020.101782.
Deininge , K. and Bye lee, D. (2012), “The ise o la ge a ms in land abundan coun ies: do hey ha e
a u u e?”,Wo ld De elopmen , Vol. 40 No. 4, pp. 701-714, doi: 10.1016/j.wo ldde .2011.04.030.
De agiache, E., Ga ella, P. and Guiso, L. (2000), “Mul iple e sus single banking ela ionship: heo y
and e idence”,Jou nal o Finance, Vol. 55 No. 3, pp. 1133-1161, doi: 10.1111/0022-1082.00243.
Do, X.L. and Baue , S. (2016), “Does c edi access a ec household income homogeneously ac oss
di e en g oups o c edi ecipien s? E idence om u al Vie nam”,Jou nal o Ru al S udies,
Vol. 47, pp. 186-203, doi: 10.1016/j.j u s ud.2016.08.001.
Du hues, T., Heidhues, F. and Buchen iede , G. (2004), “Pa icipa o y p oduc design by using conjoin
analysis in he u al inancial ma ke o No he n Vie nam”,Asian Economic Joun al, Vol. 18
No. 1, pp. 81-114, doi: 10.1111/j.1467-8381.2004.00183.x.
AJEB
9,1
122
Downloaded om h p://www.eme ald.com/ajeb/a icle-pd /9/1/105/9699955/ajeb-02-2023-0011.pd by ZBW Ge man Na ional Lib a y o Economics use on 16 Decembe 2025
Eckel, C., Johnson, C., Mon ma que e, C. and Rojas, C. (2007), “Deb a e sion and he demand o
loans pos seconda y educa ion”,Public Finance Re iew, Vol. 35 No. 2, pp. 233-262, doi: 10.1177/
1091142106292774.
Fe ando, A. and Mulie , K. (2022), “The eal e ec s o c edi cons ain s: e idence om discou aged
bo owe s”,Jou nal o Co po a e Finance, Vol. 73, pp. 1-22, doi: 10.1016/j.jco p in.2022.102171.
Foglia, A., La iola, S. and Reed z, P.M. (1998), “Mul iple banking ela ionships and he agili y o
co po a e bo owe s”,Jou nal o Banking and Finance, Vol. 22 Nos 10-11, pp. 1441-1456, doi: 10.
1016/s0378-4266(98)00058-2.
F eel, M., Ca e , S., Tagg, S. and Mason, C. (2012), “The la en demand o bank deb : cha ac e izing
discou aged bo owe s”,Small Business Economics, Vol. 38 No. 4, pp. 399-418, doi: 10.1007/
s11187-010-9283-6.
Ful on, M.E. and Reynolds, T. (2015), “The poli ical economy o ood p ice ola ili y: he case o
Vie nam”,Ame ican Jou nal o Ag icul u al Economics, Vol. 97 No. 4, pp. 1206-1226, doi: 10.
1093/ajae/aa 019.
Gamma, A.P.M., Dua e, F.D. and Espe ança, J.P. (2017), “Why discou aged bo owe s exis ? An
empi ical ( e)examina ion om less de eloped coun ies”,Eme ging Ma ke s Re iew, Vol. 33,
pp. 19-41, doi: 10.1016/j.emema .2017.08.003.
Gin
e, X. (2011), “Access o capi al in u al Thailand: an es ima ed model o o mal s. in o mal
c edi ”,Jou nal o De elopmen Economics,Vol.96No.1,pp.16-29,doi:10.1016/j.jde eco.2010.
07.001.
Goe z, A.M. and Gup a, R.S. (1996), “Who akes he c edi ? Gende , powe , and con ol o e loan use in
u al c edi p og ams in Bangladesh”,Wo ld De elopmen , Vol. 24 No. 1, pp. 45-63, doi: 10.1016/
0305-750x(95)00124-u.
Gosh, C. and He, F. (2023), “The impac o laws and ins i u ions on inancial con ac s: e idence om
ela ionship lending ac oss he wo ld”,Jou nal o Banking and Finance, Vol. 148, 106741, doi:
10.1016/j.jbank in.2022.106741.
Guiso, L. and Mine i, R. (2010), “The s uc u e o mul iple c edi ela ionships: e idence om U.S.
i ms”,Jou nal o Money, C edi and Banking, Vol. 42 No. 6, pp. 1037-1071, doi: 10.1111/j.1538-
4616.2010.00319.x.
Gu yan, J. and Cha les, K.K. (2013), “Tas e-based o s a is ical disc imina ion: he economics o
disc imina ion e u ns o i s oo s”,The Economic Jou nal, Vol. 123 No. 572, pp. F417-F432, doi:
10.1111/ecoj.12080.
Han, S. (2004), “Disc imina ion in lending: heo y and e idence”,Jou nal o Real Es a e Finance and
Economics, Vol. 29 No. 1, pp. 5-46, doi: 10.1023/b: eal.0000027199.22889.65.
Hu chins, J. (2023), “The US a m c edi sys em and ag icul u al de elopmen : e idence om ea ly
expansion, 1920-1940”,Ame ican Jou nal o Ag icul u al Economics, Vol. 105 No. 1, pp. 3-26,
doi: 10.1111/ajae.12290.
Kahneman, D. and T e sky, A. (1979), “P ospec heo y: an analysis o decision unde isk”,
Econome ica, Vol. 47 No. 2, pp. 263-291, doi: 10.2307/1914185.
Kahneman, D. and T e sky, A. (1992), “Ad ances in p ospec heo y: cumula i e ep esen a ion o
unce ain y”,Jou nal o Risk and Unce ain y,Vol.5No.4,pp.297-323,doi:10.1007/
b 00122574.
Kahneman, D., Kne sch, J.L. and Thale , R.H. (1991), “Anomalies: he endowmen e ec , loss a e sion, and
s a us quo bias”,Jou nal o Economic Pe spec i es, Vol. 5 No. 1, pp. 193-206, doi: 10.1257/jep.5.1.193.
Kass, R.E. and Ra e y, A.E. (1995), “Bayes ac o s”,Jou nal o he Ame ican S a is ical Associa ion,
Vol. 90 No. 430, pp. 773-795, doi: 10.2307/2291091.
Ki os, S. and Meshesha, G.B. (2022), “Fac o s a ec ing a me s’access o o mal inancial c edi in
basona wo ana dis ic , No h Showa Zone, Amha a egional s a e, E hiopia”,Cogen
Economics and Finance, Vol. 10 No. 1, pp. 1-22, doi: 10.1080/23322039.2022.2035043.
Asian Jou nal o
Economics and
Banking
123
Downloaded om h p://www.eme ald.com/ajeb/a icle-pd /9/1/105/9699955/ajeb-02-2023-0011.pd by ZBW Ge man Na ional Lib a y o Economics use on 16 Decembe 2025
Koma ek, A.M., De Pin o, A. and Smi h, V.H. (2020), “A e iew o ypes o isks in ag icul u e: wha
we know and wha we need o know”,Ag icul u al Sys ems, Vol. 178, pp. 1-10, doi: 10.1016/j.
agsy.2019.102738.
Kompas, T., Che, T.N., Nguyen, T.M.H. and Nguyen, H.Q. (2012), “P oduc i i y, ne e u ns, and
e iciency: land and ma ke e o m in Vie namese ice p oduc ion”,Land Economics, Vol. 88
No. 3, pp. 478-495, doi: 10.3368/le.88.3.478.
Kon, Y. and S o ey, D.J. (2003), “A heo y o discou aged bo owe s”,Small Business Economics,
Vol. 21 No. 1, pp. 37-49, doi: 10.1023/a:1024447603600.
Kuhn, L. and Bobojono , I. (2023), “The ole o isk a ioning in u al c edi demand and up ake:
lessons om Ky gyzs an”,Ag icul u al Finance Re iew, Vol. 83 No. 1, pp. 1-20, doi: 10.1108/a -
04-2021-0039.
La ine, M. and Sche ish, M. (1999), “Bayes ac o s: wha hey a e and wha hey a e no ”,The
Ame ican S a is ician, Vol. 53 No. 2, pp. 119-122, doi: 10.1080/00031305.1999.10474443.
Le, K.N. (2021), “Economic ole o educa ion in ag icul u e: e idence om u al Vie nam”,Jou nal o
Economics and De elopmen , Vol. 23 No. 1, pp. 47-58, doi: 10.1108/jed-05-2020-0052.
Le, K.N. and T uong, D.K. (2019), “T ade c edi use by sh imp a me s in Ca Mau p o ince”,Jou nal o
Economics and De elopmen , Vol. 21 No. 2, pp. 270-284, doi: 10.1108/jed-09-2019-0030.
Le enson, A.R. and Willa d, K.L. (2000), “Do i ms ge he inancing hey wan ? Measu ing c edi
a ioning expe ienced by small businesses in he U.S”,Small Business Economics, Vol. 14 No. 2,
pp. 83-94, doi: 10.1023/a:1008196002780.
Louis, T.A. (2005), “In oduc ion o Bayesian me hods II: undamen al concep s”,Clinical T ials, Vol. 2
No. 4, pp. 291-294, doi: 10.1191/1740774505cn099oa.
Migheli, M. (2024), “Land-use igh s and in o mal c edi in u al Vie nam”,I alian Economic Jou nal,
Vol. 10 No. 1, pp. 409-434, doi: 10.1007/s40797-023-00227-5.
Mu o, P. and Pe uzzi, V. (2022), “Rela ionship lending and he use o ade c edi : he ole o ela ional
capi al and p i a e in o ma ion”,Small Business Economics, Vol. 59 No. 1, pp. 327-360, doi: 10.
1007/s11187-021-00537-x.
Nguyen, T.H., Nguyen, M.H., T oege, M. and Nguyen, A.T.H. (2021), “Deb a e sion, educa ion, and
c edi sel - a ioning in SMEs”,Small Business Economics, Vol. 57 No. 3, pp. 1125-1143, doi: 10.
1007/s11187-020-00329-9.
Nguyen, T.T., Nguyen, T.T., Do, M.H., Nguyen, D.L. and G o e, U. (2022), “Shocks, ag icul u al
p oduc i i y, and na u al esou ce ex ac ion in u al Sou heas Asia”,Wo ld De elopmen ,
Vol. 159, 106043, doi: 10.1016/j.wo ldde .2022.106043.
Ogane, Y. (2023), “The numbe o bank ela ionships and bank lending o in o ma ionally opaque
SMEs”,Paci ic-Basin Finance Jou nal, Vol. 80, 102082, doi: 10.1016/j.pac in.2023.102082.
Pe ick, M. (2004), “A mic o-econome ic analysis o c edi a ioning in he Polish a m sec o ”,
Eu opean Re iew o Ag icul u al Economics, Vol. 31 No. 1, pp. 77-101, doi: 10.1093/e ae/31.1.77.
Pham, B.D. and Izumida, Y. (2002), “Ru al de elopmen inance in Vie nam: a mic oeconome ic
analysis o household su eys”,Wo ld De elopmen , Vol. 30 No. 2, pp. 319-335, doi: 10.1016/
s0305-750x(01)00112-7.
Phan, D.K., Gan, C., Na ea, G.V. and Cohen, D.A. (2013), “Fo mal and in o mal u al c edi in he
Mekong Ri e Del a o Vie nam: in e ac ion and accessibili y”,Jou nal o Asian Economics,
Vol. 26, pp. 1-13, doi: 10.1016/j.asieco.2013.02.003.
Popo , A. and Ongena, S. (2011), “In e bank ma ke in eg a ion, loan a es, and i m le e age”,Jou nal
o Banking and Finance, Vol. 35 No. 3, pp. 544-559, doi: 10.1016/j.jbank in.2010.08.011.
Reime s, M. and Klasen, S. (2012), “Re isi ing he ole o educa ion o ag icul u al p oduc i i y”,
Ame ican Jou nal o Ag icul u al Economics, Vol. 95 No. 1, pp. 131-152, doi: 10.1093/ajae/aas118.
Sengup a, R. (2014), “Lending o unc edi wo hy bo owe s”,Jou nal o Financial In e media ion,
Vol. 23 No. 1, pp. 101-128, doi: 10.1016/j.j i.2013.07.001.
AJEB
9,1
124
Downloaded om h p://www.eme ald.com/ajeb/a icle-pd /9/1/105/9699955/ajeb-02-2023-0011.pd by ZBW Ge man Na ional Lib a y o Economics use on 16 Decembe 2025
S igli z, J.E. and Weiss, A. (1981), “C edi a ioning in ma ke s wi h impe ec in o ma ion”,Ame ican
Economic Re iew, Vol. 71, pp. 393-410.
Taue , L. (1995), “Age and a me p oduc i i y”,Re iew o Ag icul u al Economics, Vol. 17 No. 1,
pp. 63-69, doi: 10.2307/1349655.
T an, H.Q. (2018), “Land accumula ion in he Mekong Ri e Del a o Vie nam: a ques ion e isi ed”,
Canadian Jou nal o De elopmen S udies, Vol. 39 No. 2, pp. 199-214.
T an, T.K.V., Elahi, E., Zhang, L., Abid, M., Pham, Q.T. and T an, T.D. (2018), “Gende di e ences in
o mal c edi app oaches: u al household in Vie nam”,Asian-Paci ic Economic Li e a u e,
Vol. 32 No. 1, pp. 131-138, doi: 10.1111/apel.12220.
Ullah, R., Shi ako i, G.P. and Ali, G. (2015), “Fac o s a ec ing a me s’ isk a i ude and isk
pe cep ions: he case o Khybe Pakh unkhwa, Pakis an”,In e na ional Jou nal o Disas e Risk
Reduc ion, Vol. 13, pp. 151-157, doi: 10.1016/j.ijd .2015.05.005.
Van de Schoo , R., Depaoli, S., King, R., K ame , B., Ma ens, K., Tadesse, M.G., Vannucci, M., Gelman,
A., Veen, D., Willemsen, J. and Yau, C. (2021), “Bayesian s a is ics and modelling”,Na u e
Re iews Me hods P ime s, Vol. 1, p. 1, doi: 10.1038/s43586-020-00001-2.
Co esponding au ho
Le Khuong Ninh can be con ac ed a : [email p o ec ed]
Fo ins uc ions on how o o de ep in s o his a icle, please isi ou websi e:
www.eme aldg ouppublishing.com/licensing/ ep in s.h m
O con ac us o u he de ails: [email p o ec ed]
Asian Jou nal o
Economics and
Banking
125
Downloaded om h p://www.eme ald.com/ajeb/a icle-pd /9/1/105/9699955/ajeb-02-2023-0011.pd by ZBW Ge man Na ional Lib a y o Economics use on 16 Decembe 2025
