Perceived corruption in business environment: exploring the underlying causes
Abstract
This paper presents an empirical contribution to the literature that tends to explain variations in corruption perceptions. Drawing on theoretical assumptions about the impact of corruption on the quality of the business environment, we explore possible determinants of perceived corruption. We argue that explaining the determinants of corruption perceptions could be a valuable input for creating effective anti-corruption policies. In the paper, we perform a detailed analysis on the case of Serbia, a transition country with relatively widespread corruption in the business environment. Using the non-parametric analysis of variance test, we examine the significance of differences in the perceived spread of corruption across a number of independent variables (regions, size of locality, industry, size, and years of operation). Specifying an ordered logistic regression model, we estimate the relation between corruption perceptions and a number of potential causes: frequencies of unofficial payments or gifts for different purposes, the impact that unofficial payments to public officials have on the business operations of the respondents, as well as attitudes of the respondents about the quality of the judicial system and political stability. We find that perceptions of corruption as an obstacle to business operations can be linked to the frequency of unofficial payments to deal with customs and taxes, as well as the impact of unofficial payments to local or regional government officials on business performance of the respondents. The results provide useful insights into what policy measures are necessary to reduce the level of corruption, as well as how the effects of such measures can be assessed.
Full text
48 2018, XXI, 4
Ekonomie
DOI: 10.15240/ ul/001/2018-4-004
In oduc ion
Business en i onmen is an impo an
de e minan o na ional compe i i eness and
sus ainable long- e m economic g ow h.
Business en i onmen quali y de e mines he
isk as well as e u ns om in es men and
he e o e a ec s in es men decisions.
The business en i onmen is la gely, i no
o ally, beyond he con ol o he i ms and hei
managemen . I esul s om exis ing poli ical,
legal and egula o y amewo k; mac oeconomic
policies; ins i u ional in as uc u e; social and
cul u al con ex wi hin which ansac ions
ake place, he quali y o physical and social
in as uc u e and many o he ac o s. Empi ical
esea ch has shown ha co up ion ep esen s
a se ious obs acle o en ep eneu ship and
business (Kau mann & Wei, 2000; Meon &
Sekka , 2005). Hellman e al. (2000) conside
co up ion, in addi ion o go e nance quali y and
s a e cap u e, a ac o ha signi i can ly shapes
business en i onmen in ansi ion coun ies.
Go e nmen has a i al ole in c ea ing and
ensu ing he s able and heal hy socio-economic
en i onmen o businesses. By ensu ing lowe
ansac ion cos s on indi iduals and i ms, well-
egula ed business en i onmen allows hem o
ope a e mo e e i cien ly. Acco ding o Gu ie
(2004) co up ion and poo ins i u ions a e key
de e minan s o exis ing poo egula ion and
misgo e nance. B een and Gillande s (2012)
conside a coun y’s exis ing s ock o egula ion
as a p oduc o i s ecen his o y o ins i u ional
quali y and co up ion. Main i ndings o hei
esea ch show ha he le el o co up ion is
he mos impo an ac o ha de e mines he
quali y o he business en i onmen , umping
he quali y o ins i u ions and a ange o o he
indica o s. The e o e, de elopmen and
implemen a ion o app op ia e s a egies,
ocused on con inuous imp o emen o
business en i onmen a e o g ea impo ance.
Al hough some au ho s jus i y co up ion on
he g ound ha i helps i ms o deal wi h obs acles
caused by ine i cien public adminis a ion
(D ehe & Gassebne , 2013; Le , 1964), co up
business en i onmen o ces i ms o alloca e
signi i can pa o hei human, i nancial and
o he esou ces o co up p ac ices ins ead o
ocusing on inno a ion and de elopmen . La ge
companies a e in be e posi ion because hey
ha e esou ces and capaci ies o p o ec hei
business in e es s and cope wi h co up ion,
while small and medium-sized en e p ises a e
mo e s ongly a ec ed by co up ion. The e
is also e idence ha he ex en o co up ion
a ies in di e en egions o he same coun y
(Linha o á & Volejníko á, 2015). This s udy
ends o con ibu e o he li e a u e on he
causes o co up ion. Pa icula ly, we empi ically
es he ele ance o a ious de e minan s o
co up ion in he business en i onmen , on he
example o a single ansi ion economy. Se bia
is a coun y whe e co up ion is pe cei ed as
a majo impedimen o in es ing and doing
business e i cien ly (S ošić, Nikolić, & Đukić,
2011). In ou analysis, we use da a on he
pe cep ions o co up ion h oughou Se bian
business en i onmen , as well as he eal cases
o co up ion, in he o m o in o mal paymen s
o gi s o a ious pu poses. We in es iga e o
wha ex en co up ion ep esen s an obs acle
o business ansac ions and o e po en ial
explana ions o he de e minan s o such
a i udes.
The pape is o ganized as ollows. Sec ion 1
p esen s a b ie o e iew o he main i ndings
o he li e a u e on co up ion, i s de e minan s
and he impac on he business en i onmen .
Sec ion 2 desc ibes he da a sou ces and explains
he p oposed me hodological p ocedu es. In
sec ions 3 and 4, he esul s o empi ical es s a e
p esen ed and discussed. The pape ends wi h
a summa y o he main i ndings.
PERCEIVED CORRUPTION IN BUSINESS
ENVIRONMENT: EXPLORING
THE UNDERLYING CAUSES
Ma ija Džunić, Na aša Golubo ić
EM_4_2018.indd 48EM_4_2018.indd 48 28.11.2018 13:12:4828.11.2018 13:12:48
49
4, XXI, 2018
Economics
1. Li e a u e Re iew
Co up ion is a complex social phenomenon ha
has a long his o y. Fo ms, causes, mechanisms
and conseqences o co up ion ep esen
a popula subjec o scien i i c inqui ies (Mau o,
1995; Paldam, 2002; Rose-Acke man, 1999;
Shlei e & Vishny, 1993). Ye , despi e nume ous
poli ical, legal, economic and sociological
s udies, he e is no globally accep ed de i ni ion
o co up ion. Co up ion is mos commonly
de i ned as he misuse o he abuse o powe
o pe sonal gain and exis s in bo h public and
p i a e sec o s.
Co up ion is p edominan ly associa ed wi h
he public sec o . Mau o (1995) deals only wi h
go e nmen co up ion. Shlei e and Vishny
(1993) ocus hei a en ion on co up ion in
he public sec o , whe e co up ion is a “sale
by go e nmen o i cial o go e nmen p ope y
o pe sonal gain” (p. 599). Pe haps he mos
comp ehensi e de i ni ion o co up ion is gi en
by Tanzi (1995), who associa es co up ion
wi h he delibe a e iola ion o impa iali y in
decision making p ocess, o he pu pose o
app op ia ing bene i s o onesel o o ela ed
indi iduals.
Rega dless o i s o ms, i is widely belie ed
ha co up ion ep esen s a conside able
impedimen o sus ainable g ow h and
de elopmen (Mau o, 1995; Tanzi, 1998).
High le els o co up ion a ec in es men
decisions and economic g ow h (Mau o, 1997).
Go e nance and co up ion issues, acco ding
o Kau man (2005), ep esen key cons ain s
o in es men and business. Business
en i onmen ha is conduci e o g ow h has
many dimensions, ela ed o he quali y o
ins i u ions, go e nmen policy, egula o y
amewo k e c. (UNODC, 2013). Besides
poli ical and economic s abili y, a sound
business en i onmen o a g ea deal depends
on he con ol o co up ion. Pe cep ions abou
business en i onmen a e a undamen al
ac o ha shapes decisions o i ms. I i ms
pe cei e business en i onmen as a ou able,
hey will be mo e inclined o in es . Co up ion
ep esen s a kind o ax on economic ac i i y,
which aises ansac ion cos s (Shlei e &
Vishny, 1993). Th ough ising unce ain y and
educed p o i abili y o en e p ises, co up ion
educes he willingness o en e p ises o
in es . I will in u n educe he a ac i eness
o en ep eneu ship; en ep eneu ial alen will
be di e ed o less p oduc i e ac i i ies, wi h
he nega i e impac on he inno a ion and
hus economic g ow h (Ha is & Me we, 2012,
p. 171). The e is also ou pu loss due o he
misalloca ion o esou ces (Shlei e & Vishny,
1993), dis o ions o incen i es, s eng hening
o he in o mal economy, changing s uc u e
o in es men and go e nmen spending.
Co up ion nega i ely a ec s coun y’s capaci y
o a ac o eign di ec in es men , which
is a majo ehicle o echnology ans e
(Abed & Da odi, 2002; Sma zynska & Wei,
2000). Indices o co up ion send ou signals
o po en ial in es o s, p o iding in o ma ion
necessa y o he assessmen o a coun y
isk and he cos o doing business, and o
he decision whe e o in es . Howe e , some
au ho s conside co up ion as bene i cial
because i helps o o e come he ine i ciency
o he go e nmen egula ions and a oid
bu eauc a ic obs acles (Le , 1964). Acco ding
o Ali and Isse (2003), b ibe y is a subs i u e o
highe wages, meaning ha i sa es money o
he go e nmen .
In he poli ical sphe e, co up ion causes
he dec ease o o e s’ con i dence in elec ed
ep esen a i es, unde mines he ule o law,
and unde mines he epu a ion o go e nmen
and o he ins i u ions. Co up ion a ec s he
c edibili y o poli ical ins i u ions in on o
ci izens and he es o he wo ld. Acco ding o
Hodgson and Jiang (2007, p. 1057) co up ion
educes le els o us in dealing wi h bo h
business and he s a e. I e odes con i dence in
public ins i u ions, diminishes s a e’s abili y o
p o ide public goods and se ices and c ea e
conduci e en i onmen o p i a e sec o
de elopmen .
Fu he mo e, co up ion a ec s o he
impo an de e minan s o business
en i onmen , such as social capi al (Pu nam,
1993; F i zen, Se i zlew, & S endsen, 2014).
Reliance on in o mal con ac s and amily,
iendship, e hnic and o he ies caused by
co up ion leads o he loss o bene i s o
wide coope a ion and compe i ion (G ae &
S endsen, 2013). I unde mines he e i cacy
o he o mal ules ha a e e y impo an o
he e i cien ma ke economy. S ong in o mal
ne wo ks wi h bonding social ies and lack o
e i cien o mal ins i u ions c ea e e ile g ound
o u he expansion o co up ion (Lipse &
Lenz, 2000; Ha is, 2007). In he long un,
co up ion causes he inc ease o inequali y and
po e y, c ea ing a gene al clima e o insecu i y,
EM_4_2018.indd 49EM_4_2018.indd 49 28.11.2018 13:12:4828.11.2018 13:12:48
50 2018, XXI, 4
Ekonomie
lack o us in ins i u ions and he gene al
dissa is ac ion o all social s a ums (Ro hs ein
& Uslane , 2005).
In ansi ion coun ies, he collapse o
he exis ing ins i u ions o cen ally-planned
economy and slow eme gence o new
ins i u ions compa ible wi h ma ke economy
has es ablished ins i u ional acuum and e ile
g ound o public sec o co up ion. Ins i u ional
weakness, such as poo ly de i ned p ope y
igh s, weak accoun abili y o public o i cials and
limi ed poli ical compe i ion, has had a p o ound
impac on he suscep ibili y o ansi ion
coun ies o co up ion. A he beginning o
ansi ion hese coun ies we e ocused on
he p oblem o economic ans o ma ion and
p i a iza ion. Re o ms o he s a e, public
adminis a ion and o he judicia y we e la gely
neglec ed (Acke man, 1992, p. 46). These
de elopmen s a ec ed he pace o poli ical
and economic e o ms and c ea ed a unique
ansi ion pa h o e e y coun y. Coun ies
whe e ansi ion pa h led o concen a ion o
economic powe in a se ing o weak ins i u ions
became especially ulne able o co up ion.
The egion o he Wes e n Balkans is
pe cei ed as a egion o high co up ion.
Co up ion is pe cei ed as a majo obs acle o
in es men and doing business in his egion
(EUROCHAMBRES, 2009). This is con i med
by he esul s o se e al s udies (S ošić, Nikolić,
& Đukić, 2011; Dža i ć e al., 2011; Budak & Rajh,
2011). Acco ding o business ep esen a i es in
Se bia, co up ion is he i h mos signi i can
obs acle o doing business, igh a e ax a es,
ine i cien go e nmen bu eauc acy, poli ical
ins abili y and access o i nancing (WEF, 2016).
Uni ed Na ions O i ce on D ugs and C ime
s udy abou he impac o co up ion on p i a e
en e p ises in Se bia e eals ha co up ion
is a g ea hind ance o p i a e en e p ise and
has a nega i e e ec upon p i a e in es men
(UNDOC, 2013).
The ob ious nega i e impac o co up ion
on economic and social ou comes implica es
ha explo ing possible de e minan s o
co up ion is e y impo an . Howe e , a se ious
limi a ion o s udies ha explo e causes
o co up ion ela es o he p oblems in i s
measu emen . Ac ual co up ion inciden s a e
e y ha d o de ec , since agen s end o hide
hei pa icipa ion in co up p ac ices. Fo his
eason, di ec measu es o co up ion a e a ely
a ailable o esea ch. The e o e, widely used
measu es o co up ion a e es ima es based
on pe cep ions o he equency and le el o
co up ion, cap u ed by su eys. Al hough
pe cep ions could possibly de ia e om
ac ual co up ion inciden s (T eisman, 2007),
pe cep ions based indica o s a e conside ed
a eliable indica o o co up ion. In ha sense,
when explo ing co up ion de e minan s,
empi ical s udies mainly ely on da a abou
pe cei ed co up ion.
The e a e many possible causes o
co up ion sugges ed by he li e a u e.
Economic de e minan s o co up ion mos
o en include coun y’s economic weal h (Hall
& Jones, 1999), income inequali y (Paldam,
2002), economic eedom (T eisman, 2000),
ade openness (Pe sson e al., 2003), in l a ion
and mac o-economic pe o mance (Melga e
al., 2010). S udies ela ed o poli ical causes
o co up ion ha e ocused on he impac o
poli ical ins abili y (Lede man e al., 2005),
poli ical libe ies (T eisman, 2000), elec o al
sys ems (Pe sson e al., 2003), e c.
Go e nmen is ci ed as he impo an
sou ce o co up ion because ex ensi e s a e
egula ion o he economy c ea es possibili ies
o he ex ac ion o en s. Empi ical esea ch
abou o igins o co up ion indica es ha poo ly
unc ioning ins i u ions and policies unde mine
ee ade and compe i ion, hus leading o
co up ion. Based on he analysis o ins i u ional
quali y in 18 OECD coun ies D ehe e al. (2009)
i nd ha inc ease in ins i u ional quali y educes
he le el o co up ion. Ex ensi e egula ion o
ma ke en y, in he o m o p ocedu es, cos s
and ime o s a a new business, is consis en
wi h highe le els o co up ion (Djanko e
al., 2002). A highly di e si i ed ade a i may
encou age public o i cials o ex ac b ibes om
impo e s (Ga i, 1999).
The e a e wo basic sou ces o co up ion
ela ed o go e nmen ac i i y. The i s
conce ns he way go e nmen pe o ms i s
ac i i ies, such as ax collec ion, cus oms and
licensing. Powe s o he go e nmen o i cials
a e o en accompanied by disc e ion, non-
anspa en o ambiguous p ocedu es, c ea ing
a ou able condi ions o ex ac ing b ibes.
The e o e, he ule o law and he quali y o
bu eauc acy a e impo an de e minan s o
co up ion (Rauch & E ans, 1997; Ali & Isse,
2003; B une i & Wede , 2004). Co up ion ends
o be lowe in he en i onmen cha ac e ized by
s ong and e ec i e judicial sys em, con ac
EM_4_2018.indd 50EM_4_2018.indd 50 28.11.2018 13:12:4828.11.2018 13:12:48
51
4, XXI, 2018
Economics
en o ceabili y and p o essional bu eauc acy.
The second includes poli ical ac o s in ol ed
in poli ical decision making, such as legisla ion,
p i a iza ion and con ac ing, whose ac i i ies
may cause a subs an ial dys unc ion o he
legisla i e, execu i e and judicial b anches o
powe . Ac i i ies o in e es g oups, which a e
o en illegal and non- anspa en , may in l uence
go e nmen policies, egula ions and laws, wi h
he pu pose o achie ing p i a e bene i s.
Ins i u ional and gene alized us s a e
also pe cei ed as ele an p edic o s o
co up ion (Ro hs ein & Uslane , 2005).
Co up en i onmen s a e cha ac e ized by
conside able de e io a ion o us in ins i u ions
and poli icians, as well as in business pa ne s.
2. Da a and Me hodology
Fo he pu pose o his s udy, da a om he i h
ound o Business En i onmen and En e p ise
Pe o mance Su ey – BEEPS V (EBRD,
2015) we e used. The BEEPS su ey has been
conduc ed as a join ini ia i e o he Eu opean
Bank o Recons uc ion and De elopmen and
he Wo ld Bank, since 1999, as a i m-le el
su ey based on in e iews wi h i m manage s,
wi h he aim o assessing he quali y o he
business en i onmen and main challenges o
he p i a e sec o de elopmen . The i h ound
o he su ey, implemen ed o e he pe iod
2012-2014, co e ed 15,883 en e p ises in 30
coun ies o Eas e n Eu ope and Cen al Asia.
The esea ch in his pape is based on da a om
he su ey implemen ed in Se bia. Rega ding
sampling implemen a ion, in each coun y wo
sampling ames ha e been used: he i s , an
o i cial ame o es ablishmen s supplied by
he o i cial sou ces in he coun y (Business
Regis e s Agency in he case o Se bia);
and he second, consis ing o en e p ises
in e iewed in p e ious ound o he su ey. In
pa icula , ou o he o al numbe o con ac ed
es ablishmen s in Se bia (836), he numbe o
non-eligible uni s was 73, 40 es ablishmen s
we e impossible o con ac , while 363 e used
o pa icipa e in he su ey. The numbe o
comple ed in e iews was 360, indica ing he
esponse a e o 0.43. The in e iews wi h
senio manage s we e conduc ed om Janua y
o Augus 2013. The da a used in his pape
exp ess hei a i udes abou ela ions be ween
business and go e nmen . The owne ship
s uc u e o he su eyed en e p ises is as
ollows: 317 en e p ises a e 100% owned by
domes ic p i a e capi al, 25 en e p ises a e o
100% p i a e o eign owne ship, 10 en e p ises
ha e a ce ain pe cen age o s a e capi al
( om 1 o 93%), while he es (8 en e p ises)
epo combined owne ship. All ypes o
en e p ises (mic o, small, medium and la ge)
a e encompassed by he su ey.
In he a emp o explo e he unde lying
causes o co up ion pe cep ions, we es
whe he he pe cep ions o co up ion as an
obs acle o business ansac ions di e based
on dis inc i e ea u es o he i ms. Besides
c oss- abula ions and desc ip i e s a is ics,
a nonpa ame ic es o he analysis o a iance
was pe o med, as o de e mine whe he he e
a e s a is ically signi i can di e ences in he
pe cei ed sp ead o co up ion ac oss a numbe
o independen a iables ( egions, size o locali y,
indus y, size, and yea s o ope a ion). Since he
examined a iables consis o mo e ca ego ies,
wi h a clea o de ing and equally spaced
in e als be ween he alues, he me hod o
choice is he K uskal-Wallis one-way analysis-
o - a iance es , H (K uskal & Wallis, 1952;
1953). This is a non-pa ame ic e sion o he
one-way ANOVA and an ex ension o he Mann-
Whi ney U es , used when he assump ion o
no mali y is iola ed, as in he case o da a used
o his s udy. The da a in ou sample mee all
he assump ions ha unde pin he applica ion
o K uskal-Wallis H es . All he dependen
a iables a e measu ed on a Like scale wi h
5 o mo e poin s; independen a iables consis
o wo o mo e ca ego ical independen g oups;
obse a ions a e independen ; and, mos
impo an ly, dis ibu ions o sco es ( o each
g oup o independen a iable) ha e he same
shape ( a iabili y).
Addi ionally, by speci ying an o de ed
logis ic eg ession model, we explo e he
de e minan s o he a i udes on co up ion. We
es ima e he ela ions be ween he pe cep ions
o co up ion and se e al independen
a iables ( equencies o uno i cial paymen s
and gi s o di e en pu poses, pe cei ed
impac o in o mal paymen s on go e nmen
decisions, pe cep ions o he impo ance
o poli ical s abili y and cou s o business
ope a ions). As he dependen a iable in he
model is ca ego ical and o de ed (pe cep ions
o co up ion as an obs acle o doing
business, exp essed by 5-poin Like scale),
we es ima e an o de ed logis ic eg ession
model, also known as p opo ional odds model
EM_4_2018.indd 51EM_4_2018.indd 51 28.11.2018 13:12:4928.11.2018 13:12:49
52 2018, XXI, 4
Ekonomie
(Ai chison & Sil ey, 1957; McKel ey & Za oina,
1975; Long & F eese, 2014).
Using he ologi model, we ob ain he
p obabili ies o obse ing each o he possible
ou comes i o dependan a iable (deg ees
o pe cei ed co up ion) wi hin he ange o
he es ima ed cu -poin s. In o he wo ds, he
p obabili y o ou come i is calcula ed as he
p obabili y ha he alue o he linea unc ion
o he independen a iables (including he
e o e m uj) is wi hin he ange o es ima ed
cu -poin s (ki):
(1)
In his model, he ela ionship be ween
each pai o ou come g oups is he same and i
is desc ibed by one se o coe i cien s, so only
one equa ion is es ima ed. Two p ocedu es
a e pe o med o es ing he p opo ional odds
assump ion: a likelihood a io es (Wol e &
Gould, 1998) and B an es (B an , 1990).
Based on he model, we ob ain p edic ions
ega ding he e ec s ha imp o ing business
en i onmen can ha e on co up ion pe cep ions
o he exis ing and po en ial business ac o s.
3. Resul s
3.1 Da a O e iew and Desc ip i e
S a is ics
Th oughou ansi ion coun ies, i m manage s
epo ha op h ee obs acles, as pe cei ed
by each i m, a e compe i o s’ p ac ices in he
in o mal sec o , access o i nance and elec ici y
issues (EBRD, 2015a). Howe e , co up ion
s ill emains one o he bigges obs acles, e en
hough he su ey esul s indica e he e has
been a signi i can decline in he sha e o i ms
epo ing ha in o mal paymen o gi s we e
expec ed o eques ed om hem in di e en
in e ac ions wi h go e nmen o i cials. One
o he easons o his decline, as epo ed,
migh be he in oduc ion o elec onic i ling
and paymen sys ems, which educe he
possibili ies o o i cials o eques in o mal
paymen s. S ill, he e a e a numbe o i ms ha
epo signi i can sha e o hei annual sales o
he pu pose o “ge ing hings done”.
The op h ee business en i onmen
obs acles iden i i ed by i ms in Se bia a e poli ical
ins abili y, ax a es and access o i nance
(EBRD, 2015). These esul s a e de e mined
by he in ensi e poli ical de elopmen s in
he pe iod p io o su ey – p esiden ial and
pa liamen a y elec ions, ollowed by he change
o he uling pa y. The new go e nmen has
in oduced a i scal consolida ion e o m, which
caused a ise in he ax a es. Access o i nance
has become he bigges p oblem o Se bian
i ms, due o he lack o c edi unds and
inc eased in e es a es. Besides, co up ion
appea s o be an impo an obs acle in Se bia.
As o 6.11% o he o al numbe o su eyed
i ms co up ion ep esen s he bigges obs acle
o business ope a ions.
The scope o co up ion appea s e e y
ime a i m in e ac s wi h go e nmen o i cials
o se ice p o ide s. Howe e , he pe cep ions
o i ms abou co up ion ac i i ies end o
di e , depending on hei size, indus y, egion
o owne ship s uc u e. Small i ms in Se bia
complain he mos abou co up ion, while all
he complain s a ise om he i ms ha a e
100% owned by domes ic p i a e capi al. The
majo i y o i ms ha complain abou co up ion
a e ope a ing in Belg ade, he capi al o Se bia.
A de ailed s a is ics on i ms’ pe cep ions
o co up ion as an obs acle o hei cu en
business ope a ions is p esen ed in Tab. 1 ( he
esponses a e assigned sco es om 0 = No
obs acle o 4 = Ve y se e e obs acle). The da a
indica e ha co up ion is widesp ead in he
business en i onmen . Less han a hal o he
su eyed i ms do no expe ience any p oblems
wi h co up ion, while 6.11% pe cei e co up ion
as a e y se e e obs acle o hei business.
Fu he mo e, o e 40% o he esponden s
claim ha o al annual cos o hei ope a ions
would dec ease i co up ion was no longe an
obs acle (no enclosed in he able). The mean
alue o expec ed cos dec ease epo ed by
he i ms is 11.75%.
In o de o explo e whe he such a i udes
on co up ion s em om eal cases o co up
ac i i ies o me ely pe cep ions on he sp ead o
co up ion in hei en i onmen , he esponden s
ha e been asked o epo i any gi s o in o mal
paymen s had been expec ed o eques ed
om hem when mee ing ax o i cials, ob aining
an impo license o ob aining an ope a ing
license. The su ey da a indica e ha 5-6% o
any in e ac ions wi h he go e nmen au ho i ies
a e ollowed by eques ing o expec ed gi s o
in o mal paymen s om he i ms. Addi ionally,
3 ou o 29 i ms ha ha e a emp ed o secu e
EM_4_2018.indd 52EM_4_2018.indd 52 28.11.2018 13:12:4928.11.2018 13:12:49
53
4, XXI, 2018
Economics
a go e nmen con ac decla ed ha a ce ain
pe cen age (2-8%) o he con ac alue “would
ypically be paid in in o mal paymen s o gi s
o secu e he con ac ”. Fi ms epo ha hey
pay ce ain pe cen ages o o al annual sales as
in o mal paymen s o gi s o public o i cials o
“ge hings done” wi h ega d o cus oms, axes,
licenses, egula ions, se ices e c. Abou 8% o
he su eyed i ms epo ha in hei business
en i onmen , an es ablishmen such as hei s
pays a ce ain pe cen age o o al annual sales
in in o mal paymen s o gi s o public o i cials.
The pe cep ions o he esponden s on how
common i is o i ms in hei line o business o
ha e o pay some i egula “addi ional paymen s
o gi s” o ge hings done wi h ega d o cus oms,
axes, licenses, egula ions, se ices e c. a e
p esen ed in Tab. 2 ( he esponses a e assigned
sco es om 1 = Ne e o 6 = Always). Only
a qua e o esponden s belie e ha i is ne e
common o i ms o pay addi ional paymen s
o gi s o such pu poses, while o 12% o he
su eyed i ms, i is qui e common o make such
paymen s. Beside pe cep ions on how common
i is o i ms o pay addi ional paymen s and gi s,
he esponden s ha e epo ed on he equency
o paymen s o di e en pu poses ( o deal wi h
cus oms, cou s and axes).
I is assumed ha he main eason o
i ms’ making uno i cial paymen s o gi s is
o hem o gain ad an ages in he d a ing o
laws, dec ees, egula ions, and o he binding
go e nmen decisions. The ollowing able
p esen s he ex en o wha p i a e paymen s/
gi s o o he bene i s o pa liamen a ians
and go e nmen , egional o local o i cials o
a ec hei o es o he con en o go e nmen
dec ees, had a di ec impac on business o he
su eyed i ms (Tab. 3).
No obs acle 42.50%
Mino obs acle 21.39%
Mode a e obs acle 15.28%
Majo obs acle 11.11%
Ve y se e e obs acle 6.11%
Don’ know / No answe 3.61%
To al 100%
Sou ce: BEEPS V (EBRD, 2015)
Tab. 1: To wha deg ee is co up ion an obs acle in business en i onmen
I is common o pay
addi ional paymen s
“ o ge hings done”
To deal wi h
cus oms/impo s
To deal wi h
cou s
To deal wi h axes
and ax collec ion
Ne e 25.28% 41.39% 42.50% 39.72%
Seldom 21.94% 15.83% 14.17% 17.22%
Some imes 24.17% 14.17% 14.17% 14.72%
F equen ly 8.33% 4.17% 5.56% 3.61%
Ve y equen ly 2.50% 1.11% 0.83% 1.11%
Always 0.83% 0.56% 0.56% 0.00%
Don’ know 16.94% 22.78% 22.22% 23.61%
To al 100% 100% 100% 100%
Sou ce: BEEPS V (EBRD, 2015)
Tab. 2: The equency o in o mal paymen s
EM_4_2018.indd 53EM_4_2018.indd 53 28.11.2018 13:12:4928.11.2018 13:12:49
54 2018, XXI, 4
Ekonomie
Desc ip i e s a is ics o he a iables used
in he analysis ha ollows is p esen ed in
Tab. 4. Obse a ions wi h missing alues (Don’
know, Does no apply) ha e been omi ed p io
o he calcula ions.
3.2 K uskal-Wallis Equali y-o -
Popula ions Rank Tes
The nex s ep o he analysis is o es whe he
he pe cep ions o co up ion as an obs acle in
he business en i onmen in Se bia di e based
No impac Mino Mode a e Majo Decisi e N/A
To a ec pa liamen a y
o es 59.17% 7.50% 3.61% 1.39% 0.83% 27.50%
To a ec go e nmen
dec ees 59.72% 6.39% 3.61% 1.67% 0.83% 27.78%
To a ec local and
egional o i cials 55.83% 9.17% 4.72% 1.67% 1.11% 27.50%
Sou ce: BEEPS V (EBRD, 2015)
Va iable Obs. Mean S . de . Min Max
Co up ion as obs acle o business 202 1.148515 1.307355 0 4
F equency o paymen s – cus oms 202 1.792079 1.068004 1 6
F equency o paymen s – cou s 202 1.861386 1.115514 1 6
F equency o paymen s – axes 202 1.831683 1.022817 1 5
Impac o paymen s – pa liamen a ians 202 1.301980 0.714384 1 5
Impac o paymen s – go e nmen 202 1.297030 0.740372 1 5
Impac o paymen s – local/ eg. o i cials 202 1.396040 0.829544 1 5
Cou s as obs acle o business 202 0.861386 1.226001 0 4
Poli ical ins abili y as obs acle 202 1.752475 1.333875 0 4
Sou ce: BEEPS V (EBRD, 2015)
Dependen a iable Independen
a iable d χ2 (wi h ies) p- alue
Co up ion as obs acle Indus y 2 6.882 0.0320
Region 3 9.973 0.0188
Uno i cial paymen s – Cus oms Size o locali y 3 8.838 0.0315
Uno i cial paymen s – Cou s Region 3 9.729 0.0210
Impac o paymen s – pa liamen a ians Indus y 2 8.105 0.0174
Impac o paymen s – go e nmen dec ees Indus y 2 13.052 0.0015
Sou ce: au ho ’s calcula ions based on BEEPS V (EBRD, 2015)
Tab. 3: The impac o uno i cial paymen s
Tab. 4: Desc ip i e s a is ics
Tab. 5: K uskal-Wallis es
EM_4_2018.indd 54EM_4_2018.indd 54 28.11.2018 13:12:4928.11.2018 13:12:49
55
4, XXI, 2018
Economics
on dis inc i e ea u es o he i ms ( i m size,
egion, indus y o he size o locali y whe e i ms
ope a e, numbe o yea s o ope a ion). The
da a in ou sample mee he assump ions ha
unde pin he use o K uskal-Wallis H es . The
esul s o he K uskal-Wallis H a e p esen ed in
Tab. 5 (only esul s a p < 0.05 signi i cance le el
a e displayed).
The K uskal-Wallis H es showed
ha he e was a s a is ically signi i can
di e ence in pe cei ing co up ion as an
obs acle o business ope a ions be ween he
manu ac u ing, e ail and o he se ices, as
well as be ween he ou egions o Se bia.
Also, equency o uno i cial paymen s/gi s ha
i ms would make in a gi en yea o deal wi h
cus oms di e signi i can ly be ween locali ies
o di e en sizes, while equency o uno i cial
gi s paid o deal wi h cou s a e signi i can ly
di e en be ween Se bian egions. The
esul s also indica e ha he e is a s a is ically
signi i can di e ence be ween i ms om
di e en indus ies ega ding he ex en o
which p i a e paymen s o pa liamen a ians
(paid o a ec hei o es) ha e a di ec impac
on hei business. Belonging o manu ac u ing,
e ail o o he se ices indus y is also a alid
p edic o o he ex en ha p i a e paymen s o
go e nmen o i cials (paid o a ec he con en
o go e nmen dec ees) ha e a di ec impac on
i ms’ business.
The esul s o he pe o med es p o e ha ,
depending on he indus y hey belong o, egion
hey ope a e in, o size o hei ci y/ own, i ms
ha e di e en pe cep ions o how co up ion, in
he o m o uno i cial paymen s o gi s o public
o i cials, a ec s hei business ope a ions. The
ollowing analysis is aimed a explo ing he
causes ha de e mine such pe cep ions.
3.3 Es ima ion o he Pe cei ed
Co up ion De e minan s
The su ey da a on co up ion in Se bian
business en i onmen indica e a signi i can
p e alence o co up ion, as pe cei ed by
economic ac o s. In he a emp o explo e he
unde lying causes o co up ion pe cep ions,
we es whe he he a i udes on co up ion
a e de e mined by eal cases o co up ion
o pe cep ions o he impac ha in o mal
paymen s may ha e on he business
ope a ions. Speci ying he o de ed logis ic
eg ession model, we es he explana o y
powe o se e al po en ial de e minan s o
a i udes on co up ion. The esponse a iable
in he model is de i ned as he deg ee o which
co up ion p esen s an obs acle o business
ope a ions o esponden s, measu ed by
5 poin Like scale. The explana o y a iables
used in he model include: he equencies
o uno i cial paymen s o gi s o di e en
pu poses (dealing wi h cus oms, cou s o
O de ed logis ic eg ession Numbe o obs = 202
LR chi2(5) = 50.63
P ob > chi2 = 0.0000
Pseudo R2 = 0.0893
Log likelihood = -258.2548
Co up ion as obs acle Coe . S d. E . Z P>|z| [95% Con . In e al]
Uno i cial paymen s-cus oms .3581645 .1489654 2.40 0.016 .0661977 .6501313
Uno i cial paymen s- axes .4267126 .1554440 2.75 0.006 .1220479 .7313772
Impac -local/ egional o i cials .3267984 .1629264 2.01 0.045 .0074685 .6461284
Cou s as obs acle .2146813 .1148418 1.87 0.062 -.0104044 .439767
Poli ical ins abili y as obs acle .1670583 .1030101 1.62 0.105 -.0348378 .3689545
/cu 1 2.113232 .3855439 (Ancilla y pa ame e s)
/cu 2 3.030764 .4147334
/cu 3 4.192927 .4714150
/cu 4 5.335898 .5401861
Sou ce: au ho ’s calcula ions based on BEEPS V (EBRD, 2015)
Tab. 6: Reg ession esul s o de e minan s o pe cei ed co up ion
EM_4_2018.indd 55EM_4_2018.indd 55 28.11.2018 13:12:4928.11.2018 13:12:49
56 2018, XXI, 4
Ekonomie
axes), he impac ha uno i cial paymen s o
public o i cials (pa liamen a ians, go e nmen
o i cials, local and egional o i cials) ha e on
he business ope a ions o he esponden s, as
well as a i udes o he esponden s ela ed o
he quali y o he judicial sys em and poli ical
s abili y.
The empi ical es ima es o he ologi model
a e p esen ed in Tab. 6.
The model summa y indica es he speci i cs
o he i ed model. The LR Chi-Squa e o
50.63 wi h a p- alue o 0.0000 implies ha he
model as a whole is s a is ically signi i can ,
as compa ed o he null model wi h no
p edic o s. I con i ms ha leas one o he
p edic o s’ eg ession coe i cien is no equal
o ze o. Deg ees o eedom o he Chi-Squa e
dis ibu ion used o es he LR Chi-Squa e
s a is ic a e de i ned by he numbe o p edic o s
in he model. The alue o pseudo-R-squa ed is
subjec o a a ie y o in e p e a ions; howe e
i does no indica e he p opo ion o a iance
o he esponse a iable explained by he
p edic o s, as he R-squa e ound in OLS
eg ession. Ancilla y pa ame e s (cu poin s)
shown in he eg ession ou pu a e used o
di e en ia e adjacen le els o he dependan
a iable, indica ing he cu s in he con inuous
la en a iable. Since ou esponse a iable has
5 le els, he numbe o cu poin s ( h esholds)
is 4. Fo example, cu 1 di e en ia es he lowes
le el o he esponse a iable (co up ion is no
obs acle o business ope a ions) om all he
highe le els o his a iable, gi en ha alues
o all p edic o s is ze o. In ou case, all alues
o la en a iable unde 2.11 would imply ha
co up ion is no obs acle o business, gi en ha
i ms ne e make in o mal paymen s o gi s o
deal wi h cus oms o axes; ha hese kinds o
paymen s ha e no impac on i ms’ ope a ions,
and ha cou s and poli ical ins abili y a e no
pe cei ed as obs acles o business.
In ou model, business ac o s ha a e
equen ly exposed o he unde akings o
paying uno i cial paymen s o gi s in o de
o deal wi h cus oms and axes end o
app ehend co up ion as a se e e obs acle
o hei business ope a ions. Acco ding o he
alues o he o de ed log-odds eg ession
coe i cien s, o a one uni inc ease in hese
wo p edic o s he pe cep ions o co up ion
a e expec ed o change in he o de ed log-
odds scale by 0.36 and 0.43 espec i ely, o he
a iables held cons an . Fo example, a one
uni inc ease in equency o in o mal paymen s
aimed a “ge ing hings done” in dealing wi h
he cus oms would esul in 0.36 uni inc ease
in he o de ed log-odds o esponse a iable
o be in a highe ca ego y (highe deg ee o
wha co up ion is pe cei ed as an obs acle o
business ope a ions), while o he a iables in
he model a e held cons an .
Simila o his, esponden s whose
ope a ions ha e been subjec o decisi e impac
o uno i cial paymen s, gi s o o he bene i s o
local o egional go e nmen o i cials, end o
pe cei e co up ion as ha m ul o hei business.
The o de ed log-odds es ima e o a one uni
inc ease in his p edic o on he expec ed le el
o he dependen a iable is 0.33, gi en he
o he a iables a e held cons an in he model.
On he o he hand, pe cep ions o esponden s
ela ed o he judicial e i ciency o poli ical
s abili y as de e minan s o business ope a ions
do no ha e s a is ical signi i cance (alpha le el
se a 0.05) in explaining co up ion pe cep ions.
4. Discussion and In e p e a ion
o Resea ch Resul s
I is clea om ou esul s ha i ms ha e
di e en pe cep ions o how co up ion, in he
o m o uno i cial paymen s o gi s o public
o i cials, a ec s hei business ope a ions,
depending on he indus y hey belong o,
egion hey ope a e in, o size o hei locali y
(see pa 3.3). The empi ical analysis indica es
ha he e a e signi i can di e ences in
pe cei ing co up ion as an obs acle o business
ope a ions be ween he manu ac u ing, e ail
and o he se ices, as well as be ween he
egions o he analysed coun y. Pe cep ions
on equency o uno i cial paymen s ha i ms
make o deal wi h cus oms di e signi i can ly
be ween i ms om locali ies o di e en sizes,
while pe cep ions on equency o uno i cial
paymen s o deal wi h cou s a e signi i can ly
di e en be ween egions. These i ndings
a e in line wi h he esul s o ecen empi ical
analysis o he business consequences o
local b ibe y en i onmen s (Hanoushek &
Kochano a, 2016), indica ing ha bu eauc a ic
co up ion is a local phenomenon, ha depends
no only on coun y, bu also indus y, i m and
ma ke size. Fu he mo e, such i ndings imply
he impo ance o iden i ying he po en ial
de e minan s o a i udes on co up ion.
The esul s o he o de ed logis ic eg ession
models a e mos commonly in e p e ed in
EM_4_2018.indd 56EM_4_2018.indd 56 28.11.2018 13:12:5028.11.2018 13:12:50
63
4, XXI, 2018
Economics
om Ten Yea s o C oss-Na ional Empi ical
Resea ch? Annual Re iew o Poli ical Science,
10(1), 211-244. h p s://doi.o g/10.1146/annu e .
polisci.10.081205.095418.
Van Duyne, P., & S occo, E. (2012).
Co up ion Policy in Se bia – om black box
o anspa en policy making. Ois e wijk: Wol
Legal Publishe s.
Wol e, R., & Gould, W. (1998). An
app oxima e likelihood- a io es o o dinal
esponse models. S a a Technical Bulle in,
7(42).
UNODC. (2013). Co up ion and C ime in
Se bia: The Impac o B ibe y and O he C ime
on P i a e En e p ise. Vienna: Uni ed Na ions
O i ce on D ugs and C ime.
Wo ld Economic Fo um. (2016). Global
Compe i i eness Repo . Gene a: Wo ld
Economic Fo um.
p o . d Ma ija Džunić, Ph.D.
Uni e si y o Niš
Facul y o Economics
Se bia
[email p o ec ed]
p o . d Na aša Golubo ić, Ph.D.
Uni e si y o Niš
Facul y o Economics
Se bia
[email p o ec ed]
EM_4_2018.indd 63EM_4_2018.indd 63 28.11.2018 13:12:5128.11.2018 13:12:51
64 2018, XXI, 4
Ekonomie
Abs ac
PERCEIVED CORRUPTION IN BUSINESS ENVIRONMENT: EXPLORING THE
UNDERLYING CAUSES
Ma ija Džunić, Na aša Golubo ić
This pape p esen s an empi ical con ibu ion o he li e a u e ha ends o explain a ia ions in
co up ion pe cep ions. D awing on heo e ical assump ions abou he impac o co up ion on he
quali y o he business en i onmen , we explo e possible de e minan s o pe cei ed co up ion.
We a gue ha explaining he de e minan s o co up ion pe cep ions could be a aluable inpu o
c ea ing e ec i e an i-co up ion policies. In he pape , we pe o m a de ailed analysis on he case
o Se bia, a ansi ion coun y wi h ela i ely widesp ead co up ion in he business en i onmen .
Using he non-pa ame ic analysis o a iance es , we examine he signi i cance o di e ences
in he pe cei ed sp ead o co up ion ac oss a numbe o independen a iables ( egions, size o
locali y, indus y, size, and yea s o ope a ion). Speci ying an o de ed logis ic eg ession model, we
es ima e he ela ion be ween co up ion pe cep ions and a numbe o po en ial causes: equencies
o uno i cial paymen s o gi s o di e en pu poses, he impac ha uno i cial paymen s o public
o i cials ha e on he business ope a ions o he esponden s, as well as a i udes o he esponden s
abou he quali y o he judicial sys em and poli ical s abili y. We i nd ha pe cep ions o co up ion
as an obs acle o business ope a ions can be linked o he equency o uno i cial paymen s o deal
wi h cus oms and axes, as well as he impac o uno i cial paymen s o local o egional go e nmen
o i cials on business pe o mance o he esponden s. The esul s p o ide use ul insigh s in o wha
policy measu es a e necessa y o educe he le el o co up ion, as well as how he e ec s o such
measu es can be assessed.
Key Wo ds: Co up ion, business en i onmen , in o mal paymen s.
JEL Classi i ca ion: D73, H11.
DOI: 10.15240/ ul/001/2018-4-004
EM_4_2018.indd 64EM_4_2018.indd 64 28.11.2018 13:12:5128.11.2018 13:12:51