Fe ando, Annalisa; Pál, Rozália
Wo king Pape
A e inancing condi ions a h ea o SMEs' pe o mance,
g ow h, and ans o ma ion?
ADBI Wo king Pape , No. 1467
P o ided in Coope a ion wi h:
Asian De elopmen Bank Ins i u e (ADBI), Tokyo
Sugges ed Ci a ion: Fe ando, Annalisa; Pál, Rozália (2024) : A e inancing condi ions a h ea
o SMEs' pe o mance, g ow h, and ans o ma ion?, ADBI Wo king Pape , No. 1467, Asian
De elopmen Bank Ins i u e (ADBI), Tokyo,
h ps://doi.o g/10.56506/CCHD8076
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/305427
S anda d-Nu zungsbedingungen:
Die Dokumen e au EconS o dü en zu eigenen wissenscha lichen
Zwecken und zum P i a geb auch gespeiche und kopie we den.
Sie dü en die Dokumen e nich ü ö en liche ode komme zielle
Zwecke e iel äl igen, ö en lich auss ellen, ö en lich zugänglich
machen, e eiben ode ande wei ig nu zen.
So e n die Ve asse die Dokumen e un e Open-Con en -Lizenzen
(insbesonde e CC-Lizenzen) zu Ve ügung ges ell haben soll en,
gel en abweichend on diesen Nu zungsbedingungen die in de do
genann en Lizenz gewäh en Nu zungs ech e.
Te ms o use:
Documen s in EconS o may be sa ed and copied o you pe sonal
and schola ly pu poses.
You a e no o copy documen s o public o comme cial pu poses, o
exhibi he documen s publicly, o make hem publicly a ailable on he
in e ne , o o dis ibu e o o he wise use he documen s in public.
I he documen s ha e been made a ailable unde an Open Con en
Licence (especially C ea i e Commons Licences), you may exe cise
u he usage igh s as speci ied in he indica ed licence.
h ps://c ea i ecommons.o g/licenses/by-nc-nd/3.0/igo/
ADBI Wo king Pape Se ies
ARE FINANCING CONDITIONS A
THREAT TO SMES’ PERFORMANCE,
GROWTH, AND TRANSFORMATION?
Annalisa Fe ando and Rozália Pál
No. 1467
July 2024
Asian De elopmen Bank Ins i u e
The Wo king Pape se ies is a con inua ion o he o me ly named Discussion Pape se ies;
he numbe ing o he pape s con inued wi hou in e up ion o change. ADBI’s wo king
pape s e lec ini ial ideas on a opic and a e pos ed online o discussion. Some wo king
pape s may de elop in o o he o ms o publica ion.
Sugges ed ci a ion:
Fe ando, A. and R. Pál. 2024. A e Financing Condi ions a Th ea o SMEs’ Pe o mance,
G ow h, and T ans o ma ion? ADBI Wo king Pape 1467. Tokyo: Asian De elopmen Bank
Ins i u e. A ailable: h ps://doi.o g/10.56506/CCHD8076
Please con ac he au ho s o in o ma ion abou his pape .
Email: [email p o ec ed]
Annalisa Fe ando is a senio lead economis a he Eu opean Cen al Bank, F ank u
am Main, Ge many. Rozália Pál is a senio economis a he Eu opean In es men Bank,
Luxembou g.
The iews exp essed in his pape a e he iews o he au ho and do no necessa ily
e lec he iews o policies o ADBI, ADB, i s Boa d o Di ec o s, o he go e nmen s
hey ep esen . ADBI does no gua an ee he accu acy o he da a included in his pape
and accep s no esponsibili y o any consequences o hei use. Te minology used may
no necessa ily be consis en wi h ADB o icial e ms.
Discussion pape s a e subjec o o mal e ision and co ec ion be o e hey a e inalized
and conside ed published.
We would like o hank Eddie Ge ba and he pa icipan s o he wo kshop on SME
pe o mance unde unce ain y a ADBI o hei use ul commen s on a p e ious e sion
o his pape . The opinions exp essed he ein a e hose o he au ho s and do no
necessa ily e lec hose o he Eu opean In es men Bank o he Eu opean Cen al
Bank. The usual disclaime s apply.
Asian De elopmen Bank Ins i u e
Kasumigaseki Building, 8 h Floo
3-2-5 Kasumigaseki, Chiyoda-ku
Tokyo 100-6008, Japan
Tel: +81-3-3593-5500
Fax: +81-3-3593-5571
URL: www.adbi.o g
E-mail: [email p o ec ed]
© 2024 Asian De elopmen Bank Ins i u e
ADBI Wo king Pape 1467 Fe ando and Pál
Abs ac
A ailabili y o in e nal and ex e nal inancing sou ces has an impac on i ms’ in es men s
and g ow h. E en p o i able i ms wi h su icien inancing sou ces in no mal imes can
be a ec ed by demand and supply shocks such as he COVID-19 lockdown, he ene gy
c isis, o he ecen igh ening o inancing condi ions. This pape analyzes he impac
o unding di icul ies on i ms’ in es men , pe o mance, and g ow h du ing no mal pe iods
and pe iods o ex e nal shocks using a eg ession adjus men ea men e ec app oach.
We di e en ia e among s uc u al ba ie s o ex e nal inancing and cyclical wo sening o
inancing condi ions, con olling o o he majo in es men ba ie s. We use su ey da a
collec ed om he i s o he eigh h in age o he Eu opean In es men Su ey (EIBIS).
The empi ical e idence shows ha mic o and small i ms and leading inno a o s a e
pa icula ly ulne able o de e io a ing unding condi ions. Resul s indica e ha i ms’ lagging
in digi aliza ion and g een in es men s a e acing mo e a s uc u al a he han cyclical
inancing issue. Consequen ly, policy suppo should be o ien ed owa ds hose s uc u al
inancing impedimen s.
Keywo ds: SMEs, in es men gap, ex e nal unding, in e nal unding, inancing cons ain s,
unce ain y, in es men ba ie s, i m pe o mance, g ow h, digi al and g een ansi ion
JEL Classi ica ion: C83, D22, G32
ADBI Wo king Pape 1467 Fe ando and Pál
Con en s
1. INTRODUCTION AND RESEARCH BACKGROUND ................................................. 1
2. DATA ............................................................................................................................ 3
3. ECONOMETRIC FRAMEWORK .................................................................................. 5
4. EMPIRICAL RESULTS ................................................................................................ 7
5. CONCLUSION AND POLICY IMPLICATIONS .......................................................... 15
REFERENCES ...................................................................................................................... 16
ANNEX ................................................................................................................................... 18
ADBI Wo king Pape [Do no en e ] Fe ando and Pál
1
1. INTRODUCTION AND RESEARCH BACKGROUND
The g ow h o SMEs is much mo e condi ioned by access o inance han ha o la ge
i ms (Fouejieu, Ndoye, and Sydo enko 2020). As i ms can g ow i hey in es and his
o en equi es inancial sou ces beyond hei income and sa ings, SMEs a e less likely
o be able o ob ain he addi ional ex e nal inancing hey need ega dless o he
po en ial o hei in es men p ojec s. Thei sou ces o unds a e o en limi ed o hei
in e nal unds and inancial suppo om iends and amily (A dic e al. 2013). Thei
inancing si ua ion can be wo sened in pe iods o ex e nal shocks, such as he sudden
d op o sales caused by he COVID-19 lockdowns, he ecen inc eases in p oduc ion
cos s due o high ene gy p ices, and he igh ening o ex e nal inancing condi ions
h ough in e es a e hikes o s ic e lending condi ions.
The missed in es men oppo uni ies caused by he limi ed amoun o unds a e no
only de imen al in he sho e m, especially in he p esen pe iod o s uc u al shi s
owa ds digi aliza ion and g eening, bu hey migh ha e longe - e m implica ions o
he g ow h, p oduc i i y, and compe i i eness o he Eu opean economy. The e is
e idence also ha he p omising high g ow h en e p ises ha gene a es mos o he
employmen g ow h in he economy ace inancing cons ain s and seek o al e na i e
inancing, addi ionally o he adi ional bank loans (Fe ando, Pál, and Du an e 2019).
Mo eo e , he g eening o he economy coincides wi h a simila ace o adap digi al
solu ions a he i m le el and hese ends a e mu ually suppo i e. Su ey-based
analyses o Eu opean companies show ha digi al i ms we e mo e esilien du ing he
COVID-19 pandemic, and hey we e less likely o educe employmen (Coad e al.
2023, EIB 2023). Fu he mo e, he poo deg ee o digi aliza ion among SMEs and
he digi aliza ion gap be ween digi al and nondigi al i ms widened o SMEs amid
ex e nal shocks (Te uel e al. 2023). Wi hou in es men s, such a digi al di ide migh
inc ease u he .
This pape in es iga e how inancing di icul ies a ec i ms’ in es men decisions o
di e en i m g oups. We show ha he e is a g ea e isk o missed in es men
oppo uni ies o small-sized and inno a i e i ms han o la ge companies, due o hei
ela i ely highe ulne abili y in pe iods o cyclical de e io a ion, which goes beyond he
s uc u ally high accumula ed in es men gap. Wi h ega d o he g een and digi al
ans o ma ion, we also ind e idence ha hose lagging behind in adop ing digi al
solu ions and hose in es ing o planning o in es in g een solu ions a e s uc u ally
acing mo e di icul ies in accessing ex e nal inancing. Howe e , we do no ind a
signi ican di e ence in e ms o inancing de e io a ion du ing c ises/ igh ening pe iods
be ween ans o ming and non ans o ming i ms.
Mo eo e , we show ha he inancing condi ions o i ms s ongly impac hei
in es men s and ha e a s ong in luence on he ollowing wo yea s’ pe o mance
and g ow h.
The e a e se e al s udies in he li e a u e ha ackle he impac o inancial cons ain s
on in es men , employmen , and inno a ion, e.g., Bu le and Co naggia (2011) and
Ga cía-Posada (2019) show ha i ms ope a ing in en i onmen s wi h lowe inancial
cons ain s a e able o in es mo e and inc ease p oduc ion. Fe nandes and Fe ei a
(2017) and Duygan-Bump, Le ko , and Mon o iol-Ga iga (2015) s ess he nega i e
e ec o inancial cons ain s on employmen . Fe ando and Ruggie i (2018) ind
ha lowe inancial cons ain s imp o e labo p oduc i i y, while Go odnichenko
and Schni ze (2013) documen he nega i e e ec o inancial cons ain s on a
i m’s inno a ion.
ADBI Wo king Pape [Do no en e ] Fe ando and Pál
2
By con as , he pape o Bonanno, Fe ando, and Rossi (2023) highligh s a di e en
ole o inancial cons ain s on i ms’ beha iou . Looking a he in e play be ween i ms’
e iciency, inno a ion, and access o inance, he au ho s ind ha inancially
cons ained i ms end o imp o e hei e iciency o educe hei isk o ailu e and o
main ain p o i s, ega dless o he sec o al disagg ega ion. Howe e , when inancial
cons ain s a e binding, i ms in low- ech sec o s a e induced o be mo e e icien han
hose in high- ech sec o s o enhance hei p o i abili y.
The li e a u e also shows ha inancial cons ain s can ha e ampli ied e ec s du ing
cycles: Aghion e al. (2012) ind ha o c edi -cons ained i ms, R&D in es men
plumme s du ing ecessions and in es men does no inc ease p opo ionally back
du ing up u ns, while Musso and Schia o (2008) ind ha in he p esence o
asymme ic in o ma ion and inancial cons ain s, e en small shocks may ampli y
business cycle luc ua ions.
Ou pape is close o bo h ends o li e a u e in es iga ing he ole o inancing ic ions
du ing no mal imes and pe iods o economic down u n, as p esen ed abo e, bu
we a e able o analyse hese wo ends simul aneously. Mo eo e , in his pape , we
p o ide deepe insigh s in o he e ec o in e nal and ex e nal unding condi ions and
ypes o i m cha ac e is ic s ha i migh imp o e o u he de e io a e he in es men
end amid he gi en unding condi ion. To analyze he in es men beha io o i ms in
ela ion o unding di icul ies, we c oss-check al e na i e a iables like a longe - e m
accumula ed in es men gap, o he d opping o in es men p ojec s ecen ly, planned
in es men d ops, and changes in he in es men a e.
We also show ha he p esence o ex e nal inance di icul ies conside ably a ec s
p o i abili y and asse g ow h on a e age wo yea s la e . In e es ingly, e en i ms wi h
no accumula ed in es men gap o planned in es men d op a e a ec ed in he long
e m by ex e nal unding condi ions, esul ing in lowe pe o mance and g ow h.
F om a policy pe spec i e, ou analysis poin s o he impo ance o i m-le el policy
suppo bo h in no mal imes and in pe iods o c isis and ansi ion, no only o
sho - e m su i al and s abiliza ion bu also o longe - e m a ge s. Recen ly, i
has been shown ha i ms ha bene i ed om he policy suppo du ing he
COVID-19 pandemic end o be mo e op imis ic abou hei in es men plans,
especially hose in digi al echnologies (Ha asz osi e al. 2022). In his pape , we a gue
mo e gene ally ha policy make s should p o ide special a en ion o SMEs and
leading inno a o s no only due o hei s uc u ally highe ex e nal inancial cons ain s
bu also because o he impac o cyclical de e io a ions o inancing condi ions.
Wi h espec o he g een and digi al ans o ma ion, we ound less e idence o he
impac o c isis. This is in line wi h p e ious indings sugges ing ha ecen c ises migh
ac as a push o digi aliza ion, especially du ing he COVID-19 c isis, and o g eening
and inc easing ene gy e iciency, especially du ing he ene gy p ice shock, as a
s a egy o su i al. Ne e heless, we do ind signi ican esul s indica ing g ea e
unding di icul ies s uc u ally ac oss yea s among hose i ms lagging in digi aliza ion
and g een in es men s.
The pape p oceeds as ollows. The nex wo sec ions desc ibe he da a and he
empi ical me hodology. Sec ion 4 ou lines he empi ical esul s, while Sec ion 5
concludes and p esen s some policy ecommenda ions.
ADBI Wo king Pape [Do no en e ] Fe ando and Pál
3
2. DATA
Fo ou analysis, we ely on he pooled i s o eigh h in ages o he Eu opean
In es men Bank In es men Su ey ( he EIBIS 2023), combined wi h Moody’s ORBIS
da abase. The EIBIS da abase con ains in o ma ion on mo e han 12,000 non inancial
i ms annually in he EU o he pe iod 2016–2023. EIBIS is an EU-wide su ey ha
ga he s quali a i e and quan i a i e in o ma ion on in es men ac i i ies by non inancial
co po a ions, bo h SMEs (5–250 employees) and la ge co po a es (250+ employees),
hei inancing equi emen s, and he di icul ies hey ace when unning hei business.
Using s a i ied sampling, EIBIS is in ended o be ep esen a i e ac oss all 27 membe
s a es o he EU, wi hin coun ies, ou i m size classes (mic o, small, medium, and
la ge), and ou sec o g oupings (manu ac u ing, se ices, cons uc ion, and
in as uc u e).1 Fo each i m, he su ey eplies a e linked o in o ma ion de i ed om
he annual inancial s a emen s ob ained om Moody’s ORBIS da abase.
Mo e impo an ly o ou analysis, he su ey con ains in o ma ion abou changes
in in e nal and ex e nal inancing condi ions, di icul ies in ob aining any ex e nal
inancing, long- e m in es men ba ie s, ypes o in es men ( anging om ixed
angible asse s o in angible asse s and inno a ion ypes new o i ms, new o
companies, o new o he global ma ke s), ealized in es men s, accumula ed
in es men gaps compa ed o hei needs/oppo uni ies, and in es men plans o he
nea u u e. The al e na i e in es men a iables enable us o conduc a p o ound
analysis o he inancing-in es men ela ionship. Mo eo e , in e ms o he link o
he inancial s a emen s o i ms, we a e able o check he pe o mance and g ow h
dynamics o i ms.
Ou main a iables o in e es a e he encompassing indica o s o ex e nal and in e nal
inancing di icul ies. Fo he indica o o ex e nal unding di icul ies, we also de ine
sepa a ely he cyclical and he s uc u al unding di icul ies. The s uc u al ba ie s o
i ms’ access o inance a e indica ed by he le el o he de elopmen o he inancial
sec o s and o he i m-speci ic cha ac e is ics, like hei anspa ency, c edibili y, le el
o angible asse s, p o i abili y, e c. To cap u e such elemen s on he supply side o
inancing, in his ca ego y we conside hose iable i ms ha needed a loan bu we e
ei he discou aged o ejec ed ( ully cons ained) o ecei ed less han hey needed
(quan i y cons ain s) o i was oo expensi e (p ice cons ain s). We check iabili y
by no egis e ing losses o h ee consecu i e yea s, o be ce ain ha cons ain s a e
due o ex e nal inancing condi ions and no o he weak inancial pe o mance o
he i m. To elimina e he ime- a ying cyclical componen o his a iable, we ake he
a e age alue o he i m le el a iable ac oss yea s. We cap u e sepa a ely he
igh ening inancial cycles, ega dless o he le el o de elopmen and cha ac e is ics o
he inancial sys em and i ms. Fo his, we conside he pe cep ion o i ms ega ding
he changes in hei ex e nal inancing condi ions. To elimina e in e nal (like he
success ulness/ iabili y o he business s a egy) e sus ex e nal d i e s, we exclude
om his ca ego y i ms ha egis e losses o h ee consecu i e yea s ( inancially
weak i ms ega dless o he cycle). The encompassing ex e nal unding di icul ies
indica o combines hese wo sou ces o s uc u al and cyclical inancing impasse.
Figu e 1 p esen s bo h he s uc u al and cyclical ex e nal unding di icul ies a iables
o he inancial pe iod 2015–2022 co e ed by he 2016–2023 EIBIS wa es.
1 EIBIS has been shown o be a eliable da a sou ce wi h no sys ema ic sampling bias (B u sche e al.
2020).
ADBI Wo king Pape [Do no en e ] Fe ando and Pál
4
Figu e 1: Ex e nal Funding Di icul ies – by S uc u al and Cyclical Componen s
Ou second main a iable, he in e nal unding di icul ies, is de ined as hose iable
i ms decla ing ha hei in e nal inance condi ions ha e de e io a ed. To exclude he
impac o i ms wi h inancial p oblems (zombie i ms), which a e less p obable o
in es , we conside only hose ha do no egis e losses o h ee consecu i e yea s.
Fu he mo e, ou analysis elies on se e al a iables om EIBIS 2016–2023 and O bis
2015–2022 as desc ibed in Table 1 in he Appendix. We dis inguish ou di e en
dependen a iables:
1) in es men gap;
2) d op in ealized in es men ;
3) d op in planned in es men ;
4) ne in es men a e.
The i s h ee a iables a e de i ed om he su ey esponses and a e cons uc ed as
dummies. Each is equal o 1 i : (1) i ms decla e ha in es men o e he las h ee
yea s was oo li le o ensu e he success o hei business going o wa d (in es men
gap); (2) i ms epo less in es men han in he p e ious yea ( ealized in es men
d op); 3) he o al in es men spen in he cu en o nex yea is expec ed o be less
han in he p e ious yea (planned in es men d op). The las a iable is he ne
in es men a e, which is de ined as he di e ence in ixed asse s be ween wo
subsequen yea s, o e lagged ixed asse s.
Table A1 also includes he de ini ions o se e al con ol a iables, like size classes,
sec o s, and a se o inancial a ios (le e age, p o i abili y, cash holding), as well as
se e al dummy a iables on he obs acles o in es men ac i i ies. We also con ol o
digi al and g een in es men s. Table 1 displays se e al cha ac e is ics o he i ms in
ou da ase . A ound 14% o i ms epo ha ing some ex e nal unding di icul ies, while
12% say ha hese di icul ies a e ela ed o cyclical condi ions and a ound 5% o
s uc u al issues (some o e lap is likely be ween he wo a iables). The pe cen age o
i ms signaling in e nal unding di icul ies is sligh ly lowe (12%). An in es men gap is
epo ed by 15% o i ms, while he di e ence be ween he d op in ealized in es men
(21%) and he d op in planned in es men (27%) is 6 pe cen age poin s.
0%
5%
10%
15%
20%
25%
2015 2016 2017 2018 2019 2020 2021 2022
s uc u al cyclical
ADBI Wo king Pape [Do no en e ] Fe ando and Pál
11
Table 6: De e minan s o In es men – Impac o Ex e nal/In e nal Funding
De e io a ion and Inc ease in Unce ain y, Ma ginal E ec s
(1)
(2)
Va iables
In es men Gap
Planned In es men D op
Ex e nal unding di icul ies (lag)
0.041***
0.032***
(0.008)
(0.008)
In e nal unding di icul ies (lag)
0.058***
0.148***
(0.009)
(0.009)
Cash holdings (lag)
–0.064***
–0.063***
(0.023)
(0.020)
P o i abili y (lag)
–0.309***
–0.017
(0.031)
(0.022)
Financial le e age (lag)
–0.003
0.100***
(0.016)
(0.014)
leading_inno a o s1
–0.020
–0.034***
(0.013)
(0.011)
Small
–0.019*
–0.044***
(0.011)
(0.009)
Medium
–0.043***
–0.072***
(0.011)
(0.009)
La ge
–0.052***
–0.110***
(0.011)
(0.010)
Cons uc ion
–0.026***
–0.025***
(0.009)
(0.008)
Se ices
–0.027***
–0.012
(0.008)
(0.008)
In as uc u e
0.003
–0.044***
(0.008)
(0.007)
Sou h
–0.084***
–0.018***
(0.007)
(0.007)
Wes and No h
–0.058***
0.017***
(0.008)
(0.006)
D2020
–0.010
0.178***
(0.009)
(0.008)
Obs acle – unce ain y (lag)
0.032***
0.052***
(0.009)
(0.007)
Obs acle – lack o demand
0.012*
0.028***
(0.007)
(0.006)
Obs acle – lack o skilled s a
0.020**
–0.021***
(0.008)
(0.007)
Obs acle – digi al in as uc u e
0.007
–0.009
(0.007)
(0.006)
Obse a ions
15,000
30,967
S anda d e o s in pa en heses.
*** p < 0.01, ** p < 0.05, * p < 0.1.
ADBI Wo king Pape [Do no en e ] Fe ando and Pál
12
Wi h ega d o i m cha ac e is ics, we ind ha i ms wi h g ea e inancial le e age a e
mo e likely o in es less, while cash sa ings and p o i abili y clea ly se e as a posi i e
bu e o in es men s. Leading inno a o s a e less likely o dec ease hei planned
in es men s, despi e he ela i ely wo se ex e nal unding condi ions. This can be
explained by he s ong eliance on in e nal o al e na i e sou ces as well as on he ac
ha R&D in es men s, once launched, a e planned o a longe pe iod o mul iple
yea s. Ac oss size classes, la ge -sized i ms ha e a signi ican ly lowe accumula ed
in es men gap han mic o i ms. Fu he , hey a e also less likely o d op hei u u e
in es men s han mic o/small i ms. Unce ain y, a lack o demand, and a lack o skilled
s a a e signi ican ba ie s o in es men s, hus accumula ing in es men gaps.
Ne e heless, i ms wi h a lack o skilled s a a e mo e esilien in e ms o in es men
plans. This is in line wi h he idea ha a lack o skilled s a epo ed by i ms as an
obs acle o in es men may be hough o incen i ize o he ypes o in es men s, o
ins ance in digi al o AI o subs i u e o skilled wo ke s, a he han in ainings o
enhance he skill o al eady employed wo ke s. Howe e , in o he s udies (EIB 2024) i
is ound ha high-g ow h i ms, which in es ela i ely mo e han a e age i ms, a e
mo e o en cons ained by he una ailabili y o quali ied wo ke s.
In an addi ional exe cise, we spli he ex e nal unding di icul ies in o s uc u al and
cyclical componen s. As p e iously s a ed, among i ms ha show s uc u al di icul ies
in ob aining ex e nal inance, we conside iable i ms ha needed a loan bu we e
ei he discou aged, did no ecei e i , o ecei ed less han hey needed. We g oup in
he ca ego y o i ms signaling cyclical di icul ies hose ha , ega dless o hei cu en
ex e nal inancing possibili ies, belie e ha ex e nal inancing condi ions a e wo sening.
In he econome ic analysis p esen ed in Table 7, he in es men gap is likely o
inc ease due o bo h s uc u al and cyclical componen s, wi h a highe impac o he
o me (Column 1). The es ima ed d op in planned in es men indica ed in Column 2 is
d i en ins ead by he cyclical componen , while he nega i e coe icien o he s uc u al
componen ends o mi iga e he o e all impac . This we may explain by he ac
ha hose i ms ha s uc u ally ace di icul ies in accessing ex e nal inancing ely
mo e, o exclusi ely, on in e nal inancing, so hey a e less indeb ed in gene al and
consequen ly less impac ed by igh ening o he inancing condi ions.
Impac o Ex e nal Financing Di icul ies on Pe o mance and G ow h
In his sec ion we p esen he esul s on he ex pos e ec o he p esence o ex e nal
unding di icul ies on i m pe o mance and g ow h. The p opensi y sco e is un using
he speci ica ion desc ibed in Sec ion 3. Table 8 shows he dis ibu ion o he sample
ac oss i ms wi h ex e nal unding di icul ies and in es men gaps (Panel a) and d ops
in planned in es men (Panel b).
ADBI Wo king Pape [Do no en e ] Fe ando and Pál
13
Table 7: De e minan s o In es men Gap/Planned In es men D op o Iden i y
he Impac o S uc u al Ex e nal Financing Di icul ies, Cyclical Funding
De e io a ion, and In e nal Funding De e io a ion
(1)
(2)
Va iables
In gap
Planned In D op
S uc u al ex e nal inancing di icul ies
0.090***
–0.042***
(0.010)
(0.010)
Cyclical ex e nal inancing di icul ies
0.035***
0.047***
(0.009)
(0.009)
In e nal unding de e io a ion (lag)
0.058***
0.146***
(0.009)
(0.009)
Cash holdings (lag)
–0.066***
–0.060***
(0.023)
(0.020)
P o i abili y (lag)
–0.297***
–0.022
(0.031)
(0.022)
Financial le e age (lag)
–0.017
0.111***
(0.017)
(0.014)
leading_inno a o s1
–0.022*
–0.033***
(0.013)
(0.011)
Small
–0.017
–0.044***
(0.011)
(0.009)
Medium
–0.040***
–0.073***
(0.011)
(0.009)
La ge
–0.048***
–0.111***
(0.011)
(0.010)
Cons uc ion
–0.027***
–0.025***
(0.009)
(0.008)
Se ices
–0.026***
–0.012
(0.008)
(0.008)
In as uc u e
0.002
–0.043***
(0.008)
(0.007)
Sou h
–0.082***
–0.020***
(0.007)
(0.007)
Wes and No h
–0.053***
0.014**
(0.008)
(0.006)
D2020
–0.009
0.176***
(0.009)
(0.008)
Obs acle – unce ain y (lag)
0.031***
0.052***
(0.009)
(0.007)
Obs acle – lack o demand
0.011
0.028***
(0.007)
(0.006)
Obs acle – lack o skilled s a
0.021***
–0.021***
(0.008)
(0.007)
Obs acle – digi al in as uc u e
in e nal unding di icul ies (lag)
0.008
–0.009
(0.007)
(0.006)
Obse a ions
14,992
30,944
S anda d e o s in pa en heses.
ADBI Wo king Pape [Do no en e ] Fe ando and Pál
14
Table 8: Sample Dis ibu ion o Fi ms wi h Ex e nal Financing Di icul ies
and In es men Issues
Panel A: In es men Gap
In es men Gap
0
1
To al
Ex e nal unding di icul ies
0
68%
14%
82%
1
13%
4%
18%
To al
81%
19%
100%
Panel b: Planned In es men D op
Planned In es men D op
0
1
To al
Ex e nal unding di icul ies
0
60%
22%
82%
1
11%
6%
18%
To al
71%
29%
100%
We es ima e he p opensi y sco e, which esul s om he condi ional p obabili y o
a i m signaling ex e nal unding di icul ies, gi en he alue o he obse ed i ms’
cha ac e is ics and di e en subsamples o wi h o wi hou in es men di icul ies
(in es men gap o planned in es men d op). The se o explana o y a iables chosen
mus sa is y he balancing p ope y, which equi es ha a e he ma ching, he
dis ibu ions o he co a ia es and he p opensi y sco e be ween he ea ed and he
con ol g oups a e simila . Figu e 4 con i ms ha he p opensi y sco e dis ibu ion a e
he ma ching is simila o he ea ed and con ol g oups.
Figu e 4: P opensi y Sco e Dis ibu ion Be o e and A e he Ma ching
Sou ce: EIBIS-O bis 2016–2023.
Table 9 displays ou main indings on he (ex pos ) e ec o access o ex e nal inance
on a i m’s g ow h based on he p opensi y sco e analysis. F om Column 1 o Table 9,
we see ha he p esence o ex e nal unding di icul ies has a nega i e and s a is ically
signi ican impac on he subsequen p o i abili y and g ow h. Fi ms ha aced
di icul ies in ob aining ex e nal inancing we e less p o i able (–1.26 pe cen age poin s)
and g ew ela i ely less (1.1%) han i ms ha did no ace his kind o p oblem. In he
i h column we see ha he losses in e ms o p o i abili y a e e en highe among he
subg oup o i ms ha epo ha ing had some in es men gaps in he pas (–1.74
pe cen age poin s). By con as , he dis inc ion be ween i ms wi h and wi hou u u e
plans o d op in es men does no add any u he addi ional in o ma ion on u u e
ADBI Wo king Pape [Do no en e ] Fe ando and Pál
15
pe o mance. We should bea in mind ha he e is a gap be ween planned and ealized
in es men gaps; i ms end o be mo e pessimis ic when decla ing hei plans.
In e ms o u u e asse g ow h, he loss is lowe o i ms ha , on op o inancial
p oblems, also signaled no pas in es men gap (Column 3). This inding suppo s
he idea ha hese i ms, ha ing al eady expanded acco ding o hei business needs
in he p e ious yea s, migh hold back om in es ing u he when ex e nal inance
is no easily a ailable. Resul s migh also sugges ha besides he di ec impac o
unds a ailabili y o in es men s, he e is a signi ican nega i e impac o ex e nal
inancing di icul ies on pe o mance and g ow h. Ne e heless, he coe icien o his
second g oup is no signi ican , gi en he low sample size (4% and 6% o he o al
sample, espec i ely).
Table 9: Di e en ial G ow h Ra es o Fi ms Wi h Funding Gap Ve sus Fi ms
Wi h No Funding Gap by In es men Decisions – P opensi y Sco e Resul s
( o Mo e De ails See Table A3 in he Appendix)
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
2-yea A e age
G ow h Ra e
All
Fi ms
T-s a
No
In es men
Gap
T-s a
Wi h
In es men
Gap
T-s a
No Planned
In es men
D op
T-s a
Planned
In es men
D op
T-s a
ROA
–1.26
–6.14
–1.23
5.1
–1.74
3.53
–1.214
4.4
–1.206
3.52
To al Asse s
–.011
–2.56
–0.02
2.83
–0.01
1.28
–0.02
2.3
–0.01
1.52
5. CONCLUSION AND POLICY IMPLICATIONS
In his pape we p o ide no el e idence o he nega i e impac o in e nal and ex e nal
unding condi ions on i ms’ in es men s and g ow h. We documen ha mic o and
small i ms and leading inno a o s a e mo e likely o ace bo h in e nal and ex e nal
unding di icul ies, especially in cyclically wo sening pe iods.
We show ha he p esence o ex e nal unding di icul ies has a long- e m impac on
u u e i ms’ pe o mance and g ow h. Fi ms indeed ha e mo e di icul ies gene a ing
addi ional inancial lows in hei in es men in o al asse s in he subsequen wo yea s
when hey had p e ious di icul ies collec ing ex e nal inance. The losses a e e en
highe when he same i ms had also signaled some in es men gaps in he pas .
We a gue ha policy suppo should ocus on i ms ha a e mo e ulne able o
igh ening and de e io a ing unding condi ions, especially i in e nal and ex e nal
unding condi ions a e de e io a ing simul aneously, which has been he case ecen ly
o many mic o and small i ms. Viable i ms, e en wi h high g ow h po en ial, and
leading inno a o s migh be o ced o s agna e by canceling hei in es men s due o a
lack o , o una o dable, unding. Inc easing he al e na i e inancing solu ions would i
pa icula ly he inancing need o small and inno a i e i ms, especially hose wi h high
po en ial o g ow.
In he cu en pe iod o s uc u al shi s owa ds digi aliza ion and g eening, inancing
condi ions migh play an impo an ole in ans o ming Eu opean i ms. Resul s
indica e mo e o a s uc u al a he han a cyclical inancing issue among i ms ha a e
lagging in digi aliza ion and g een in es men s. Consequen ly, policy suppo should be
o ien ed owa ds hose s uc u al impedimen s ha p e en i ms om ans o ming.
Ta ge ed policy suppo o hese speci ic in es men s is needed o close he
digi aliza ion and g eening gap among EU i ms, he eby accele a ing he g een and
ai ansi ion.
ADBI Wo king Pape [Do no en e ] Fe ando and Pál
16
REFERENCES
Abadie A., D. D ukke , L. J. He , and G. W. Imbens. 2001. Implemen ing Ma ching
Es ima o s o A e age T ea men E ec s in S a a. S a a Jou nal 1(1): 1–18.
Aghion, P., N. Be man, L. Eyma d, P. Askenazy, and G. Ce e. 2012. C edi
Cons ain s and he Cyclicali y o R&D In es men : E idence om F ance.
Jou nal o he Eu opean Economic Associa ion 10(5): 1001–1002.
A dic Alpe , O. P., M. Hommes, P. S ein, and W. Benno. 2013. Closing he C edi Gap
o Fo mal and In o mal Mic o, Small, and Medium En e p ises. Washing on,
D.C.: Wo ld Bank, Janua y 2013 G oup.
Bonanno, G., A. Fe ando, and S. P. S. Rossi. 2023. Do Inno a ion and Financial
Cons ain s A ec he P o i E iciency o Eu opean En e p ises? Eu asian
Business Re iew 13: 57–86.
B u sche , P.-B., A. Coali, J. Delano e, and P. Ha asz osi.2020. EIB G oup Su ey
on In es men and In es men Finance: A Technical No e on Da a Quali y,
Eu opean In es men Bank, Wo king Pape 2020/08. DOI 10.2867/772584
Bu le , A. W., and J. H. Co naggia. 2011. Does Access o Ex e nal Finance Imp o e
P oduc i i y? E idence om a Na u al Expe imen . Jou nal o Financial
Economics 99(1): 184–120.
Coad, A., P. Baue , C. Domnick, P. Ha asz osi, R. Pál, and M. Te uel. 2023. F om
Rapid Decline o High G ow h: Whe e in he Dis ibu ion Did COVID Hi
Ha des ? Jou nal o Small Business and En e p ise De elopmen 30(6):
1178–1209. h ps://doi.o g/10.1108/JSBED-02-2023-0061.
Duygan-Bump, B., A. Le ko , and J. Mon o iol-Ga iga. 2015. Financing Cons ain s
and Unemploymen : E idence om he G ea Recession. Jou nal o Mone a y
Economics 75: 89–100.
EIB. 2023. In es men Repo 2022/2023: Resilience and Renewal in Eu ope.
Eu opean In es men Bank.h ps://www.eib.o g/en/publica ions/20220211-
in es men - epo -2022.
———. 2024. In es men Repo 2023/2024: T ans o ming o Compe i i eness.
Eu opean In es men Bank, Luxembou g. h ps://www.eib.o g/en/publica ions/
20230323-in es men - epo -2023.
Fe nandes, A. P., and P. Fe ei a. 2017. Financing Cons ain s and Fixed-Te m
Employmen : E idence om he 2008–9 Financial C isis. Eu opean Economic
Re iew 92: 215–223.
Fe ando, A., R. Pál, and E. Du an e. 2019. Financing and Obs acles o High G ow h
En e p ises: The Eu opean Case. EIB Wo king Pape 2019/03.
Fe ando, A. and A. Ruggie i. 2018. Financial Cons ain s and P oduc i i y: E idence
om Eu o A ea Companies. In e na ional Jou nal o Finance & Economics
23(3): 257–228.
Fouejieu, A., A. Node, and T. Sydo enko. 2020. Unlocking Access o Finance o
SMEs: A C oss-Coun y Analysis. IMF Wo king Pape No. 2020/055
Ma ch 2020.
Ga cía-Posada Gómez, M. 2019. C edi Cons ain s, Fi m In es men and G ow h:
E idence om Su ey Da a. Jou nal o Banking and Finance 99: 121–114.
ADBI Wo king Pape [Do no en e ] Fe ando and Pál
17
Go odnichenko, Y., and M. Schni ze . 2013. Financial Cons ain s and Inno a ion: Why
Poo Coun ies Don’ Ca ch Up. Jou nal o he Eu opean Economic Associa ion
11(5): 1115–1152.
Ha asz osi, P., L. Mau in, R. Pál, D. Re ol ella, and W. an de Wielen. 2022. Fi m-
Le el Policy Suppo du ing he C isis: So Fa , So Good. In e na ional
Economics 171: 30–48.
Li, M. 2012. Using he P opensi y Sco e Me hod o Es ima e Causal E ec s: A Re iew
and P ac ical Guide. O ganiza ional Resea ch Me hods 00(0): 1–39.
Musso, P., and S. Schia o. 2008. The Impac o Financial Cons ain s on Fi m Su i al
and G ow h. Jou nal o E olu iona y Economics 18: 135–114.
Rosenbaum, P. R., and D. B. Rubin. 1983. The Cen al Role o he P opensi y Sco e in
Obse a ional S udies o Causal E ec s. Biome ika 70(1): 41–55.
Rosenbaum, P. R., and D. B. Rubin. 1984. Reducing Bias in Obse a ional S udies
Using Subclassi ica ion on he P opensi y Sco e. Jou nal o he Ame ican
S a is ical Associa ion 79(387): 516–524.
Te uel, M., Ama al-Ga cia S., Baue P., and A. Coad. 2023, P oduc i i y and HGEs:
Resilience and Reco e y om he COVID-19 Pandemic. Indus y and
Inno a ion, July 2023.
ADBI Wo king Pape [Do no en e ] Fe ando and Pál
18
ANNEX
Table A1: Desc ip ions and De ini ions o he Main Va iables
Va iable
Desc ip ion
Main dependen a iables
In es men gap
Fi ms decla ing ha in es men o e he las h ee yea s was oo li le o ensu e he
success o hei business going o wa d
Realized in es men
d op
Fi ms wi h less in es men han in he p e ious yea
Planned in es men
d op
Fi ms o which o al in es men expec ed o he cu en o nex yea is expec ed o
be less han in he p e ious yea
Ne in es men a e
Di e ence o ixed asse s be ween wo subsequen yea s, o e p e ious ixed asse s
Main a iables
Ex e nal unding
di icul ies
Fi ms wi h ei he s uc u al o cyclical unding di icul ies
s uc u al
Those iable i ms ha needed a loan bu we e ei he discou aged, did no ecei e i ,
o ecei ed less han hey needed. No egis e ing losses o h ee consecu i e yea s
cyclical
Fi ms expec ing hei ex e nal inancing condi ions o de e io a e. No egis e ing
losses o h ee consecu i e yea s
In e nal unding
di icul ies
Fi ms decla ing ha hei in e nal inance condi ions ha e de e io a ed. No egis e ing
losses o h ee consecu i e yea s.
Main con ol a iables
Fi m size
Fou size classes: mic o (23% o obse a ions), small (34% o obs), medium (29% o
obs), and la ge (15% o obs)
Sec o
B oad sec o g oups (dummy a iables): manu ac u ing (28% o obse a ions),
cons uc ion (22% o obs), se ices (26% o obs), and in as uc u e (23% o obs)
Coun y g oup
Coun ies a e clus e ed in o h ee g oups: “Cen e and Eas ,” “Sou h,” and
“No hwes .” “Cen e and Eas ”: BG, CZ, HR, HU, LT, LV, PL, RO, SI, SK; “Sou h”:
CY, ES, FR, GR, IT, MT, PT; “No hwes ”: AT, BE, DE, DK, EE, FI, IE, LU, NL, SE
P o i abili y
Cash low (p o i plus dep ecia ion) o e a e age o o al asse s (cu en and
p eceding yea )
Financial le e age
Sum o loans and long- e m deb o e o al asse s
Cash holdings
Amoun o cash and cash equi alen s o e o al asse s
Cash low
Ne income minus changes in wo king capi al o e o al asse s
ROA
ROA is calcula ed by di iding a i m's ne income by he a e age o i s o al asse s,
mul iplied by 100
Fi ms’ g ow h
Di e ence o o al asse s be ween wo subsequen yea s, o e p e ious o al asse s
Labou p oduc i i y
Labo p oduc i i y is calcula ed by di iding he o al ou pu by he o al numbe o
employees
Obs acle – unce ain y
The ex en o which unce ain y abou he u u e is an obs acle o in es men ac i i ies
Obs acle – lack o
demand
The ex en o which demand o p oduc and se ices is an obs acle o in es men
ac i i ies
Obs acle – lack o
skilled s a
The ex en o which a ailabili y o s a wi h he igh skills is an obs acle o in es men
ac i i ies
Obs acle – digi al
in as uc u e
The ex en o which access o digi al in as uc u e is an obs acle o in es men
ac i i ies
Leading inno a o s
Fi ms wi h (subs an ial) R&D and p oduc s new o he coun y o he global ma ke
Digi al
Fi ms ha ha e implemen ed digi al echnology in pa s o hei business o o ganized
hei en i e business a ound i
G een
Al eady in es ed o plan o in es o ackle he impac o wea he e en s o ca bon
emissions
ADBI Wo king Pape [Do no en e ] Fe ando and Pál
19
Figu e A1: In es men T ends – Compu ed wi h ORBIS Da a
Sou ce: O bis 2015–2023.
Figu e A2: In es men T ends – Compu ed Wi h EIBIS Da a
Sou ce: EIBIS 2016–2023.
-0.5
-0.4
-0.3
-0.2
-0.1
0
0.1
0.2
0.3
2003 2006 2009 2012 2015 2018 2021
Pe cen age change o in es men s
ne in es men g oss in es men
8%
9%
10%
11%
12%
13%
14%
15%
16%
2000 2003 2006 2009 2012 2015 2018 2021
Cash holdings o o al asse s
0%
20%
40%
60%
2016 2017 2018 2019 2020 2021 2022 2023
Sha e o i ms wi h in es men d op
( ealized and planned)
In es men d op in p e ious yea
Planned in es men d op o he cu en o nex pe iod
ADBI Wo king Pape [Do no en e ] Fe ando and Pál
20
Figu e A3: Ex e nal Funding Di icul ies and hei Impac
on In es men s – Annually
Ex e nal unding di icul ies and hei impac
on accumula ed in es men gap
Ex e nal unding di icul ies and hei impac
on planned in es men d op
No e: Ba s ep esen con idence in e al a 95% le el.
Sou ce: EIBIS-O bis 2023.
Figu e A4: Expec ed P obabili y o In es men Gap and Planned In es men D op
Wi h and Wi hou Ex e nal Funding Di icul ies – Annually
In es men gap
Planned in es men d op
No e: Ba s ep esen con idence in e al a 95% le el.
Sou ce: EIBIS-O bis 2023.
0%
5%
10%
15%
20%
25%
0%
5%
10%
15%
20%
25%
30%
2016
2017
2018
2019
2020
2021
2022
2023
Impac o ex e nal unding di icul ies in p e ious
yea on in es men gap, seconda y axes
Sha e o i ms wi h ex e nal unding di icul ies
0%
5%
10%
15%
20%
25%
0%
5%
10%
15%
20%
25%
30%
2016
2017
2018
2019
2020
2021
2022
2023
Impac o ex e nal unding di icul ies on planned
in es men d op, seconda y axes
Sha e o i ms wi h ex e nal unding di icul ies
0%
5%
10%
15%
20%
25%
30%
35%
40%
2017 2018 2019 2020 2021 2022 2023
No ex e nal unding di icul ies - lagged
Fi ms wi h ex e nal unding di icul ies - lagged
0%
10%
20%
30%
40%
50%
60%
2016 2017 2018 2019 2020 2021 2022 2023
No ex e nal unding di icul ies
Fi ms wi h ex e nal unding di icul ies
