Designing a country's small and medium-sized enterprise development index using firm-level data: The case of Thailand

Shinozaki, Shigehi o; Miyakawa, Daisuke
Wo king Pape
Designing a coun y's small and medium-sized en e p ise
de elopmen index using i m-le el da a: The case o
Thailand
ADB Economics Wo king Pape Se ies, No. 785
P o ided in Coope a ion wi h:
Asian De elopmen Bank (ADB), Manila
Sugges ed Ci a ion: Shinozaki, Shigehi o; Miyakawa, Daisuke (2025) : Designing a coun y's small
and medium-sized en e p ise de elopmen index using i m-le el da a: The case o Thailand, ADB
Economics Wo king Pape Se ies, No. 785, Asian De elopmen Bank (ADB), Manila,
h ps://doi.o g/10.22617/WPS250228-2
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/322378
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/3.0/igo/
ASIAN DEVELOPMENT BANK
ASIAN DEVELOPMENT BANK
6 ADB A enue, Mandaluyong Ci y
1550 Me o Manila, Philippines
www.adb.o g
ADB ECONOMICS
WORKING PAPER SERIES
NO. 785
June 2025
Designing a Coun y’s Small and Medium-Sized En e p ise De elopmen Index
Using Fi m-Le el Da a
The Case o Thailand
This pape employs p obabilis ic p incipal componen analysis o de elop a new way o quan i a i ely
assess wha a ec s mic o, small, and medium-sized en e p ise (MSME) de elopmen na ionally, by using
g anula i m-le el panel da a o 49,565 MSMEs in Thailand. The es ima ion esul s ound a po en ial
disp opo iona e e ec o MSME policy in e en ions du ing and a e he co ona i us disease pandemic.
This unde sco es he impo ance o using a ocused app oach when designing policies o MSME
de elopmen o acili a e mo e sus ainable, esilien p i a e sec o g ow h.
Abou he Asian De elopmen Bank
ADB is a leading mul ila e al de elopmen bank suppo ing inclusi e, esilien , and sus ainable g ow h ac oss
Asia and he Paci ic. Wo king wi h i s membe s and pa ne s o sol e complex challenges oge he , ADB
ha nesses inno a i e inancial ools and s a egic pa ne ships o ans o m li es, build quali y in as uc u e,
and sa egua d ou plane . Founded in 1966, ADB is owned by 69 membe s—50 om he egion.
DESIGNING A COUNTRY’S SMALL
AND MEDIUM-SIZED ENTERPRISE
DEVELOPMENT INDEX USING
FIRM-LEVEL DATA
THE CASE OF THAILAND
Shigehi o Shinozaki and Daisuke Miyakawa
ASIAN DEVELOPMENT BANK
The ADB Economics Wo king Pape Se ies
p esen s esea ch in p og ess o elici commen s
and encou age deba e on de elopmen issues
in Asia and he Paci ic. The iews exp essed
a e hose o he au ho s and do no necessa ily
e lec he iews and policies o ADB o
i s Boa d o Go e no s o he go e nmen s
hey ep esen .
ADB Economics Wo king Pape Se ies
Shigehi o Shinozaki and Daisuke Miyakawa
No. 785 | June 2025
Shigehi o Shinozaki ([email protected] g)
is a senio economis a he Economic Resea ch
and De elopmen Impac Depa men ,
Asian De elopmen Bank. Daisuke Miyakawa
([email protected]) is a p o esso a Waseda
Uni e si y and chie economis o UTokyo Economic
Consul ing Inc.
Designing a Coun y’s Small and Medium-Sized En e p ise
De elopmen Index Using Fi m-Le el Da a:
The Case o Thailand
C ea i e Commons A ibu ion 3.0 IGO license (CC BY 3.0 IGO)
© 2025 Asian De elopmen Bank
6 ADB A enue, Mandaluyong Ci y, 1550 Me o Manila, Philippines
Tel +63 2 8632 4444; Fax +63 2 8636 2444
www.adb.o g
Some igh s ese ed. Published in 2025.
ISSN 2313-6537 (p in ), 2313-6545 (PDF)
Publica ion S ock No. WPS250228-2
DOI: h p://dx.doi.o g/10.22617/WPS250228-2
The iews exp essed in his publica ion a e hose o he au ho s and do no necessa ily e lec he iews and policies
o  he Asian De elopmen Bank (ADB) o i s Boa d o Go e no s o he go e nmen s hey ep esen .
ADB does no gua an ee he accu acy o he da a included in his publica ion and accep s no esponsibili y o any
consequence o hei use. The men ion o speci ic companies o p oduc s o manu ac u e s does no imply ha hey
a e endo sed o ecommended by ADB in p e e ence o o he s o a simila na u e ha a e no men ioned.
By making any designa ion o o e e ence o a pa icula e i o y o geog aphic a ea in his documen , ADB does no
in end o make any judgmen s as o he legal o o he s a us o any e i o y o a ea.
This publica ion is a ailable unde he C ea i e Commons A ibu ion 3.0 IGO license (CC BY 3.0 IGO)
h ps://c ea i ecommons.o g/licenses/by/3.0/igo/. By using he con en o his publica ion, you ag ee o be bound
by he e ms o his license. Fo a ibu ion, ansla ions, adap a ions, and pe missions, please ead he p o isions
and e ms o use a h ps://www.adb.o g/ e ms-use#openaccess.
This CC license does no apply o non-ADB copy igh ma e ials in his publica ion. I he ma e ial is a ibu ed
oano he sou ce, please con ac he copy igh owne o publishe o ha sou ce o pe mission o ep oduce i .
ADB canno be held liable o any claims ha a ise as a esul o you use o he ma e ial.
Please con ac pubsma ke [email protected] g i you ha e ques ions o commen s wi h espec o con en , o i you wish
oob ain copy igh pe mission o you in ended use ha does no all wi hin hese e ms, o o pe mission o use
heADB logo.
Co igenda o ADB publica ions may be ound a h p://www.adb.o g/publica ions/co igenda.
No e:
In his publica ion, “B” e e s o bah .
ABSTRACT
Unde s anding he business en i onmen and s uc u al issues ha limi g ow h is c i ical when
designing an e ec i e na ional policy amewo k o p i a e sec o de elopmen —especially o
mic o, small, and medium-sized en e p ises (MSMEs). Gi en he limi ed MSME da a a ailable,
his pape employs p obabilis ic p incipal componen analysis o de elop a new way o
quan i a i ely assess wha a ec s MSME de elopmen na ionally, by using g anula i m-le el
panel da a o 49,565 MSMEs in Thailand as a case s udy. The es ima ion esul s ound a po en ial
disp opo iona e e ec o MSME policy in e en ions du ing and a e he co ona i us disease
pandemic. Go e nmen assis ance o MSMEs likely helped Bangkok-based i ms ease he
nega i e pandemic e ec s, especially in manu ac u ing. Howe e , i did no help local MSMEs—
ega dless o sec o —as hei ope a ional pe o mance de e io a ed bo h du ing and a e he
pandemic. This unde sco es he impo ance o using a ocused app oach when designing policies
o MSME de elopmen o acili a e mo e sus ainable, esilien p i a e sec o g ow h.
Keywo ds: SME de elopmen , access o inance, inancial inclusion, SME policy, p obabilis ic
p incipal componen analysis, Thailand
JEL codes: D22, G20, L20, L50

1. In oduc ion
Mac o and mic o policy implemen a ion should be consis en wi h assis ance measu e a ge s.
While highly agg ega ed a iables such as g oss domes ic p oduc (GDP) and he in la ion a e
ha e been used as key measu es in guiding policy design, policy make s a e now paying mo e
a en ion o g anula da a such as i m-le el inancial s a emen s. This e lec s he need o
conside he he e ogenous s a us and cha ac e is ics o policy a ge s—such as mic o, small, and
medium-sized en e p ises (MSMEs)—so policy implemen a ion is mo e e ec i e and e icien
(Gou inchas e al. 2020; Ebeke e al. 2021). In his pape , we empi ically measu e summa ized
business ac i i y using g anula da a o cons uc i m ac i i y indexes ha accoun o he
he e ogenei y among policy a ge s.
MSMEs ha e a la ge impac on business ac i i y, job c ea ion, and economic ou pu , helping d i e
g ow h ac oss de eloping Asia and he Paci ic. Go e nmen s in he egion hus use a ious
assis ance measu es o na ional MSME de elopmen — o p omo e young and women
en ep eneu s, adop and comme cialize echnology, expand MSME in e na ional ma ke access,
de elop human capi al and wo ke skills, and boos access o inance. Howe e , cons ain s on
MSME de elopmen emain in mos coun ies, aising he ques ion o how go e nmen s can
design e ec i e policies ha be e each MSMEs. Unde s anding he business en i onmen and
s uc u al p oblems associa ed wi h business g ow h is c i ical o design an e ec i e na ional
policy amewo k o MSME de elopmen . Howe e , limi ed MSME da a make i di icul o do so.
Gi en he gene al use ulness o cons uc ing i m ac i i y indexes, a ious global o ganiza ions
such as he O ganisa ion o Economic Co-ope a ion and De elopmen , he Economic Resea ch
Ins i u e o ASEAN and Eas Asia, and he In e na ional T ade Cen e ha e al eady begun he
p ocess. They p opose a quali a i e app oach using assessmen ma ices o pe o mance a ings
o median compa isons based on a ailable da a o e alua e MSME de elopmen condi ions (ADB
2022). Despi e hese e o s, he disc e ion associa ed wi h ways o cons uc measu es, he cos
o p oducing eliable measu es, and he di icul y o in e p e ing esul s a e also conside ed majo
issues public o ganiza ions mus add ess. P eceding s udies such as ADB (2022) and Shinozaki
e al. (2024) aim o o e come his di icul y by applying an empi ical me hod—p obabilis ic
p incipal componen analysis (P-PCA)— o coun y-le el annual da a, which is immune o
disc e ion and easy o implemen , hus allowing o s aigh o wa d in e p e a ion.
One ca ea o p eceding s udies is ha cons uc ing ac i i y indexes o MSMEs—such as he
Small and Medium-Sized En e p ise De elopmen Index (SME-DI)—do no necessa ily conside
he he e ogenei y among policy a ge s. This d awback e lec s he ac ha ADB (2022) and
Shinozaki e al. (2024) used agg ega e da a such as coun y-le el mac oeconomic a iables o
cons uc an SME-DI ha co e s he en i e (o sub) Asian egion(s). In his pape , we ollow ADB
(2022) and Shinozaki e al. (2024) and u he apply he P-PCA me hod o disagg ega ed da a so
we can conside he he e ogenous s a us among policy a ge s, o example, ac oss a ious
indus ies and egions in cons uc ing an SME-DI.
He e, we use p op ie a y i m-le el panel da a ob ained om a p i a e da a ende in Thailand
(Dun & B ads ee ) o apply he P-PCA me hod o he g anula da a. Speci ically, we conduc
se e al exe cises o u he de elop he SME-DI. Fi s , we apply he P-PCA me hod o indus y-
egion-yea le el disagg ega ed panel da a (cons uc ed om i m-yea le el g anula da a
speci ically on Thailand’s MSMEs) o c ea e an index accoun ing o MSME ac i i ies in a single
coun y. This index is u he b oken down by indus y- and/o egional-le el indica o s as well as
indica o s explici ly accoun ing o MSMEs’ eal and inancial ac i i ies. Second, using he smalle
da ase andomly chosen by he en i e g anula da ase , we cons uc indus y- egion-yea le el
2
disagg ega ed panel da a and e- un he P-PCA algo i hm. Compa ing his esul wi h ha based
on he en i e MSME da ase , we examine whe he o no he ela i ely small amoun o da a can
ep oduce an SME-DI consis en wi h ha ob ained om he en i e da ase . This p o ides use ul
in o ma ion p ac i ione s can easily implemen — o example, su ey analyses o cons uc a
eliable indica o summa izing MSME ac i i ies. Thi d, we u he apply he P-PCA me hod o i m-
le el “ aw” g anula da a (MSME da a no agg ega ed by indus y- egion bu used as is) and
di ec ly ob ain he indica o s ha accoun o MSME ac i i ies in Thailand. Again, his exe cise
con ibu es o ou deepe unde s anding o he alue o su ey analyses in cons uc ing a eliable
indica o summa izing MSME ac i i ies.
2. Thailand’s MSME Landscape and Policy Suppo Measu es
Like in o he coun ies, MSMEs play a c ucial ole in d i ing Thailand’s economy.1 O e ime, hei
numbe s ha e consis en ly inc eased, e en du ing he co ona i us disease (COVID-19)
pandemic ha s a ed in ea ly 2020. MSMEs inc eased in numbe by 0.9% in 2020 and 1.4% in
2021. As o end-2023, he e we e 3.2 million MSMEs, accoun ing o 99.5% o all en e p ises
(ADB 2024). By sec o , he la ges sha e was in wholesale and e ail ade (41.8%), ollowed by
o he se ices (including accommoda ion and ood se ices, 40.4%) and manu ac u ing (16.0%).
MSMEs we e sp ead ac oss he coun y wi h 83.6% in he p o inces and 16.4% in he capi al ci y
Bangkok. As o end-2023, MSMEs employed 12.9 million wo ke s o 70.4% o he o al wo k o ce,
which has g adually expanded du ing he pos -pandemic eco e y.
While GDP g ow h in Thailand slowed in 2023 (2.0%), MSME ou pu emained ela i ely obus .
Despi e he pandemic challenges, GDP ebounded quickly o MSMEs— om a 9.6% decline in
2020 o 4.2% g ow h in 2021, a V-shaped eco e y. As o end-2023, MSME ou pu was B6.3
illion, accoun ing o 35.2% o GDP, up 3.6% om 2022. Among MSMEs, o he se ices
con ibu ed mos (41.2%), ollowed by manu ac u ing (30.0%) and wholesale and e ail ade
(21.8%) (ADB 2024).
MSME expo s also ebounded quickly om he pandemic impac , g owing by a ema kable
22.0% in 2021 a e a 17.1% d op in 2020. By 2023, MSME expo alue eached B1.3 illion o
13.4% o o al expo alue. MSME expo s g ew by a obus 24.3% om 2022. Majo expo
des ina ions included o he Sou heas Asian coun ies, he People’s Republic o China (PRC), he
Uni ed S a es, and he Eu opean Union, which oge he accoun ed o nea ly 80% o o al MSME
expo alue. Nea ly all majo MSME expo ma ke s expanded p ima ily due o hei own
economic eco e y (O ice o Small and Medium En e p ises P omo ion [OSMEP] 2022). In 2021,
expo olumes o gems and jewel y, wood, ag icul u al p oduce, i on and s eel, and plas ic
p oduc s g ew as es . By con as , MSME expo s such as suga , au omobile pa s, and ubbe
p oduc s declined in olume. MSME impo s also expanded, eaching B1.4 illion in 2023,
accoun ing o 14.0% o o al impo alue, a 16.2% inc ease om 2022 (ADB 2024).
The go e nmen ’s quick policy ac ions suppo ing MSMEs and wo ke s du ing he pandemic likely
con ibu ed o he lowe pandemic impac on businesses in Thailand. Fo deb inancing, he
cen al bank (Bank o Thailand) p o ided se e al liquidi y suppo measu es o banks o inance
MSMEs hu by he pandemic. These included capi al bu e s o banks, de e ed p incipal
paymen s, educed in e es a es, and new so loans/c edi lines o MSMEs. The go e nmen
1 In Thailand, MSMEs a e de ined as ei he (i) manu ac u ing o (ii) se ices and ading, using he numbe o employees
and e enue as c i e ia (ADB 2024). Fo manu ac u ing (including ag icul u e), an MSME is de ined as a i m wi h up
o 200 employees o annual e enue o less han B500 million. Fo se ices and ading, an MSME can ha e up o
100 employees o annual e enue below B300 million.
3
used alue added ax (VAT) e unds o domes ic en ep eneu s and educed social secu i y
con ibu ions and wi hholding axes o businesses. I also p o ided B5,000 mon hly handou s o
sel -employed and laid-o wo ke s, and pa ial (50%) sala y payou s o hose displaced. To help
lowe business cos s, wa e and elec ici y paymen s we e suspended (ADB 2020).
Gi en he impo ance o MSMEs in he Thailand economy, he go e nmen de eloped a medium-
e m s a egy o p omo e MSMEs. The cu en SME P omo ion Plan 2023–2027 aims o c ea e a
s ong and p og essi e en i onmen o make MSMEs mo e compe i i e. I has h ee s a egic
pilla s: (i) inclusi e de elopmen ac oss all MSME sec o s, (ii) iden i y ma ke oppo uni ies, and
(iii) c ea e a suppo i e MSMEs ecosys em ha imp o es access o inance, echnology,
inno a ion, skills de elopmen , big da a, he legal amewo k, and policy in o ma ion (ADB 2024).
3. Empi ical App oaches
This s udy uses he P-PCA empi ical app oach. As de ailed in ADB (2022) and Shinozaki e al.
(2024), he idea behind he P-PCA me hod is basically same as s anda d p incipal componen
analysis (PCA), which ex ac s common ac o s sha ed by mul iple ime se ies da a. One unique
ea u e o he P-PCA me hod is i s obus ness agains missing eco ds, which is likely he case in
de eloping coun ies. This i e a ion o he PCA calcula ion and in e pola ion p ocesses gi es us
he p incipal componen s (and ac o loadings) based on he da a su e ing om missing eco ds.
The de ailed P-PCA model is explained in Appendix 1.2
The ADB (2022) and Shinozaki e al. (2024) main analysis akes ou a limi ed numbe o common
ac o s sha ed by, o example, mul iple ime-se ies mac oeconomic da a accoun ing o he en i e
(o sub) Asian egion(s). Unlike p e ious s udies, we ocus on a single coun y (Thailand) as a
case s udy, and ollow ou analy ical s eps. Fi s , we use g anula MSME panel da a (agg ega ing
ime-se ies indus y- egion le el g anula i m da a) o ake ou he common ac o s as a baseline
analysis. Second, we b eak down he analysis by indus y and egion, and use ha nexus o ake
ou he common ac o s. Thi d, he same p ocedu es a e used o analyze he end o eal o
inancial ac i i ies o i ms. And ou h, we conduc small sample exe cises o see o wha ex en
he dynamics o he ex ac ed p incipal componen s a e in luenced by sample size.
4. Da a
The i m-le el panel da a used he e come om a p i a e da a ende in Thailand (Dun &
B ads ee ). The da a consis o annual equency i m-le el in o ma ion o e he pe iods om
2016 o 2023.3 In his i m-le el da a, we ha e i m-yea unbalanced panel da a o 11 a iables,
consis ing o hose associa ed wi h i ms’ eal and inancial ac i i ies (Table 1). The o iginal da a
co e 52,086 i ms wi h no missing da a on employmen and o al 370,676 samples o 2016–
2023. F om hese, we ex ac MSMEs ha i he c i e ia o i ms in manu ac u ing o ag icul u e
wi h up o 200 employees and hose in se ices and o he indus ies wi h up o 100 employees–
49,565 MSMEs wi h a o al o 353,373 samples o 2016–2023, equi alen o 95% o o al i m
samples.4 This MSME da ase is analyzed in his s udy.
2 The Appendix is a ailable a h p://dx.doi.o g/10.22617/WPS250228-2.
3 Th oughou his pape , we call a i m’s inancial s a emen da a s o ed in he o iginal da abase ending in yea YYYY
as he yea = YYYY. Fo example, i a i m’s inancial s a emen accoun s o he accoun ing pe iod om Janua y 2023
o Decembe 2023, we call his da a as yea = 2023.
4 A i m is classi ied as an MSME i i mee s he c i e ia o a leas 1 yea du ing he da a pe iod. This classi ica ion is
i o he employmen h eshold o he na ional MSME de ini ion in Thailand (see oo no e 1).
4
The indus y classi ica ion o he da a e e s o nine ca ego ies based on he s anda d indus ial
classi ica ion sys em: (i) ag icul u al, o es y, and ishing; (ii) mining; (iii) cons uc ion; (i )
manu ac u ing; ( ) anspo a ion; ( i) wholesale; ( ii) e ail; ( iii) inancial se ices; and (ix) o he
se ices.5 The egional classi ica ion o he da a e e s o Bangkok (capi al ci y) and six o icial
egions: Cen al (excep Bangkok), Eas e n, No h, No heas , Sou he n, and Wes e n Thailand.
To cons uc he inpu da a o he P-PCA me hod, we agg ega e each a iable in each yea using
hese indus y and egion classi ica ions.
Fo he small sample exe cises, we andomly choose a se o da a om he en i e MSME da ase .
This p o ides us an expe imen al en i onmen whe e we use in o ma ion on he limi ed numbe
o su eyed i ms o compu e he SME-DI. Whe e he numbe o andomly chosen i ms is limi ed,
we apply he P-PCA di ec ly o hose selec ed i m-le el panel da a o ob ain he SME-DI based
on he pseudo su ey da a.6 In o he wo ds, we es ima e he ac o s mo ing behind hose i m-
le el da a se ies and a la ge numbe o ac o loadings o each i m- a iable pai .
Table 1: MSME Da a in Thailand
A. Va iables Used
ID
Va iable name
De ini ion
Real/Financial
1
sls_ e n_am
Sales e enue
Real
2
emp_cn
Numbe o employees
Real
3
g s_p _o _lss_am
G oss p o i /loss
Real
4
be _ ax_ne _p _am
Ne p o i be o e axes
Real
5
o _ase _am
To al asse s
Real
6
w kg_capl_am
Wo king capi al
Real
7
o _liab_am
To al liabili ies
Financial
8
d_db _am
T ade deb o s
Financial
9
bnk_loan_od_am
Bank loans ou s anding/o e d a
Financial
10
d_c _am
T ade c edi
Financial
11
o _cu _liab_am
To al cu en liabili ies
Financial
5 In his indus y classi ica ion, o example, “Wholesale” is ea ed as a dis inc indus y classi ica ion om “Re ail” while
hese wo indus ies a e o en placed in one ca ego y. Gi en ha he da a con ain a su icien numbe o obse a ions
o hese wo indus y classi ica ions and ha we p e e a la ge numbe o indus ies o de e mine he numbe o semi-
agg ega e ime se ies da a used in applying he P-PCA, we ea “Wholesale” as a dis inc indus y classi ica ion om
“Re ail.”
6 The applica ion o he P-PCA me hod o he en i e da a o i m-yea le el g anula da a equi es a ce ain le el o
compu ing powe , which we do no ha e o his pape . Re ining he compu ing p ocess would be one o he mos
impo an ways o ob ain SME-DI in he u u e om he aw da a (i.e., i m-yea le el da a).
Con inued on he nex page
11
B. Rela ion Be ween he Second Fac o and Va iables
C. Rela ion Be ween he Thi d Fac o and Va iables
Sou ce: Au ho s’ calcula ions.
Bank Loan
O e d a
Employees
To al
G oss
P o i /Loss
Amoun
Ne P o i
Be o e
Taxes
Sales
Re enue
Amoun
To al
Asse s
Amoun
To al
Cu en
Liabili ies
Amoun
To al
Liabili ies
Amoun
T ade
C edi
T ade
Deb o s
Wo king
Capi al
Amoun
(-1.0, -0.9) 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0130 0.0000 0.0000
(-0.9, -0.8) 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0130 0.0000 0.0260 0.0000 0.0000
(-0.8, -0.7) 0.0000 0.0000 0.0000 0.0000 0.0130 0.0000 0.0000 0.0260 0.0000 0.0130 0.0000
(-0.7, -0.6) 0.0260 0.0130 0.0130 0.0000 0.0130 0.0000 0.0260 0.0000 0.0130 0.0000 0.0000
(-0.6, -0.5) 0.0130 0.0000 0.0000 0.0130 0.0000 0.0000 0.0000 0.0000 0.0130 0.0130 0.0000
(-0.5, -0.4) 0.0000 0.0000 0.0519 0.0130 0.0000 0.0000 0.0130 0.0260 0.0000 0.0130 0.0000
(-0.4, -0.3) 0.0000 0.0000 0.0130 0.0130 0.0000 0.0130 0.0130 0.0000 0.0000 0.0000 0.0130
(-0.3, -0.2) 0.0130 0.0260 0.0000 0.0000 0.0000 0.0260 0.0130 0.0130 0.0000 0.0130 0.0000
(-0.2, -0.1) 0.0000 0.0000 0.0000 0.0000 0.0000 0.0260 0.0000 0.0130 0.0130 0.0000 0.0130
(-0.1, -2.22e-16) 0.0130 0.0000 0.0130 0.0260 0.0000 0.0130 0.0130 0.0130 0.0130 0.0130 0.0000
(-2.22e-16, 0.1) 0.0130 0.0000 0.0000 0.0130 0.0000 0.0130 0.0000 0.0000 0.0000 0.0260 0.0000
(0.1, 0.2) 0.0130 0.0130 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0130
(0.2, 0.3) 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0390
(0.3, 0.4) 0.0000 0.0260 0.0000 0.0130 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
(0.4, 0.5) 0.0000 0.0130 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
(0.5, 0.6) 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
(0.6, 0.7) 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0130
(0.7, 0.8) 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
(0.8, 0.9) 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
(0.9, 1.0) 0.0000 0.0000 0.0000 0.0000 0.0649 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
mean -0.285* 0.032 -0.401* -0.156 0.45 -0.143* -0.455* -0.402* -0.555* -0.269* 0.154
Bank Loan
O e d a
Employees
To al
G oss
P o i /Loss
Amoun
Ne P o i
Be o e
Taxes
Sales
Re enue
Amoun
To al
Asse s
Amoun
To al
Cu en
Liabili ies
Amoun
To al
Liabili ies
Amoun
T ade
C edi
T ade
Deb o s
Wo king
Capi al
Amoun
(-1.0, -0.9) 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
(-0.9, -0.8) 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
(-0.8, -0.7) 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
(-0.7, -0.6) 0.0000 0.0130 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0130
(-0.6, -0.5) 0.0000 0.0130 0.0000 0.0000 0.0000 0.0000 0.0000 0.0130 0.0000 0.0000 0.0130
(-0.5, -0.4) 0.0000 0.0130 0.0000 0.0000 0.0000 0.0130 0.0130 0.0000 0.0130 0.0000 0.0130
(-0.4, -0.3) 0.0000 0.0260 0.0000 0.0260 0.0000 0.0260 0.0000 0.0130 0.0000 0.0000 0.0000
(-0.3, -0.2) 0.0000 0.0000 0.0000 0.0000 0.0000 0.0130 0.0000 0.0000 0.0000 0.0000 0.0130
(-0.2, -0.1) 0.0130 0.0130 0.0000 0.0130 0.0000 0.0130 0.0260 0.0260 0.0260 0.0260 0.0260
(-0.1, -2.22e-16) 0.0130 0.0000 0.0260 0.0130 0.0000 0.0130 0.0130 0.0000 0.0000 0.0260 0.0000
(-2.22e-16, 0.1) 0.0260 0.0130 0.0000 0.0000 0.0260 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
(0.1, 0.2) 0.0000 0.0000 0.0000 0.0260 0.0000 0.0130 0.0130 0.0390 0.0260 0.0260 0.0000
(0.2, 0.3) 0.0130 0.0000 0.0000 0.0000 0.0649 0.0000 0.0000 0.0000 0.0130 0.0000 0.0130
(0.3, 0.4) 0.0000 0.0000 0.0130 0.0000 0.0000 0.0000 0.0130 0.0000 0.0000 0.0000 0.0000
(0.4, 0.5) 0.0130 0.0000 0.0130 0.0000 0.0000 0.0000 0.0130 0.0000 0.0000 0.0130 0.0000
(0.5, 0.6) 0.0000 0.0000 0.0130 0.0000 0.0000 0.0000 0.0000 0.0000 0.0130 0.0000 0.0000
(0.6, 0.7) 0.0130 0.0000 0.0130 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
(0.7, 0.8) 0.0000 0.0000 0.0130 0.0130 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
(0.8, 0.9) 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
(0.9, 1.0) 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
mean 0.171 -0.365* 0.378* 0.017 0.183* -0.196* 0.035 -0.108 0.04 0.061 -0.299*

12
b. MSMEs ope a ing in Bangkok (Capi al Ci y)
Figu e 3 depic s he op h ee ac o s in he case o egional-le el indica o s such as o he capi al
ci y, Bangkok. In his es ima ion, h ee ac o s explain 88% o he da a a ia ion. 9 Table 4
summa izes he dis ibu ion o he es ima ed impac s o each a iable on h ee ac o s o he case
o Bangkok.
The associa ions be ween he h ee ac o s and he a iables in he case o Bangkok sugges
se e al hings. The i s ac o implies ha o e he cou se o he COVID-19 pandemic, many
MSMEs ope a ing in Bangkok had a highe deb bu den unde lowe p o i abili y and sales
e enue. The second ac o sugges s ha p o i shows a “hump” shape o e ime. This g oup o
MSMEs likely aced p o i losses du ing and a e he pandemic. The hi d ac o is nega i ely
co ela ed wi h mos inancial-side a iables while showing a posi i e co ela ion wi h sales
e enue (no associa ion wi h o he eal-side a iables o employmen and p o i s). MSMEs loca ed
in Bangkok we e hu inancially in some indus ies no well co e ed by sales e enue.
This es ima ion sugges s ha Bangkok-based MSMEs hu by he pandemic by sales and p o i
we e likely o ob ain mo e bank c edi . Go e nmen assis ance measu es du ing he pandemic
migh ha e been disp opo iona ely alloca ed o capi al ci y-based MSMEs in some indus ies.
Figu e 3: SME-DI o Bangkok (Capi al Ci y)
SME-DI = Small and Medium-Sized En e p ise De elopmen Index.
Sou ce: Au ho s’ calcula ions.
9 The con ibu ion o each es ima ed p incipal componen (PC) is 65% o PC1, an addi ional 16% o PC2, and a u he
7% o PC3, o a o al o 88% explained.
-0.8
-0.6
-0.4
-0.2
0
0.2
0.4
0.6
0.8
2016 2017 2018 2019 2020 2021 2022 2023
Comp 1 Comp 2 Comp 3
Weigh s
13
Table 4: Fac o Loadings based on P obabilis ic P incipal Componen Analysis—
Bangkok (Capi al Ci y)
A. Rela ion Be ween he Fi s Fac o and Va iables
B. Rela ion Be ween he Second Fac o and Va iables
Bank Loan
O e d a
Employees
To al
G oss
P o i /Loss
Amoun
Ne P o i
Be o e
Taxes
Sales
Re enue
Amoun
To al
Asse s
Amoun
To al
Cu en
Liabili ies
Amoun
To al
Liabili ies
Amoun
T ade
C edi
T ade
Deb o s
Wo king
Capi al
Amoun
(-1.0, -0.9) 0.0000 0.0101 0.0000 0.0101 0.0000 0.0101 0.0101 0.0101 0.0101 0.0000 0.0000
(-0.9, -0.8) 0.0000 0.0101 0.0101 0.0101 0.0000 0.0000 0.0000 0.0000 0.0000 0.0101 0.0101
(-0.8, -0.7) 0.0000 0.0000 0.0000 0.0101 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
(-0.7, -0.6) 0.0000 0.0000 0.0000 0.0000 0.0101 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
(-0.6, -0.5) 0.0000 0.0000 0.0000 0.0000 0.0404 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
(-0.5, -0.4) 0.0000 0.0000 0.0101 0.0000 0.0000 0.0000 0.0000 0.0000 0.0101 0.0101 0.0000
(-0.4, -0.3) 0.0000 0.0000 0.0000 0.0000 0.0202 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
(-0.3, -0.2) 0.0101 0.0101 0.0202 0.0000 0.0000 0.0000 0.0000 0.0000 0.0101 0.0000 0.0000
(-0.2, -0.1) 0.0000 0.0000 0.0303 0.0202 0.0000 0.0000 0.0000 0.0000 0.0000 0.0101 0.0101
(-0.1, -2.22e-16) 0.0101 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0101 0.0000 0.0000
(-2.22e-16, 0.1) 0.0101 0.0000 0.0101 0.0000 0.0101 0.0000 0.0202 0.0000 0.0000 0.0000 0.0000
(0.1, 0.2) 0.0101 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0101 0.0000
(0.2, 0.3) 0.0000 0.0000 0.0101 0.0000 0.0000 0.0000 0.0101 0.0101 0.0000 0.0000 0.0101
(0.3, 0.4) 0.0000 0.0101 0.0000 0.0000 0.0000 0.0000 0.0101 0.0101 0.0101 0.0101 0.0101
(0.4, 0.5) 0.0202 0.0000 0.0000 0.0000 0.0000 0.0101 0.0000 0.0000 0.0202 0.0000 0.0202
(0.5, 0.6) 0.0000 0.0000 0.0000 0.0202 0.0000 0.0202 0.0202 0.0202 0.0101 0.0101 0.0000
(0.6, 0.7) 0.0202 0.0000 0.0000 0.0101 0.0101 0.0101 0.0000 0.0202 0.0000 0.0101 0.0000
(0.7, 0.8) 0.0101 0.0101 0.0000 0.0101 0.0000 0.0101 0.0202 0.0202 0.0101 0.0202 0.0101
(0.8, 0.9) 0.0000 0.0404 0.0000 0.0000 0.0000 0.0303 0.0000 0.0000 0.0000 0.0000 0.0202
(0.9, 1.0) 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
mean 0.306* 0.28 -0.197* -0.034 -0.317* 0.51* 0.242 0.399* 0.096 0.192 0.318
Bank Loan
O e d a
Employees
To al
G oss
P o i /Loss
Amoun
Ne P o i
Be o e
Taxes
Sales
Re enue
Amoun
To al
Asse s
Amoun
To al
Cu en
Liabili ies
Amoun
To al
Liabili ies
Amoun
T ade
C edi
T ade
Deb o s
Wo king
Capi al
Amoun
(-1.0, -0.9) 0.0000 0.0000 0.0303 0.0101 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
(-0.9, -0.8) 0.0000 0.0000 0.0202 0.0000 0.0101 0.0101 0.0202 0.0000 0.0101 0.0000 0.0101
(-0.8, -0.7) 0.0101 0.0000 0.0101 0.0000 0.0000 0.0000 0.0000 0.0000 0.0202 0.0303 0.0000
(-0.7, -0.6) 0.0101 0.0000 0.0101 0.0000 0.0101 0.0000 0.0000 0.0202 0.0000 0.0000 0.0000
(-0.6, -0.5) 0.0000 0.0000 0.0101 0.0000 0.0101 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
(-0.5, -0.4) 0.0000 0.0101 0.0000 0.0101 0.0000 0.0000 0.0101 0.0101 0.0000 0.0101 0.0000
(-0.4, -0.3) 0.0101 0.0000 0.0000 0.0101 0.0000 0.0000 0.0000 0.0000 0.0101 0.0101 0.0000
(-0.3, -0.2) 0.0000 0.0000 0.0000 0.0101 0.0000 0.0000 0.0000 0.0101 0.0000 0.0000 0.0000
(-0.2, -0.1) 0.0000 0.0101 0.0000 0.0000 0.0000 0.0000 0.0000 0.0202 0.0000 0.0000 0.0000
(-0.1, -2.22e-16) 0.0101 0.0101 0.0101 0.0101 0.0000 0.0000 0.0101 0.0000 0.0101 0.0000 0.0202
(-2.22e-16, 0.1) 0.0101 0.0202 0.0000 0.0000 0.0000 0.0404 0.0202 0.0000 0.0000 0.0000 0.0000
(0.1, 0.2) 0.0000 0.0303 0.0000 0.0000 0.0000 0.0202 0.0101 0.0202 0.0000 0.0101 0.0101
(0.2, 0.3) 0.0000 0.0000 0.0000 0.0000 0.0000 0.0101 0.0000 0.0101 0.0101 0.0202 0.0101
(0.3, 0.4) 0.0202 0.0000 0.0000 0.0202 0.0101 0.0101 0.0202 0.0000 0.0000 0.0000 0.0101
(0.4, 0.5) 0.0000 0.0000 0.0000 0.0101 0.0000 0.0000 0.0000 0.0000 0.0202 0.0101 0.0000
(0.5, 0.6) 0.0000 0.0101 0.0000 0.0101 0.0505 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
(0.6, 0.7) 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0101
(0.7, 0.8) 0.0202 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0101 0.0000 0.0000
(0.8, 0.9) 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0202
(0.9, 1.0) 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
mean 0.038 0.07 -0.71* -0.043 0.117 0.026 -0.14 -0.189 -0.083 -0.208 0.219
Con inued on he nex page
14
C. Rela ion Be ween he Thi d Fac o and Va iables
Sou ce: Au ho s’ calcula ions.
c. Manu ac u ing i ms by egion
He e we in es iga e how he me hod we use can highligh he he e ogenei y among MSMEs
h ough he indus y- egion nexus. Figu e 4 depic s h ee ac o s ob ained om he da a
accoun ing o manu ac u ing i ms ope a ing in Bangkok, while Figu e 5 depic s hose in
manu ac u ing ou side Bangkok (local a eas).10 I should be no ed ha in his exe cise we inpu
he aw i m le el da a o MSMEs in o he model ins ead o he agg ega e MSME da a se ies
(Appendix 2A).
We can immedia ely iden i y ha he i s ac o s in each igu e—which accoun o he la ges
a ia ion o he da a among he h ee ac o s—show di e en dynamics. Using he i s ac o s
om Figu es 4 and 5 as an example, we can depic he i m’s ac i i ies in hose wo cases. F om
he associa ions be ween he i s ac o in Figu e 4 and each a iable, we ound ha a la ge
numbe o capi al ci y-based manu ac u ing MSMEs likely educed hei employmen wi h la ge
bo owing while no necessa ily seeing sales e enue and p o i de e io a e du ing and pos
pandemic, al hough he es ima es we e no s a is ically signi ican (Appendix 2A1). F om he
associa ions be ween he i s ac o in Figu e 5 and each a iable, we ound ha many local
manu ac u ing MSMEs (non-capi al ci y-based i ms) expe ienced lowe p o i s om he pandemic
onse un il 2022. Thei p o i s likely imp o ed in 2023 (Appendix 2A2).
10 The con ibu ion o each es ima ed p incipal componen (PC) o Figu e 4 is 72% o PC1, an addi ional 14% o PC2,
and a u he 5% o PC3, o a o al o 91% explained. Fo Figu e 5, i is 53% o PC1, an addi ional 20% o PC2, and
a u he 11% o PC3, o a o al o 84% explained.
Bank Loan
O e d a
Employees
To al
G oss
P o i /Loss
Amoun
Ne P o i
Be o e
Taxes
Sales
Re enue
Amoun
To al
Asse s
Amoun
To al
Cu en
Liabili ies
Amoun
To al
Liabili ies
Amoun
T ade
C edi
T ade
Deb o s
Wo king
Capi al
Amoun
(-1.0, -0.9) 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
(-0.9, -0.8) 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
(-0.8, -0.7) 0.0101 0.0101 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
(-0.7, -0.6) 0.0000 0.0000 0.0000 0.0000 0.0000 0.0101 0.0101 0.0101 0.0101 0.0000 0.0000
(-0.6, -0.5) 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0101 0.0303 0.0000 0.0303 0.0000
(-0.5, -0.4) 0.0000 0.0000 0.0101 0.0000 0.0000 0.0101 0.0202 0.0101 0.0202 0.0000 0.0000
(-0.4, -0.3) 0.0101 0.0000 0.0202 0.0000 0.0000 0.0202 0.0202 0.0000 0.0000 0.0101 0.0303
(-0.3, -0.2) 0.0202 0.0000 0.0101 0.0202 0.0000 0.0000 0.0000 0.0101 0.0101 0.0000 0.0101
(-0.2, -0.1) 0.0000 0.0101 0.0101 0.0101 0.0101 0.0101 0.0000 0.0303 0.0202 0.0000 0.0101
(-0.1, -2.22e-16) 0.0202 0.0101 0.0101 0.0101 0.0101 0.0202 0.0101 0.0000 0.0000 0.0101 0.0202
(-2.22e-16, 0.1) 0.0101 0.0000 0.0000 0.0101 0.0000 0.0101 0.0101 0.0000 0.0101 0.0202 0.0101
(0.1, 0.2) 0.0000 0.0000 0.0202 0.0000 0.0101 0.0101 0.0101 0.0000 0.0101 0.0101 0.0000
(0.2, 0.3) 0.0000 0.0202 0.0000 0.0202 0.0101 0.0000 0.0000 0.0000 0.0101 0.0000 0.0101
(0.3, 0.4) 0.0101 0.0303 0.0101 0.0101 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
(0.4, 0.5) 0.0101 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0101 0.0000
(0.5, 0.6) 0.0000 0.0000 0.0000 0.0101 0.0101 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
(0.6, 0.7) 0.0000 0.0000 0.0000 0.0000 0.0404 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
(0.7, 0.8) 0.0000 0.0101 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
(0.8, 0.9) 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
(0.9, 1.0) 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
mean -0.099 0.148 -0.102 0.083 0.373* -0.206* -0.291* -0.374* -0.184* -0.156 -0.138*
15
Figu e 4: Manu ac u ing—Bangkok
Sou ce: Au ho s’ calcula ions.
Figu e 5: Manu ac u ing—Ou side Bangkok
Sou ce: Au ho s’ calcula ions.
16
This es ima ion sugges s ha manu ac u ing MSMEs based in Bangkok cos cu s h ough layo s
and success ully ob ained bank c edi du ing he pandemic, a ibu ed o he lowe impac o he
pandemic on hei sales e enue and p o i s. Eme gency go e nmen assis ance measu es likely
acili a ed hei access o bank c edi . By con as , local manu ac u ing MSMEs likely saw p o i s
all du ing he pandemic, as hey we e no well co e ed by go e nmen inancial assis ance
( hough hey imp o ed in 2023).
d. Ag ibusinesses by egion
A simila exe cise was conduc ed on ag icul u al MSMEs by egion. The i s ac o s we e
ob ained om he da a accoun ing o ag ibusiness i ms ope a ing in Bangkok and ag ibusiness
MSMEs ou side he capi al (Figu es 6 and 7) (Appendix 2B).11
The es ima es simply show ha p o i abili y de e io a ed amid he COVID-19 pandemic. Capi al
ci y-based ag icul u al MSMEs likely de e io a ed in sales e enue and g oss p o i wi h no
e ec i e suppo om any imp o ed wo king capi al condi ions du ing he pandemic, al hough
es ima es we e no s a is ically signi ican (Appendix 2B1). Simila ly, local ag icul u al MSMEs
consis en ly saw hei p o i s de e io a e o e he cou se o he pandemic, accele a ed by a
cons an slowdown in access o bank c edi , e en as wo king capi al condi ions imp o ed
(es ima es s ongly signi ican ) (Appendix 2B2).
Figu e 6: Ag icul u e—Bangkok
Sou ce: Au ho s’ calcula ions.
11 The con ibu ion o each es ima ed p incipal componen (PC) o Figu e 6 is 69% o PC1, an addi ional 19% o PC2,
and a u he 4% o PC3, o a o al o 92% explained. Fo Figu e 7, i is 59% o PC1, an addi ional 16% o PC2, and
a u he 11% o PC3, o a o al o 86% explained.

17
Figu e 7: Ag icul u e—Ou side Bangkok
Sou ce: Au ho s’ calcula ions.
Real and inancial b eakdown o he SME-DI
In ou baseline analysis, we simul aneously used all a iables—ca ego ized as bo h eal and
inancial. One possible, use ul exe cise is o use he da a accoun ed o by he eal and inancial
a iables sepa a ely so we can see he ex en o which he baseline esul is d i en by each ype
o a iable.
The op h ee ac o s based on he da a can be shown sepa a ely as eal o inancial a iables
(Figu es 8 and 9). In bo h es ima ions, hose h ee ac o s explain 87% o he da a a ia ion.12 A
able on ac o loadings is a ached (Appendix 2C).
The mos impo an ac is ha he dynamics o he ac o s in he wo panels a e simila . Mo e
p ecisely, he i s ac o s in he wo igu es a e inc easing while he second ac o s a e in u-shape.
The hi d ac o s exhibi a mo e complica ed end in bo h he eal and inancial cases. Findings
om hese sepa a e exe cises by eal and inancial a iables a e consis en wi h he baseline
esul s using all a iables (see Table 2). Fo example, he i s ac o in eal a iables shows ha
MSME sales e enues and p o i s de e io a ed du ing he pandemic while hey likely s eng hened
wo king capi al and he wo k o ce (Appendix 2C1). The inancial a iables show ha MSME
bo owing om banks likely inc eased wi h mo e cus ome s no ye paying o hei p oduc s and
se ices (Appendix 2C2). Al hough i is na u al o see hese links be ween a iables o e episodes
such as he pandemic, his app oach would be po en ially use ul o see whe he o no eal and
inancial a iables accoun o MSME business dynamics, and how hey do so.
12 The con ibu ion o each es ima ed p incipal componen (PC) o Figu e 8 is 58% o PC1, an addi ional 19% o PC2,
and a u he 10% o PC3, o a o al o 87% explained. Fo Figu e 9, i is 62% o PC1, an addi ional 18% o PC2,
and a u he 7% o PC3, o a o al o 87% explained.
18
Figu e 8: SME-DI on Real Va iables
SME-DI = Small and Medium-Sized En e p ise De elopmen Index.
Sou ce: Au ho s’ calcula ions.
Figu e 9: SME-DI on Financial Va iables
SME-DI = Small and Medium-Sized En e p ise De elopmen Index.
Sou ce: Au ho s’ calcula ions.
19
Small sample exe cises
As men ioned, by using he pseudo da ase andomly chosen om all g anula MSME da a and
e- unning he P-PCA algo i hm, we can see he ex en o which he ela i ely small numbe o
da a can p oduce a SME-DI consis en wi h ha using he en i e MSME da ase . This expe imen
is use ul o p ac i ione s conduc ing, o example, su ey analyses o cons uc a eliable indica o
summa izing MSME ac i i ies. By doing so, we cons uc ed wo se s o SME-DI. Panel A (Panel
B) o Figu e 10 depic s an SME-DI based on he andomly chosen subsample ha accoun s o
50% (10%) o he en i e MSME da a.13 A ac o loadings able is a ached (Appendix 2D).
Fi s , we can check he consis ency be ween Figu e 1 and Figu e 10A. The SME-DI based on
50% o he da ase basically eplica es ha based on all MSME da a. O e he pe iods o ou
analysis, he i s ac o mo ed om a low o high le el, he second ac o exhibi ed a U-shape
ansi ion, wi h he hi d ac o showing mo e complica ed dynamics. A ca e ul inspec ion o he
associa ion be ween each ac o and each a iable sugges s ha he associa ion is compa ible
wi h Figu e 1 and Table 2. Fo example, in he i s ac o based on a 50% sample, MSME sales
e enues and ne p o i s be o e ax de e io a ed o e he cou se o he pandemic while hey
ecei ed wo king capi al inancing om banks, which con ibu ed o s eng hening hei balance
shee s: a simila esul as in he baseline es ima e using all MSME da a (Appendix 2D1).
Figu e 10: Small Sample Exe cises
A. 50% Sample
13 The con ibu ion o each es ima ed p incipal componen (PC) o Figu e 10A is 44% o PC1, an addi ional 16% o
PC2, and a u he 12% o PC3, o a o al o 72% explained. Fo Figu e 10B, i is 41% o PC1, an addi ional 16% o
PC2, and a u he 12% o PC3, o a o al o 69% explained.
Con inued on he nex page
20
B. 10% Sample
Sou ce: Au ho s’ calcula ions.
Second, we can con i m none heless ha i is no necessa ily easy o eplica e a SME-DI based
on he MSME sample by using andomly chosen and smalle samples. The consis ency be ween
he wo se s o es ima ed ac o s wo sens as he size o he subsample becomes smalle . Figu e
10B accoun s o he h ee ac o s based on a andomly chosen 10% size o he en i e da ase . I
ails o ollow he o iginal SME-DI based on he en i e da ase . Also, he associa ion be ween each
ac o and each a iable does no necessa ily sha e he same ea u es as in he ull sample
(Appendix 2D2). To summa ize, he small sample exe cise helps us con i m ha a ce ain
educ ion in sample size does no c ea e any se ious p oblems in es ima ing he SME-DI. As his
c i ically depends on he a ge g oup, i would be in e es ing o see how his empi ical s udy
applies owa d o he da ase s.
5.2. Fi m-Yea Le el Raw G anula Da a
As a second se o esul s, we p esen he SME-DI based on i m-yea le el “ aw” g anula da a,
o MSME da a no agg ega ed a he indus y- egion le el bu used as is. Figu e 11 shows he op
h ee ac o s ob ained om he P-PCA me hod applied o 10% andomly chosen da a. These
h ee ac o s explain 62% o he en i e a ia ion o he da a.14 Table 5 summa izes he dis ibu ion
o he es ima ed impac s o each a iable on he h ee ac o s. Simila o he baseline exe cise
using indus y- egion-yea le el agg ega e da a, applying he P-PCA yields he same numbe o
es ima ed ac o loadings as ha o i ms (a ound 5,000 i ms depending on choice o a iables).
Fo each pai , he impac s a e compu ed as he mul iplica ion o he sign o he ac o loading o
each ac o and sha e o he a ia ion o each a iable o ha o he a ia ion o he co esponding
ac o .
Fi s , we can iden i y ha he i s ac o in Figu e 11 is a mi o image o he i s ac o in Figu e
1. We ind ha he p o i has a posi i e (nega i e) associa ion wi h he i s ac o in Figu e 11
(Figu e 1). This sugges s ha o some ex en he 10% andomly chosen aw da a ep oduce he
SME-DI based on mo e agg ega ed (bu no ull) da a.
14 The con ibu ion o each es ima ed p incipal componen (PC) is 43% o PC1, an addi ional 12% o PC2, and a
u he 7% o PC3, o a o al o 62% explained.
ASIAN DEVELOPMENT BANK
ASIAN DEVELOPMENT BANK
6 ADB A enue, Mandaluyong Ci y
1550 Me o Manila, Philippines
www.adb.o g
ADB ECONOMICS
WORKING PAPER SERIES
NO. 785
June 2025
Designing a Coun y’s Small and Medium-Sized En e p ise De elopmen Index
Using Fi m-Le el Da a
The Case o Thailand
This pape employs p obabilis ic p incipal componen analysis o de elop a new way o quan i a i ely
assess wha a ec s mic o, small, and medium-sized en e p ise (MSME) de elopmen na ionally, by using
g anula i m-le el panel da a o 49,565 MSMEs in Thailand. The es ima ion esul s ound a po en ial
disp opo iona e e ec o MSME policy in e en ions du ing and a e he co ona i us disease pandemic.
This unde sco es he impo ance o using a ocused app oach when designing policies o MSME
de elopmen o acili a e mo e sus ainable, esilien p i a e sec o g ow h.
Abou he Asian De elopmen Bank
ADB is a leading mul ila e al de elopmen bank suppo ing inclusi e, esilien , and sus ainable g ow h ac oss
Asia and he Paci ic. Wo king wi h i s membe s and pa ne s o sol e complex challenges oge he , ADB
ha nesses inno a i e inancial ools and s a egic pa ne ships o ans o m li es, build quali y in as uc u e,
and sa egua d ou plane . Founded in 1966, ADB is owned by 69 membe s—50 om he egion.
DESIGNING A COUNTRY’S SMALL
AND MEDIUM-SIZED ENTERPRISE
DEVELOPMENT INDEX USING
FIRM-LEVEL DATA
THE CASE OF THAILAND
Shigehi o Shinozaki and Daisuke Miyakawa