Business Licensing

AfghanlsLan Þubllc Þollcy 8esearch CrganlzaLlon


May 2014

Worklng Þaper



8us|ness L|cens|ng |n Afghan|stan:
Þrocedura| keform or Inst|tut|ona|
kegress|on?
www.appro.org.af 2

Acknow|edgements
1hls paper ls based on Lwo research pro[ecLs funded by Lhe World 8ank's lnLernaLlonal llnance
CorporaLlon (llC) ln 2010 and 2011.


About the kesearchers
1he AÞÞ8C research Leam responslble for Lhls paper are (ln alphabeLlcal order): Ahmad Shaheer
Anll, Saeed ÞarLo, Lhsan SaadaL, Mohsln usyan, and !os WlnLers.

Saeed ÞarLo and !os WlnLers auLhored Lhls reporL. 8ebecca Cang edlLed an earller verslon of Lhls
reporL.

AÞÞ8C ls graLeful Lo Lhe many Lraders who agreed Lo be lnLervlewed aL mulLlple Llmes and spenL
many hours wlLh AÞÞ8C researchers. AÞÞ8C ls also graLeful Lo numerous lndlvlduals from
AfghanlsLan Chamber of Commerce and lndusLrles and Lhe MlnlsLry of Commerce and lndusLry.


About AÞÞkC
AfghanlsLan Þubllc Þollcy 8esearch CrganlzaLlon (AÞÞ8C) ls an lndependenL soclal research
organlzaLlon promoLlng soclal and pollcy learnlng Lo beneflL developmenL and reconsLrucLlon
efforLs ln AfghanlsLan. AÞÞ8C ls a non-proflL, non-governmenL organlzaLlon, headquarLered ln
kabul, AfghanlsLan. AÞÞ8C's mlsslon ls Lo measure developmenL progress agalnsL sLraLeglc
reconsLrucLlon ob[ecLlves Lo provlde lnslghLs on how Lo lmprove performance agalnsL Lhe
developmenL mllesLones seL by Lhe Afghan governmenL and lnLernaLlonal donors. AÞÞ8C
conducLs applled research, carrles ouL evaluaLlons, and provldes Lralnlng on pollcy analysls,
research meLhods, MonlLorlng and LvaluaLlons, and research meLhods.

lor more lnformaLlon, see: www.appro.org.af
ConLacL: mall[appro.org.af


AÞÞ8C ls responslble for all omlsslons and errors.


ÞhoLograph: !os WlnLers





© 2014. AfghanlsLan Þubllc Þollcy 8esearch CrganlzaLlon. Some rlghLs reserved. 1hls publlcaLlon
may be reproduced, sLored ln a reLrleval sysLem or LransmlLLed for non-commerclal purposes
only and wlLh wrlLLen credlL Lo AÞÞ8C and Lhe auLhor(s). Where Lhls publlcaLlon ls reproduced,
sLored or LransmlLLed elecLronlcally, a llnk Lo AÞÞ8C's webslLe aL www.appro.org.af should be
provlded. Any oLher use of Lhls publlcaLlon requlres prlor wrlLLen permlsslon whlch may be
obLalned by wrlLlng Lo: mall[appro.org.af
www.appro.org.af 3
1ab|e of Contents

Lxecut|ve Summary......................................................................................................4
key kecommendat|ons ................................................................................................S
1. Introduct|on.............................................................................................................7
2. Cb[ect|ves and Methodo|ogy....................................................................................8
3. Cverv|ew: 8us|ness L|cens|ng |n Afghan|stan............................................................9
4. I|nd|ngs..................................................................................................................11
4.1 New L|censes................................................................................................................ 11
4.2 L|cense kenewa|s ......................................................................................................... 1S
4.3 I|nd|ngs from Survey Data............................................................................................ 18
S. D|scuss|on ..............................................................................................................22
6. kecommendat|ons .................................................................................................24
Annex 1: Survey I|nd|ngs, Demograph|c Cbservat|ons ...............................................27
Append|x 1: Internat|ona| Chamber of Commerce Ant|-corrupt|on ku|es of Conduct for
8us|nesses and Nat|ona| Governments.......................................................................30

www.appro.org.af 4
Lxecut|ve Summary
Slnce 2004 Lhe CovernmenL of AfghanlsLan has shown a greaL wllllngness Lo reform lLs prlvaLe
secLor, easlng Lhe way for new buslnesses and lmprovlng Lhe operaLlng envlronmenL for Lhose
already ln exlsLence. 1hls openness Lo change was noLed by Lhe World 8ank, whlch descrlbed
AfghanlsLan as Lhe Lop 'confllcL-affecLed' buslness reformer of 2004. Cf parLlcular noLe were
reforms LhaL cuL new buslness appllcaLlon procedures from 28 Lo 1, appllcaLlon processlng Llmes
from 90 Lo 7 days, and new reglsLrles for properLy and credlL Lo faclllLaLe Lhe provlslon of loans.
AL Lhe London Conference ln early 2006 Lhe CovernmenL of AfghanlsLan furLher commlLLed Lo
slmpllfylng and sLreamllnlng lnvesLmenL-relaLed leglslaLlon, regulaLlons, and procedures by Lhe
end of 2007. 1hls commlLmenL was lncorporaLed lnLo Lhe AfghanlsLan naLlonal uevelopmenL
SLraLegy (AnuS) for 2008 - 2013, whlch prlorlLlzes poverLy reducLlon and susLalnable
developmenL Lhrough prlvaLe secLor-led markeL growLh and reform as a ma[or feaLure of
naLlonal reconsLrucLlon. lnLernaLlonally funded reconsLrucLlon and developmenL lnlLlaLlves
coupled wlLh ongolng lmprovemenLs Lo AfghanlsLan's buslness envlronmenL conLlnue Lo
faclllLaLe a surge ln Lhe number of reglsLered domesLlc and lnLernaLlonal buslnesses.

uesplLe compleLlng a number of ln-depLh sLrucLural reforms, Lhe CovernmenL of AfghanlsLan
and lLs lnLernaLlonal donors have been less successful ln susLalnlng and lnsLlLuLlonallzaLlon of
Lhese pro[ecLs. ln parLlcular, llLLle aLLenLlon has been pald Lo effecLlve monlLorlng of procedural
changes or Lo creaLlng lncenLlves Lo brlng abouL behavloural change among publlc and prlvaLe
acLors. 1he resulL has been a sLrong Lendency for lnsLlLuLlonal regresslon Lo pre-reform sLaLus,
and Lhus renewed opporLunlLles for bureaucraLlc delay and corrupLlon. An lmporLanL example of
Lhls, and Lhe rlsks Lo economlc growLh and good governance lL creaLes, can be seen ln reforms
Lo Lhe buslness llcenslng secLor. Whlle successes were lnlLlally dramaLlc, reduclng requlred
procedural sLeps Lo elghL from over flfLy beLween 2004 and 2006, by Lhe Llme of Lhls sLudy,
requlred sLeps were back up Lo over LwenLy and appllcanLs almosL unlversally expecLed Lo pay
some form of llllclL fee, usually aL mulLlple sLages of Lhe process and aL splrallng cosLs. 1hls
regresslon demonsLraLes Lhe vulnerablllLy of buslness acLors and Lhe sLrengLh of offlclal
lmpunlLy - an awareness LhaL has led Lo a loss of confldence among buslness acLors, and a
correspondlng lncrease ln negaLlve lncenLlves Lo engage ln corrupL pracLlces.

1hrough Lhe lens of buslness llcenslng reforms, Lhls worklng paper examlnes progress and
remalnlng challenges under Lhe economlc reconsLrucLlon models currenLly ln use by Lhe
CovernmenL of AfghanlsLan and lLs lnLernaLlonal donors. 1o Lhls end, Lhe paper compares Lhe
procedural sLeps for buslness llcenslng and renewal offlclally mandaLed by Lhe MlnlsLry of
Commerce and lndusLry and Lhose acLually needed Lo compleLe Lhe process. Survey daLa and
resulLlng analysls Lhen plnpolnL areas of conLlnued dlfflculLy for appllcanLs/lnvesLors, lncludlng
boLLlenecks, dupllcaLlve or unnecessary processes, and opporLunlLles for corrupL pracLlces Lo
emerge and become flrmly embedded.

1he followlng recommendaLlons are lnLended Lo supporL Lhe CovernmenL of AfghanlsLan, lLs
formal lnsLlLuLlons, and lLs lnLernaLlonal donors ln Lhelr ongolng efforLs Lo brlng susLalnable
sLrucLural reform and lncreased Lransparency and efflclency Lo Lhe buslness llcenslng reglme ln
AfghanlsLan.

www.appro.org.af 3
key kecommendat|ons
• ln addlLlon Lo sLrucLural reform, efforLs have Lo be made Lo affecL and shape acLual
behavlour by Lhe prlvaLe and publlc users of Lhe sysLem. Þubllc campalgns and changes
ln Lhe educaLlon sysLem are among Lhe Lools LhaL may be used Lo ralse awareness abouL
corrupLlon.
• 8elaLlng Lo Lhe above polnL, Lhere needs Lo be recognlLlon LhaL publlc awareness
campalgns, unllke sLrucLural reform, can Lake years Lo become lnsLlLuLlonallzed and
have a measurable lmpacL on Lhe general publlc's perspecLlve on corrupLlon.
• CurrenLly, buslness llcenses are valld for only one year. Much of Lhe bureaucraLlc delay
and opporLunlLles for corrupLlon LhroughouL Lhe buslness llcenslng reglme could be
ellmlnaLed by exLendlng Lhe duraLlon of buslness llcenses from one Lo Lhree, or even
flve years.
• 1he mosL ob[ecLlonable requlremenL of Lhe MlnlsLry of llnance, accordlng Lo llcense
holders and llcense renewers, ls LhaL of presenLlng buslness accounLs as parL of Lhe
llcenslng renewal process. 1haL Lhls ls a legal requlremenL - and noL a !" $%&'(
mechanlsm for sollclLlng brlbes by MlnlsLry offlclals - musL be made clear Lo Lhe
appllcanLs. 1hese requlremenLs, as well as a clear and conclse explanaLlon for Lhelr
need, should be posLed on Lhe MlnlsLry's webslLe and ln all offlces where renewal
procedures Lake place.
• Afghan buslnesses should be mandaLed by law Lo esLabllsh fully audlLable accounLlng
sysLems. Þossesslon of compleLe and audlLable accounLs wlll acL as a deLerrenL Lo
corrupL offlclals aL several key phases of Lhe llcenslng process.
• Cnly deslgnaLed governmenL offlclals should be permlLLed Lo vlew buslness accounLs. A
ma[or concern of buslness acLors ls LhaL wldespread knowledge regardlng Lhelr level of
wealLh may lncrease rlsks of harassmenL, exLorLlon, kldnapplng, or oLher forms of
crlmlnal acLlvlLy. Lnsurlng confldenLlallLy wlLhln Lhe llcenslng sysLem ls a key lncenLlve Lo
securlng Lhe full parLlclpaLlon of buslness acLors.
• 1he Afghan buslness cusLom of glvlng or recelvlng small paymenLs ln exchange for
servlces, whlch can be seen as peLLy corrupLlon, ls llkely Lo conLlnue for Lhe foreseeable
fuLure. Whlle a compleLe ban ls noL pracLlcal or advlsable, lL ls posslble Lo regulaLe and
monlLor whaL consLlLuLes an accepLable graLulLy and when a graLulLy becomes a brlbe.
1he lnLernaLlonal Chamber of Commerce (lCC) 8ules of ConducL and 8ecommendaLlons
for CombaLlng LxLorLlon and 8rlbery provlde a helpful model for such pro[ecLs.
• 1he Afghan Chamber of Commerce and lndusLry (ACCl) requlres ongolng Lechnlcal
supporL Lo help consolldaLe lLs auLhorlLy as a buslness-secLor assoclaLlve lnsLlLuLlon.
ACCl musL be asslsLed Lo bulld lLs leglLlmacy wlLhln Lhe buslness communlLy by
exhlblLlng an ablllLy Lo perform key Lasks, parLlcularly Lhrough advocacy campalgn.
• CompeLlLlon over access Lo resources and scope of auLhorlLy among key mlnlsLrles and
lnsLlLuLlons ls undermlnlng naLlonal efforLs Lo lmprove Lhe buslness operaLlng
envlronmenL. ConsulLaLlon wlLh represenLaLlves of key mlnlsLrles, AfghanlsLan
lnvesLmenL SupporL Agency, ACCl and Lhe wlder buslness communlLy are needed Lo
explore long Lerm sLraLegles for addresslng amblgulLles ln Lhe dlvlslon of labour and
resources among mlnlsLrles and admlnlsLraLlve deparLmenLs, as a key componenL of Lhe
broader goal of lnsLlLuLlng good governance.
• 8eform efforLs should embrace grassrooLs lnnovaLlons almed aL lncreaslng lnsLlLuLlonal
efflclency, such as Lhe role of admlnlsLraLlve flxers, or )(*+,-".)%/,. 8y formally
www.appro.org.af 6
leglLlmlzlng such pracLlces, Lhey can be regulaLed Lo ellmlnaLe opporLunlLles for
corrupLlon whlle mlnlmlzlng LransacLlon cosLs for llcense appllcanLs. 8egulaLlons could
lnclude formal cerLlflcaLlon, reporLlng requlremenLs, flnanclal dlsclosure, and Lax
paymenL on commlsslons earned among oLhers.
www.appro.org.af 7
1. Introduct|on
Slnce 2004 Lhe CovernmenL of AfghanlsLan has shown greaL wllllngness Lo reform Lhe prlvaLe
secLor, easlng Lhe way for new buslnesses Lo esLabllsh, and lmprovlng Lhe operaLlng
envlronmenL for Lhose already ln exlsLence. 1hls openness Lo change was noLed by Lhe World
8ank, whlch descrlbed AfghanlsLan as Lhe Lop 'confllcL-affecLed' buslness reformer of 2004.
1
Cf
parLlcular noLe were reforms LhaL reduced Lhe sLeps ln new buslness reglsLraLlon process from
28 Lo 1, appllcaLlon processlng Llmes from 90 Lo 7 days, and new reglsLrles for properLy Lo serve
as collaLeral for obLalnlng loans.
2
AL Lhe London Conference ln early 2006 Lhe CovernmenL of
AfghanlsLan furLher commlLLed Lo slmpllfylng and sLreamllnlng lnvesLmenL-relaLed leglslaLlon,
regulaLlons, and procedures by Lhe end of 2007.
3
1hls commlLmenL was lncorporaLed lnLo
AfghanlsLan naLlonal uevelopmenL SLraLegy (AnuS) for 2008 - 2013, whlch prlorlLlzes poverLy
reducLlon and susLalnable developmenL Lhrough prlvaLe secLor led markeL growLh and reform as
a ma[or feaLure of naLlonal reconsLrucLlon.
4
lnLernaLlonally funded reconsLrucLlon and
developmenL lnlLlaLlves coupled wlLh ongolng lmprovemenLs Lo AfghanlsLan's buslness
envlronmenL are lnLended Lo faclllLaLe a surge ln Lhe number of reglsLered domesLlc and
lnLernaLlonal buslnesses.

A key ob[ecLlve for Lhe CovernmenL of AfghanlsLan and lLs lnLernaLlonal donors remalns
regulaLory reform and sLrucLural change Lo bulld on prevlous lmprovemenLs Lo Lhe buslness
envlronmenL. 1hrough conLlnued reforms naLlonal and lnLernaLlonal sLakeholders hope Lo
susLaln exlsLlng domesLlc and lnLernaLlonal prlvaLe lnvesLmenL ln AfghanlsLan and Lo aLLracL new
lnvesLors ln Lhe fuLure. Much progress has been made ln Lhls regard slnce Lhe esLabllshmenL of
AfghanlsLan lnvesLmenL SupporL Agency (AlSA) ln 2003 and Lhrough exLenslve supporL Lo Lhe
MlnlsLry of Commerce and lndusLry (MoCl). Cver 20,000 prlvaLe buslnesses have been
reglsLered by AlSA offlces LhroughouL Lhe counLry, whlle approxlmaLely flfLy new buslnesses a
day are belng reglsLered wlLh Lhe MoCl.

1here ls general saLlsfacLlon wlLh lmprovemenLs Lo AfghanlsLan's buslness operaLlng
envlronmenL among domesLlc and lnLernaLlonal buslnesses. Powever, lL ls also evldenL LhaL a
number of boLLlenecks conLlnue Lo lmpose slgnlflcanL burdens on Lhe buslness communlLy
lncludlng varlous forms of delay, exLorLlon, and onerous bureaucraLlc requlremenLs. 1hls paper
ls based on Lwo research pro[ecLs whlch mapped and assessed Lhe processes of acqulrlng a
buslness llcense and renewlng an exlsLlng buslness llcense. 1he flndlngs reporLed here are
lnLended Lo supporL Lhe CovernmenL of AfghanlsLan and lLs lnLernaLlonal donors ln Lhelr
conLlnued efforLs Lo sLreamllne new buslness reglsLraLlon and buslness llcense renewal
procedures.

SecLlon 2 ouLllnes Lhe sLudy's ob[ecLlves and meLhodology. SecLlon 3 examlnes reforms Lo daLe
ln Lhe arena of buslness llcenslng and llcense renewal, hlghllghLlng Lhe work of Lhe ueparLmenL
of Llcenslng, MlnlsLry of Commerce and lndusLry (uL-MoCl), Lhelr key accompllshmenLs and
remalnlng challenges. SecLlon 4 provldes a comparlson beLween offlclal and acLual procedures

1
World 8ank (2006), ºuolng 8uslness ln 2006: CreaLlng !obs," 6.
2
01+!2
3
AfghanlsLan CompacL (2006).
4
AfghanlsLan naLlonal uevelopmenL SLraLegy (2008).
www.appro.org.af 8
for acqulrlng new llcenses and llcense renewals, whlle SecLlon 3 offers an analysls of Lhe lmpacL
of Lhls dlscrepancy on buslness acLors ln AfghanlsLan. SecLlon 6 concludes and provldes a serles
of recommendaLlons for conslderaLlon by domesLlc and lnLernaLlonal pollcy makers Lo supporL
ongolng reforms ln llcenslng and llcense renewal ln AfghanlsLan.
2. Cb[ect|ves and Methodo|ogy
1hls paper reporLs on Lhe flndlngs from Lwo separaLely commlssloned sLudles. 1he flrsL sLudy
focused on acqulrlng a buslness llcense wlLh a focus on Lhe MlnlsLry of Commerce and lndusLry.
1he second sLudy focused on renewlng an exlsLlng buslness llcense wlLh a focus on Lhe MlnlsLry
of llnance (Mol). 1he comblned ob[ecLlves of Lhese sLudles were Lhus Lo:
- uocumenL offlclal procedures requlred Lo obLaln and renew buslness llcenses Lhrough Lhe
ueparLmenL of Llcenslng, MlnlsLry of Commerce and lndusLry
- uocumenL Lhe sLeps, Llme, and cosLs of obLalnlng and renewlng a buslness llcense ln
pracLlce lncludlng llclL and llllclL requlremenLs
- ÞlnpolnL areas of conLlnued dlfflculLy for appllcanLs lncludlng boLLlenecks, dupllcaLlve or
unnecessary processes, and corrupLlon, and
- CeneraLe recommendaLlons for fuLure lnLervenLlons ln llcenslng reform.

1he flrsL sLudy was carrled ouL ln consulLaLlon and cooperaLlon wlLh uL-MoCl. lnlLlally,
researchers used a mapplng exerclse Lo dellneaLe offlclal and acLual sLeps requlred Lo obLaln
new and renew exlsLlng buslness llcenses Lhrough Lhe uL-MoCl. 1he offlclal sLeps were
deLermlned Lhrough lnLervlews wlLh key lnformanLs aL uL-MoCl, whlle Lhe acLual sLeps were
mapped ouL by underLaklng Lhe process of acqulrlng a Lrade llcense and Lhrough focus group
dlscusslons wlLh buslness acLors on Lhe Loplc of llcense renewal.

8esearchers Lhen soughL Lo verlfy Lhe appllcanLs' acLual experlences and ldenLlfy Lhe consLralnLs
ln Lhe llcenslng processes. 1hls was achleved by convenlng Lwo focus group dlscusslons ln March
and Aprll 2011 wlLh acLlve Lraders ln Lhe offlces of AÞÞ8C and AfghanlsLan Chamber of
Commerce and lndusLry (ACCl). 1he resulLlng daLa was furLher verlfled Lhrough a survey of
llcense appllcanLs aL uL-MoCl conducLed over a perlod of Lhree weeks durlng !une and !uly
2011, resulLlng ln daLa from 132 appllcanLs for boLh new llcenses and llcense renewals.

ldenLlfylng process boLLlenecks ln buslness llcenslng renewal carrled ouL aL Mol was Lhe focus of
Lhe second sLudy. 1o galn lnslghL lnLo common boLLlenecks, researchers soughL Lo conducL ln-
depLh lnLervlews wlLh offlclals ln relevanL deparLmenLs aL Mol as well as offlcers aL Lhe Plgh
Cfflce of CverslghL (PCC). lormal approaches Lo Mol offlclals for lnLervlews were meL wlLh
lndlfference and a general lack of wllllngness Lo engage. As an alLernaLlve, Lwo consulLaLlon
meeLlngs were organlzed aL AÞÞ8C's offlces ln SepLember 2011 wlLh lnvlLees from Mol, PCC,
oLher relevanL bodles, and llcenslng renewal appllcanLs. Whlle Lhls dld noL generaLe new daLa on
Mol pracLlces, AÞÞ8C researchers were able Lo gauge Mol offlclals' reacLlons Lo Lhe sLudy's
prellmlnary flndlngs, lncludlng allegaLlons of wldespread corrupLlon LhroughouL Lhe llcenslng
renewal process. 1hese consulLaLlon meeLlngs also served Lo faclllLaLe dlalogue among
sLakeholders, wlLh Lhe goal of moblllzlng pollLlcal wlll Lo address ldenLlfled lssues ldenLlfled by
Lhe research.

www.appro.org.af 9
lL should be noLed LhaL Lhese sLudles focused only on Lhe experlences of Afghan naLlonal lndlvlduals and
flrms. AlLhough one forelgn flrm was engaged durlng Lhe survey phase, no forelgn lndlvlduals were
lncluded as key lnformanLs or focus group parLlclpanLs. lorelgn buslness acLors' experlences wlLh llcenslng
or oLher admlnlsLraLlve Lasks ls an lmporLanL area for research, however, wlLh parLlcular emphasls on
forelgn lnvesLors' responses Lo corrupLlon and Lhe lmpacL of Lhose responses on corrupL pracLlces.
3. Cverv|ew: 8us|ness L|cens|ng |n Afghan|stan
uesplLe slgnlflcanL growLh ln prlvaLe secLor acLlvlLy as a resulL of reconsLrucLlon and reform
efforLs slnce 2001 a number of ma[or consLralnLs remaln, wlLh poor governance and
lnsLlLuLlonallzed corrupLlon assumlng promlnence.
3
A 2008 survey commlssloned by Lhe World
8ank and Lhe unlLed klngdom's ueparLmenL for lnLernaLlonal uevelopmenL (uflu) reporLed LhaL
alLhough prlvaLe secLor acLors were ºquleLly resolvlng and managlng Lhe bewllderlng array of
consLralnLs Lhey face," growLh and markeL developmenL conLlnued Lo be hlndered by weak
pollcy enforcemenL, dlsorder, corrupLlon, and access Lo resources among oLher facLors.
6
ln
parLlcular, corrupLlon and weak pollcy enforcemenL were seen by Lhe respondenLs as a
consLralnL Lo buslness reglsLraLlon and llcenslng.
7
1hese consLralnLs are ºmore or less Lhe same"
as Lhose reporLed ln a slmllar survey from 2003, buL wlLh ºmarked deLerloraLlon ln pollcy
enforcemenL and securlLy relaLed Lo crlme, LhefL, and dlsorder."
8
AfghanlsLan's buslness
regulaLory envlronmenL ls currenLly ranked 167 ouL of 183 economles, down from 162 ln 2009.
9


As ln oLher counLrles, operaLlng a buslness ln AfghanlsLan requlres a llcense. Llcenses for
lndlvldual Lraders or Lradlng companles are awarded by Lhe ueparLmenL of Llcenslng aL Lhe
MlnlsLry of Commerce and lndusLry (uL-MoCl), whlle Lhe AfghanlsLan lnvesLmenL SupporL
Agency (AlSA) granLs all oLher Lypes of llcenses ln con[uncLlon wlLh Lhe relevanL mlnlsLrles. Whlle
lndlvldual Lraders are requlred Lo reglsLer only wlLh Lhe uL-MoCl, Lradlng companles are
addlLlonally requlred Lo reglsLer wlLh Lhe Commerclal CourL. All llcenses are valld for a perlod of
one year, afLer whlch Lhe llcense holder musL apply for renewal.

AlLhough Lhls sLudy focuses on Lhe llcenslng processes of MoCl, a brlef descrlpLlon of AlSA's role
and funcLlons ls lmporLanL for undersLandlng Lhe full conLexL buslness llcenslng ln AfghanlsLan.

3
See AÞÞ8C's Worklng Þaper on CorrupLlon and ÞrlvaLe SecLor uevelopmenL, avallable aL:
hLLp://www.appro.org.af/ÞubllcaLlons.hLml.
6
World 8ank / uflu (2008). 1he AfghanlsLan lnvesLmenL CllmaLe ln 2008. (kabul: World 8ank and uflu),
§3 and 6.
7
1he reporL noLes: ºLarger flrms generally percelve corrupLlon as a blgger lssue, alLhough many small
flrms reporL lnformal glfLs or paymenLs ln Lhelr deallngs wlLh publlc offlclals, especlally for Lhe
reglsLraLlon process, securlng operaLlng llcenses, or obLalnlng a waLer connecLlon. Cver half Lhe flrms ln
PeraL, !alalabad, and kabul reporLed corrupLlon as a severe consLralnL. Cver 20 percenL of flrms ln
PeraL, kandahar, and Mazar-e-Sharlf (compared wlLh [usL 4° of flrms ln kabul and 6° of Lhose ln
!alalabad) reporLed LhaL an lnformal glfL was expecLed or requesLed for Lhe reglsLraLlon process. lor an
operaLlng llcense, Lwo ouL of Lhree flrms ln Mazar-e-Sharlf and Lwo ouL of flve ln !alalabad reporL paylng
brlbes. 1o secure an lmporL llcense, abouL half of Lhe flrms ln Mazar-e-Sharlf and PeraL, and a Lhlrd of
Lhose ln !alalabad, pay a brlbe or make a glfL." World 8ank/uflu (2008), §2.31.
8
01+!2
9
CurrenL rank derlved from Lhe World 8ank's Lase of uolng 8uslness lndex, accessed aL
hLLp://daLa.worldbank.org/lndlcaLor/lC.8uS.LASL.xC on nov. 6, 2011. 2009 daLa clLed ln World 8ank
(2009), §9.
www.appro.org.af 10
AlSA was launched ln 2003 as parL of efforLs Lo sLreamllne Lhe buslness llcenslng process, due ln
large parL Lo percelved bureaucraLlc weaknesses aL Lhe uL-MoCl. AlSA was creaLed Lo serve as a
'one-sLop-shop' for obLalnlng or renewlng llcenses for a range of buslness models. AlSA lssues
llcenses for all prlvaLe secLor buslnesses excepL Lradlng companles (reglsLered wlLh MoCl) and
small reLall buslnesses (reglsLered wlLh Lhe kabul MunlclpallLy).
10
AppllcanLs are asslsLed ln
compleLlng Lhe forms requlred Lo lnlLlaLe Lhe reglsLraLlon process or Lo renew a llcense. Cn
behalf of Lhe buslness appllcanL, and dependlng on Lhe speclflc requlremenLs for each llcense
Lype, AlSA asslsLs Lhe appllcanLs ln followlng up on Lhelr llcense appllcaLlons Lhrough Lhe
MlnlsLry of !usLlce, MoCl, Lhe Commerclal CourL, and Mol. Securlng a new buslness llcense
under AlSA's sLreamllned appllcaLlon sysLem Lakes an average of flve weeks whlle llcense
renewal can Lake several monLhs. A key dlfference beLween AlSA and uL-MoCl ls LhaL AlSA
compleLes all sLages of Lhe reglsLraLlon processes on behalf of Lhe appllcanL whlle uL-MoCl
procedures requlre Lhe appllcanL Lo underLake Lhese processes hlm/herself or uLlllze Lhe
servlces of seml-offlclal agenLs.

lL ls worLh polnLlng ouL LhaL AlSA was creaLed ln 2004 Lo remove bureaucraLlc hurdles Lo
buslness acLlvlLy and as a non-governmenLal body LhaL acLed efflclenLly and effecLlvely ln
supporL of Lhe fasL emerglng prlvaLe secLor. AlSA's role was Lo ease Lhe admlnlsLraLlve burden
on Lhe uL-MoCl and ald lL Lo re-assume lLs rlghLful role as a key governmenL mlnlsLry.
11
8efore
2004 acqulrlng a buslness llcense wlLh Lhe uL-MoCl was esLlmaLed Lo Lake as many as 33 sLeps
and up Lo elghL weeks. 1hrough use of Lhe ºlnvesLor 8oad Map" dlagnosLlc Lool
12
, Lhe
CovernmenL of AfghanlsLan and lLs advlsors successfully reduced Lhe number of slgnaLures
needed Lo obLaln a llcense Lo flve and Lhe Llme requlred Lo beLween flve and seven days. Whlle
applylng for a buslness llcense wlLh uL-MoCl ls more efflclenL Lhan lL was ln 2004, uL-MoCl
procedures conLlnue Lo be hampered by excesslve bureaucracy and opporLunlLles for
corrupLlon, however. 1here ls sLrong evldence boLh wlLhln Lhls sLudy and oLhers Lo supporL Lhe
World 8ank's flndlng LhaL Lhe uL-MoCl's llcenslng process has ºsllpped" from Lhe sLreamllned
flve sLeps back Lo a burdensome 17 sLages for lndlvldual Lraders and 18 for lncorporaLed Lradlng
companles.
13


8enewlng an exlsLlng buslness llcense ls descrlbed as even more compllcaLed and Llme-
consumlng Lhan applylng for a new llcense, glven hlgher lncenLlves for corrupLlon and Lax fraud
as well as bureaucraLlc lnefflclencles aL relevanL governmenL mlnlsLrles. lncreased lncenLlves for
fraud are llkely relaLed Lo Lhe precondlLlon LhaL all Lax paymenLs musL be flnallzed wlLh Mol for
Lhe prevlous flnanclal year before a renewal can be lssued. 8ecause many buslnesses avold
paymenL of buslness Laxes or keep less-Lhan-accuraLe accounLs of Lhelr acLlvlLles, llcense
renewal appllcanLs are commonly sub[ecL Lo an lnvesLlgaLlon by Mol offlclals. 1hose wlLh sound
record keeplng are noL lmmune from lnspecLlon elLher. lnspecLlons usually Lake around Lwo
monLhs. 1he llkellhood of belng lnspecLed aL a slow pace, comblned wlLh lack of lnsLlLuLlonal

10
See www.alsa.org.af for furLher deLalls.
11
lL ls worLh noLlng LhaL AlSA's very success has become an obsLacle Lo MoCl galnlng full leglLlmacy and
becomlng esLabllshed as a key mlnlsLry charged wlLh faclllLaLlng and overseelng prlvaLe secLor
developmenL, however. AlSA's currenL sLrengLh and success lncreaslngly ralse quesLlons regardlng
MoCl's conLlnulng weaknesses. Þaradoxlcally, lL ls now MoCl's need for conLlnued supporL Lo
accompllsh whaL AlSA has already achleved.
12
lor more on Lhe uSAlu/World 8ank funded ºlnvesLor 8oadmap," see
hLLp://www.commerce.gov.af/en/page/333.
13
World 8ank / uflu (2008).
www.appro.org.af 11
overslghL, low salarles, and llmlLed capaclLy aL MoCl and Mol ln parLlcular creaLes an
envlronmenL LhaL enables and LoleraLes corrupL behavlor. 1hus, for a muLually beneflclal fee,
pald by Lhe appllcanL or a )(*+,-".)%/ (agenL) on Lhe appllcanL's behalf, llcense renewal
appllcaLlons are fasL-Lracked whlle Lhe governmenL sLands Lo lose Laxes payable by buslnesses.

1o undersLand where and how Lhe currenL weaknesses and boLLlenecks are occurrlng, a
comparlson was made beLween Lhe offlclally mandaLed and acLual sLeps for obLalnlng a new
llcense and renewlng an exlsLlng buslness llcense Lhrough uL-MoCl. 1he nexL secLlon reporLs on
Lhe flndlngs from Lhls comparlson based on Lhe daLa collecLed from key lnformanLs and survey
parLlclpanLs.
4. I|nd|ngs
4.1 New L|censes
Cfflclally, obLalnlng a new llcense Lhrough uL-MoCl should Lake no more Lhan Lwo days provldlng Lhe
paperwork ls ln order and necessary clearances have been obLalned by Lhe appllcanL from Lhe relevanL
mlnlsLrles.
14
1he sLeps are descrlbed as follows:

1. Approach ulrecLoraLe of 1rade Llcenses (u1L), uL-MoCl, for referral Lo Lhe SecLlon for
lndlvlduals 1raders or SecLlon for Companles.
2. CollecL and compleLe Lhe 8eglsLraLlon lorm.
3. SubmlL Lhe 8eglsLraLlon lorm Lo Lhe uL-MoCl 1reasury ueparLmenL. CollecL Lhe 8equesL for
ÞaymenL lorm Lo be Laken Lo a deslgnaLed 8ank for deposlL.
4. ueposlL paymenL for Lhe llcense fee aL Lhe 8ank. CollecL recelpL.
3. SubmlL recelpL Lo Lhe SecLlon for lndlvldual 1raders or SecLlon for Companles, for
verlflcaLlon of Lhe deposlL.
6. Þroceed Lo Lhe CenLral 8eglsLraLlon Cfflce, Mol, Lo reglsLer Lhe buslness. CollecL
conflrmaLlon leLLer.
7. SubmlL conflrmaLlon leLLer Lo Lhe SecLlon for lndlvldual 1raders or SecLlon for Companles Lo
obLaln clearance for lssuance of llcense.
8. SubmlL llcense clearance Lo Lhe Pead of uL-MoCl, for slgnaLure and lssuance.

1he offlclal cosLs assoclaLed wlLh obLalnlng a new llcense are llsLed ln 1able 1. Cfflclals aL uL-
MoCl sLaLe LhaL beLween 1.3 and 2 days are requlred Lo compleLe Lhe llcenslng process, aL a
LoLal dlrecL cosL of 2,700 Afghanls (Afs)
13
for an lndlvldual llcense.

1o verlfy Lhls lnformaLlon, a llcense appllcanL was followed by an AÞÞ8C researcher Lo
documenL Lhe varlous sLages of llcense acqulslLlon Lhrough uL-MoCl. ln conLrasL Lo Lhe offlclally
sLaLed elghL sLeps, Lhe documenLaLlon Lhe acLual process revealed LhaL Lhere were many more
sLeps Lhan elghL. A key flndlng of Lhls phase of Lhe sLudy was Lhe prevalence of flxers, or
komlshenkars, LhroughouL Lhe Lwo processes of acqulrlng a new llcense and renewlng an
exlsLlng one. SLudles slnce around 2007 have shown a slgnlflcanL lncrease ln Lhe use of Lhe

14
AppllcanLs are requlred Lo have key documenLs ln place before beglnnlng Lhe process, lncludlng an
'organlzaLlonal consLlLuLlon' descrlblng Lhe naLure of Lhe buslness and Lhe roles of key lndlvlduals who
make up Lhe flrm.
13
1he sLandard exchange raLe for 30 Afghanls ls 1 uSu.
www.appro.org.af 12
komlshenkars as a means Lo overcome Lhe mulLlple burdens of mulLl-sLep, compllcaLed, and
ofLen obscure admlnlsLraLlve processes.
16
1he lnLervlewed Lraders reporLed LhaL Lhelr own
deallngs wlLh offlclals and Lhose conducLed Lhrough komlshenkars are plagued by corrupLlon,
boLh aL Lhe lnsLlLuLlonal level and Lhrough smaller, more arblLrary brlbes.

1ab|e 1. L|cense Acqu|s|t|on Costs (Cff|c|a|)
uomesLlc lndlvldual 1rader Llcense 33 uSu
uomesLlc Company 1rader Llcense 60 uSu
lorelgn lndlvldual 1rader Llcense 1,400 uSu (lncludes 30 uSu as ºspeclal charge"
17
)
lorelgn Company 1rader Llcense 1,400 uSu (lncludes 30 uSu as ºspeclal charge")
lorelgn-uomesLlc !olnL venLure 1rader Llcense 1,400 uSu (lncludes 30 uSu as ºspeclal charge")
1radlng Agency of a lorelgn Company ln
AfghanlsLan
1,900 uSu (lncludes 20 uSu as ºspeclal charge")
1ranslL and lrelghL lorwardlng 8uslness Llcense 60 uSu
CooperaLlve Company Llcense SLarLlng aL 60 uSu (lncreases dependlng on
amounL of Lhe sLarL-up caplLal requlred)
8rokerlng Llcense 33 uSu
Source: uL-MoCl

Slnce 2006 Lhe prevalence of corrupLlon has lncreased and lL has become more lnsLlLuLlonallzed
accompanled wlLh Lhe lnevlLable backslldlng of sLrucLural reforms. komlshenkars are LhoughL Lo
parLlclpaLe ln Lhe glvlng and Laklng of brlbes 44° of Lhe Llme ln 2010.
18
lor a shorL Llme,
komlshenkars were formally recognlzed by Lhe cusLoms deparLmenL as seml-offlclal clerks for
Lhe processlng of Lraders' paperwork.
19
AlLhough Lhls ls yeL anoLher layer of formallzed
corrupLlon faced by buslness acLors, ln many ways Lhe komlshenkars acL as a buffer beLween
corrupL or less capaclLaLed offlclals and Lraders by navlgaLlng Lhe fasLesL and leasL expenslve
paLh Lhrough whaL ls a complex, unclear, and aL Llmes predaLory process.
20


1he acLual process for acqulrlng a new llcense ls as follows:

1. SubmlL a wrlLLen leLLer of requesL for a llcense Lo uL-MoCl. LeLLer wrlLers are locaLed aL Lhe
enLrance of Lhe MoCl, Lhelr fee ls approxlmaLely 30 Afs and Lhls Lakes abouL Len mlnuLes.
! AL Lhls polnL, Lhe appllcanL can choose Lo use Lhe servlces of any number of flxers who
congregaLe ln fronL of Lhe MoCl and are ofLen called over by Lhe leLLer wrlLers Lo asslsL
poLenLlal cllenLs. A flxer usually charges 7,000 Afs for compleLlon of Lhe llcenslng
process, offerlng dellvery ln approxlmaLely Lhree days.
2. Þroceed Lo Lhe SecLlon for lndlvldual 1raders or SecLlon for Companles, uL-MoCl.
! SubmlL Lhe llcense requesL leLLer Lo Lhe relevanL secLlon.
! CollecL a Þersonal lnformaLlon lorm and an AssessmenL lorm for a fee of 220 Afs. (1he
fee llsLed on Lhe Þersonal lnformaLlon lorm ls 100 Afs, Lhe AssessmenL lorm does noL
noLe a prlce.)

16
lWA (2007).
17
lL ls unclear whaL Lhls 'speclal charge' covers, or wheLher ls lL parL of scheduled fees.
18
lWA (2007) and lWA (2010).
19
key lnformanL lnLervlewed ln kabul, !une 2011. SubsequenLly, ln !uly-AugusL 2011, key lnformanLs from
Lhe CusLoms ueparLmenL of Lhe MlnlsLry of llnance reporLed LhaL komlshenkars were no longer
formally recognlzed ln Lhe cusLoms sysLem.
20
lWA (2010).
www.appro.org.af 13
! CompleLe Lhe flrsL parL of Lhe Þersonal lnformaLlon lorm.
3. Þroceed Lo Lhe MlnlsLry of lnLerlor (Mol) Lo submlL parL Lwo of Lhe Þersonal lnformaLlon
lorm, Lhe Crlmlnal Clearance lorm. 1hls clearance usually Lakes abouL 24 hours.
4. Pavlng collecLed Lhe Crlmlnal Clearance lorm, proceed Lo Lhe Mol Lo submlL parL Lhree of
Lhe Þersonal lnformaLlon lorm, whlch lncludes a requesL for a 1ax ldenLlflcaLlon number
(1ln).
3. 8eLurn Lo Lhe SecLlon for lndlvldual 1raders or SecLlon for Companles, uL-MoCl, Lo submlL
Lhe Mol sLamped Crlmlnal Clearance lorm. CollecL llcense fee paymenL form Lo be Laken Lo
Lhe uL-MoCl 1reasury ueparLmenL.
6. Þroceed Lo Lhe 1reasury ueparLmenL for verlflcaLlon of Lhe sLamped Crlmlnal Clearance
lorm and Lhe llcense fee paymenL form.
7. Þroceed Lo Lhe uL-MoCl ueparLmenL of 8eglsLraLlon, where Lhe appllcanL ls lssued Lwo
requesLs for paymenL (one for 1,300 Afs and Lhe oLher for 130 Afs).
8. Þroceed Lo Lhe deslgnaLed 8ank, Lo submlL Lwo fees (1,300 and 130 Afs). CollecL recelpLs.
9. 8eLurn Lo Lhe 1reasury ueparLmenL Lo submlL recelpLs for verlflcaLlon, afLer whlch Lhe
appllcanL ls referred Lo Lhe ueparLmenL of 8eglsLraLlon.
10. Þroceed Lo Lhe ueparLmenL of 8eglsLraLlon, where Lhe paymenL ls agaln verlfled and Lhe
appllcanL ls referred back Lo Lhe offlce of Lhe u1L, uL-MoCl.
11. Þroceed Lo Lhe u1L Lo recelve an order for lssuance of a 1ln.
12. Þroceed Lo Lhe CenLral 8eglsLraLlon Cfflce, Mol, Lo submlL Lhe 1ln requesL. 1he lssuance of
1ln usually Lakes 24 hours.
13. 8eLurn Lo Lhe ueparLmenL of 8eglsLraLlon, uL-MoCl, Lo recelve Lhe 8eglsLraLlon lorm. 1he
appllcanL ls Lhen lssued Lwo more requesLs for paymenL: 300 Afs for Lhe ºrlghL Lo reglsLer"
(3%4 " 5%1'6 and 666 Afs for Lhe ºllrsL AppllcaLlon Surcharge" (3%478 9,-%:"-).
14. 8eLurn Lo Lhe 8ank Lo deposlL second round of fees.
13. 8eLurn Lo Lhe ueparLmenL of 8eglsLraLlon Lo submlL paymenL recelpLs for verlflcaLlon. 1he
appllcanL ls Lhen referred back Lo Lhe CenLral 8eglsLraLlon Cfflce, Mol.
16. 8eLurn Lo Lhe CenLral 8eglsLraLlon Cfflce for submlsslon of Lhe appllcanL's compleLed and
slgned 8eglsLraLlon lorm. 1he appllcanL ls Lold Lo reLurn Lhe nexL day Lo collecL Lhe verlfled
and reglsLered form.
17. 8eLurn Lo Lhe CenLral 8eglsLraLlon Cfflce Lo collecL 8eglsLraLlon lorm. 1he appllcanL ls Lold Lo
dellver parL of Lhe form Lo Lhe Archlves ueparLmenL, uL-MoCl.
18. Þroceed Lo Lhe Archlves ueparLmenL Lo submlL parL of Lhe 8eglsLraLlon lorm.
19. 8eLurn Lo Lhe SecLlon for lndlvldual 1raders or SecLlon for Companles Lo collecL a clearance
form for submlsslon Lo Lhe SecreLarlaL, uL-MoCl.
20. Þroceed Lo Lhe SecreLarlaL, uL-MoCl, Lo submlL Lhe llcenslng clearance form. AL Lhls polnL,
Lhe llcense ls lssued ln larsl and Lngllsh and senL Lo Lhe offlce of Lhe u1L, uL-MoCl, for
slgnaLure.
21. Þroceed Lo Lhe u1L. 1he ulrecLor slgns and formally lssues Lhe appllcanL hls/her llcense.

1he 132 appllcanLs surveyed aL uL-MoCl reporLed a greaL deal of varlaLlon ln acLual cosLs for
obLalnlng a new llcense (llgure 1), as well as Lhe unavallablllLy of recelpLs for all of Lhe 'fees'
requesLed by offlclals (llgure 2).

www.appro.org.af 14
I|gure 1. Average Cost per L|cense 1ype


Shadowlng one llcense appllcanL as parL of Lhls sLudy revealed LhaL Lhe acLual cosL of an
lndlvldual Lrader's llcense LoLaled 3,060 Afghanls, excludlng lndlrecL LransporLaLlon cosLs and
oLher Lravel expenses (1able 2). 1hls ls an average dlfference ln dlrecL cosLs of 360 Afs per
llcenslng appllcanL, noL lncludlng Lhe more slgnlflcanL cosLs assoclaLed wlLh procedural delay,
Lravel and losL economlc opporLunlLy.
21


I|gure 2. Þercentage of App||cants kece|v|ng kece|pts for A|| Þayments


Cn average, around flfLy new appllcanLs per day begln Lhe process of obLalnlng a new buslness
llcense. 1he amounL of money LhaL leaks from Lhe process ls Lhus roughly 360 Afs per appllcanL
slx days a week, or approxlmaLely a leak of 108,000 Afs per week.





21
1ravel and LransporL relaLed expenses from kandahar Lo kabul, for example, can cosL around 400 uSu
aL a mlnlmum and range up Lo Lhousands of dollars.
www.appro.org.af 13
1ab|e 2. Ind|rect Costs for Cut-of-kabu| App||cants
1rave| Costs - Cne Way (USD)
Þrov|nce
Iood |n kabu|
(USD per day)
Accommodat|on
(USD per n|ght)
A|r Land
kandahar Lo kabul 10 - 40 10 - 300 100 10 - 40
PeraL Lo kabul // // 120 20 - 40
Mazar Lo kabul // // 100 10 - 20
Pelmand Lo kabul // // 100 7 - 20
!alalabad Lo kabul // // no Commerclal lllghLs 4 - 10
Chaznl Lo kabul // // no Commerclal lllghLs 4 - 10
ÞakLla Lo kabul // // no Commerclal lllghLs 4 - 6
kunduz Lo kabul // // no Commerclal lllghLs 10 - 20
8amyan Lo kabul // // no Commerclal lllghLs 10 - 13
Þarwan Lo kabul // // no Commerclal lllghLs 3 - 3
khosL Lo kabul // // no Commerclal lllghLs 10 -13

1he process descrlbed above ls esLlmaLed Lo Lake an appllcanL beLween Lhree and flve days of
conLlnuous efforL, requlrlng a LoLal of 23 slgnaLures or verlflcaLlons of slgnaLures. uslng a
komlshenkar doubles Lhe cosL of obLalnlng a llcense buL lL also cuLs Lhe average processlng Llme
ln half.
4.2 L|cense kenewa|s
8enewlng a llcense wlLh Lhe uL-MoCl ls a far slower and, by all accounLs, much more palnful
process. 1he offlclal cosLs assoclaLed wlLh Lhe renewal process are llsLed ln 1able 3. Cfflclally,
Lhe process for renewlng a llcense Lhrough uL-MoCl ls as follows:

1. SubmlL an up Lo daLe buslness balance sheeL Lo Lhe 8evenue ueparLmenL and ConLrol
SecLlon, Mol, Lo secure a leLLer of clearance Lo proceed wlLh Lhe renewal process.
2. Cnce clearance ls recelved from Lhe Mol, submlL a wrlLLen leLLer of requesL Lo MoCl for Lhe
renewal of an explred llcense. 1he MoCl wlll refer Lhe requesL Lo Lhe ulrecLoraLe of 1rade
Llcenses (uL), MoCl, for a fee of 30 Afs.
3. Þroceed Lo Lhe uL-MoCl Lo collecL Lwo forms for a charge of 220 Afs (no recelpLs are lssued).
4. Þroceed Lo Lhe ueparLmenL of 8eglsLraLlon, uL-MoCl. Cn Lhe condlLlon LhaL Lhe appllcanL
has already reglsLered Lhe requesL for renewal wlLh Lhe uL-MoCl, Lhe ueparLmenL of
8eglsLraLlon lssues a requesL for paymenL of 1,930 Afs.
3. Þroceed Lo Lhe deslgnaLed 8ank for paymenL and collecLlon of Lhe recelpL.
6. 8eLurn Lo Lhe ueparLmenL of 8eglsLraLlon, where Lhe recelpL ls verlfled and reglsLered. 1he
bank recelpL ls sLamped and reLurned Lo Lhe appllcanL. CollecL Lwo furLher requesLs for
paymenL (130 Afs each) Lo Lhe ueparLmenL of 8eglsLraLlon, for 'sLamplng servlces' and
phoLocopylng charges.
7. Þroceed Lo Lhe 8evenue ueparLmenL and LhereafLer Lhe ConLrol SecLlon, Mol, Lo presenL
buslness accounLs ln order Lo secure clearance on Lax paymenLs and oLher buslness charges
(e.g., lmporL Laxes).
8. Cn recelpL of clearance from Lhe ConLrol ueparLmenL and a referral back Lo Lhe uL-MoCl,
Lhe appllcanL may reLurn Lo uL-MoCl Lo collecL hls/her renewed llcense.
www.appro.org.af 16

1ab|e 3. L|cense kenewa| Costs (Cff|c|a|)
uomesLlc lndlvldual 1rader Llcense 33 uSu
uomesLlc Company 1rader Llcense 30 uSu
lorelgn lndlvldual 1rader Llcense 400 uSu (lncludes 30 uSu as ºspeclal charge"
22
)
lorelgn Company 1rader Llcense 400 uSu (lncludes 30 uSu as ºspeclal charge")
lorelgn-uomesLlc !olnL venLure 1rader Llcense 400 uSu (lncludes 30 uSu as ºspeclal charge")
1radlng Agency of a lorelgn Company ln
AfghanlsLan
900 uSu (lncludes 20 uSu as ºspeclal charge")
1ranslL and lrelghL lorwardlng 8uslness Llcense 30 uSu
CooperaLlve Company Llcense SLarLs from 30 uSu buL may be hlgher dependlng on
amounL of Lhe sLarL-up caplLal
8rokerlng Llcense 33 uSu
Source: uL-MoCl

1o documenL Lhe acLual process, cosLs, and lengLh of Llme for llcense renewals focus group
dlscusslons and lndlvldual lnLervlews were held wlLh a number of experlenced Lraders, Mol
offlclals, and offlcers of Lhe PCCAC.

1he llcenslng renewal process Lakes slgnlflcanLly longer Lhan securlng a new llcense malnly
because Lhe flnanclal sLakes are much hlgher ln Lhe renewal of llcenses, boLh for Lhe appllcanLs
and offlclals lnvolved. 1o lllusLraLe, wlLhouL clearance from Mol Lhe appllcaLlon for renewal
cannoL be submlLLed Lo uL-MoCl. And, wlLhouL a valld llcense goods cannoL be cleared Lhrough
Lhe CusLoms sysLem, rlsklng slgnlflcanL losses Lo lmporLlng Lraders who are requlred Lo pay
sLorage fees Lo Lhe porLs where Lhelr goods are belng held. 1hls rlsk ls compounded for Lraders
of perlshable goods who also lncur spollaLlon-relaLed cosLs.

unllke ln securlng new llcenses, where use of admlnlsLraLlve flxers was seen as more of an lssue
of convenlence, Lraders ln Lhe focus group descrlbed flxers as a pracLlcal necesslLy ln Lhe llcense
renewal process. Lven wlLh Lhelr addlLlonal connecLlons, securlng a llcense renewal can Lake a
flxer as long as Lhree Lo four monLhs and can cosL a conslderable amounL ln llllclL fees. 1raders'
declslon-maklng on wheLher Lo use Lhe servlces of a flxer musL also Lake lnLo accounL Lhe facL
LhaL llcense renewal ls requlred annually.

1he flrsL serlous boLLleneck ln Lhe llcenslng renewal process occurs even prlor Lo submlLLlng an
appllcaLlon Lo Lhe MoCl for llcense renewal, when Lraders are requlred Lo presenL Lhelr buslness
accounLs Lo Lhe Mol Lo obLaln a leLLer of clearance. Cfflclally, Lhe 8evenue ueparLmenL and
ConLrol SecLlon of Lhe Mol are requlred Lo examlne renewal appllcanLs' buslness accounLs Lo
ensure Laxes, duLles, renLs, and oLher obllgaLlons are up Lo daLe. Much of Lhls daLa ls ln facL
already avallable Lo Lhem, ln Lhe daLabase of A?SCuuA, AfghanlsLan's new, fully auLomaLed
cusLoms sysLem. Whlle Lhere are concerns regardlng Lhe accuracy of some of Lhe daLa, Lhe facL
remalns LhaL Lhe AS?CuuA daLabase can and should be used as a ma[or source for much of Lhe
lnformaLlon Mol offlclals demand from llcense renewal appllcanLs.

knowlng LhaL Lhls daLabase exlsLs, requesLs for deLalled lnformaLlon by Mol offlclals are Lhus
undersLood by mosL Lraders Lo be lndlrecL requesLs for brlbes. Whlle Mol offlclals are legally

22
lL ls unclear whaL Lhls 'speclal charge' covers, or wheLher lL ls parL of scheduled fees.
www.appro.org.af 17
mandaLed Lo requesL lnformaLlon LhaL ls requlred by offlclal forms, such as Lhe number of
people employed ln an appllcanL's flrm, renL pald for buslness premlses, properLles owned, dally
sales flgures and so forLh, requesLs are more commonly made noL Lo gaLher Lhe requlred
lnformaLlon buL lnsLead Lo coerce Lraders lnLo paylng brlbes. 8eyond offlclals' lgnorance of daLa
already ln Lhelr possesslon, Lhe connecLlon beLween opporLunlLles for corrupLlon and requesLs
for lnformaLlon ls furLher evldenced by Lhe facL LhaL no maLLer wheLher Lhe appllcanL's flles are
orderly and up-Lo-daLe or genulnely ln need of lnvesLlgaLlon, appllcanLs are almosL unlversally
expecLed Lo proffer 'donaLlons' or glfLs of appreclaLlon Lo secure Lhe Mol's clearance of Lhelr
accounLs. Moreover, each dlvlslon of Lhe Mol lnvolved ln Lhe renewal process ls llkely Lo
demand such a paymenL, ranglng from abouL 600 uSu Lo recelve Lhe appllcaLlon Lo 1000 uSu
for flnal conflrmaLlon and clearance.

1hls sLage of Lhe process alone can Lake many days or even weeks dependlng on Lhe appllcanL's
wllllngness Lo furnlsh Lhe offlcer ln charge wlLh adequaLe paymenL. lor an appllcanL whose
accounLs are noL well malnLalned Lhe llkellhood of paylng mulLlple and excesslve llllclL fees ls
even hlgher. A breakdown of llcense appllcanLs' percepLlons on Lhe causes of boLLlenecks aL Lhe
Mol can be seen ln llgure 3, wlLh un[usLlfled and undocumenLed requesLs for cash paymenLs as
Lhe mosL clLed source.
23


I|gure 3. 8ott|enecks at the M|n|stry of I|nance

.=132

Cnce Lhe leLLer of clearance ls secured from Lhe Mol, appllcanLs face furLher admlnlsLraLlve
boLLlenecks LhroughouL Lhe process, almosL unlversally found when excesslve or dupllcaLlve
bureaucraLlc requlremenLs lncrease opporLunlLles for corrupLlon. 1hese lnclude, for example,
demands for a 'volunLary' graLulLy of beLween 2,000 and 4,000 Afs aL Lhe CenLral SLaLlsLlcal
Cfflce where Lraders are requlred Lo reglsLer Lhelr llcenses. AnoLher common hurdle ln Lhe
renewal process ls ln Lhe u1L, MoCl, where Lraders are requesLed Lo pay beLween 200 and 2,000
Afs as an appllcaLlon processlng fee. Some Lraders reporLed LhaL renewal appllcaLlons were
occaslonally denled by Lhe u1L on Lhe grounds LhaL Lhe goods Lo be lmporLed under Lhe
renewed llcense were currenLly belng lmporLed by oLher llcense holders, Lraders allege LhaL
such refusals are drlven by pressure from hlgher-level offlclals and oLher Lraders who wlsh Lo
monopollze Lhe lmporL markeL for parLlcular goods. AlLhough Lhe u1L has a relaLlvely clean

23
lurLher survey daLa ls avallable ln SecLlon 4.3.
www.appro.org.af 18
repuLaLlon for brlbe Laklng among Lraders (desplLe Lhe paLLerns descrlbed above), flxers
rouLlnely pay furLher graLulLles Lo u1L offlcers of beLween 100 Lo 300 Afs Lo fasL Lrack Lhelr
cllenLs' appllcaLlons.

Much of Lhe corrupLlon ln Lhls arena has become lnsLlLuLlonallzed and ls susLalned Lhrough LaclL
agreemenLs beLween Mol offlcers, flxers, and Lhe renewal appllcanLs Lhemselves. AlLhough
ofLen seen as vlcLlms of corrupLlon, Lraders have aL leasL Lwo lncenLlves noL Lo presenL accuraLe
buslness accounLs or oLher lnformaLlon and lnsLead offer Lo pay brlbes. llrsL, buslness accounLs
and oLher flnanclal daLa can reveal a Lrader's neL worLh Lo oLhers, whlch may lncrease Lhe
llkellhood of kldnapplng, exLorLlon or oLher securlLy LhreaLs for many wealLhy Afghans. Second,
lL ls wldely recognlzed LhaL many Lraders undervalue Lhelr lmporLs, buslness slze, proflLs and
properLles ln order Lo pay less duLles and Laxes Lo Lhe sLaLe.

knowlng Lhls, Mol offlcers seek ouL lncompleLeness, lrregularlLles, and oLher lnadequacles ln
appllcanLs' paperwork ln order Lo [usLlfy Lhelr demand for lnformal paymenLs. WhaL has
developed ls a relaLlvely flxed equlllbrlum beLween offlclals and appllcanLs, a schedule of llllclL
paymenLs whlch ls undocumenLed yeL undersLood by all parLlclpanLs. Whlle Lhere are flrms ln
kabul LhaL offer accounL preparaLlon servlces for beLween 10,000 and 20,000 Afs, many Lraders
prefer Lo conLlnue paylng Lhe Mol Lo keep Lhelr acLual holdlngs unreporLed. 8uL cosLs are
splrallng as corrupLlon becomes lncreaslngly lnsLlLuLlonallzed, lnformanLs suggesLed LhaL
recenLly Lraders have been known Lo pay as much as 200,000 Afs [usL Lo geL Lhrough Lhe flrsL
sLage of Mol clearance. ln order Lo compleLe Lhe renewal process, Lraders descrlbe paylng up Lo
300,000 Afghanls ln LoLal, dependlng on Lhe slze of Lhe buslness, adequacy of buslness records
and Lhe appllcanL's ablllLy Lo pay. lL ls lmporLanL Lo sLaLe agaln LhaL Lhls process ls requlred on an
annual basls.
4.3 I|nd|ngs from Survey Data
ln addlLlon Lo quallLaLlve lnLervlews, lnformanLs were asked Lo parLlclpaLe ln a quanLlLaLlve
survey. 1he survey requesLed LhaL parLlclpanLs provlde deLalled lnformaLlon abouL amounLs pald
durlng Lhe llcenslng renewal process and rank Lhe varlous aspecLs of obLalnlng or renewlng a
buslness llcense accordlng Lo Lhelr Lransparency, helpfulness and ease.

Cf Lhe 132 survey parLlclpanLs almosL 40° reporL paylng beLween 10,000 and 20,000 Afs ln llllclL
fees, whlle over 20° reporL paylng more Lhan 100,000 Afs (llgure 4).

www.appro.org.af 19
I|gure 4. 8reakdown of Amounts Þa|d to kenew a 8us|ness L|cense

.=132

Clven Lhe prevalence of uslng admlnlsLraLlve flxers, renewal appllcanLs were asked Lo esLlmaLe
Lhe cosLs lncurred when uLlllzlng flxers ln Lhe renewal process. When a flxer was used, cosLs
were reporLedly lower, Lhough as seen ln llgure 3, llllclL fees sLlll generally LoLaled beLween
10,000 and 20,000 Afs.

I|gure S. Lst|mated Iees When Us|ng I|xers |n L|cense kenewa|s

.=132

When fees pald by appllcanL Lraders lndependenLly and fees pald Lhrough Lhe use of a flxer are
compared, lL ls clear LhaL Lhe use of a flxer accrues real savlngs Lo Lhe renewal appllcanL.
Averaged ouL, lndlvldual Lraders can spend 200,000 Afs on a renewal appllcaLlon Lhey submlL
Lhemselves, savlng as much as 100,000 Afs Lhrough Lhe parLlal use of a flxer, and almosL 140,000
Afs when a flxer ls used for Lhe whole process (llgure 6).

www.appro.org.af 20
I|gure 6. Average Costs for L|cense kenewa|

.=132

Survey parLlclpanLs were furLher asked wheLher Lhey had recelved any form of recelpL for
paymenLs made Lo Mol and MoCl offlclals durlng Lhe renewal process. Cver 93 percenL of Lhose
surveyed reporLed maklng undocumenLed paymenLs (no recelpLs) Lo Lhe Mol, whlle [usL under
30 percenL reporLed maklng undocumenLed paymenLs Lo Lhe MoCl aL some sLage of Lhe llcense
renewal process (llgure 7).

I|gure 7. Þayments w|thout kece|pt

.=132

Þerhaps noL surprlslngly glven Lhe resulLs descrlbed above, Lhe vasL ma[orlLy of llcense renewal
appllcanLs rely on flxers. WhaL ls lnLeresLlng, however, ls LhaL for Lhe ma[orlLy of Lhose surveyed
(approxlmaLely 80 percenL), Lhe mosL compelllng reason Lo uLlllze Lhe servlces of a flxer were
noL relaLed Lo money, buL Lo savlng Llme (llgure 8). Cf addlLlonal noLe ls Lhe Lhlrd mosL common
reason for uslng a flxer, whlch ls Lraders' unfamlllarlLy wlLh renewal procedures.

www.appro.org.af 21
I|gure 8. keasons for Us|ng a I|xer

.=132

ln a separaLe componenL of Lhe survey, Lraders were asked Lo rank Lhe mlnlsLrles and speclflc
deparLmenLs lnvolved ln llcenslng renewal accordlng Lo Lhelr Lransparency, helpfulness and ease
of use.
24
1he leasL LransparenL polnLs of conLacL accordlng Lo respondenLs were Lhe banks
deslgnaLed for paymenL of llcense fees, Lhe ConLrol SecLlon of Lhe Mol, and Lhe CenLral
8eglsLraLlon Cfflce of Lhe Mol. ln Lerms of helpfulness, Lhe leasL helpful enLlLles were reporLed
Lo be Lhe bank, Lhe ConLrol SecLlon and CenLral 8eglsLraLlon Cfflce aL Lhe Mol. 1he ulrecLoraLe
of 1rade Llcenses and Lhe 1reasury ueparLmenL aL MoCl were also ldenLlfled by llcense renewal
appllcanLs as relaLlvely dlfflculL sLages ln Lhe process. ÞaymenL of fees aL deslgnaLed banks was
conslsLenLly ranked lowesL for Lransparency, helpfulness, and ease, parLlcularly by Lhose
acqulrlng a new llcense.

Abuse of auLhorlLy by offlclals LhroughouL Lhe buslness llcenslng secLor noL only deLracLs from
economlc growLh by ralslng LransacLlon cosLs for Lraders, buL has Lhe knock-on effecL of
undermlnlng Lhe leglLlmacy of Lhe llcenslng reglme as a whole. An example of Lhls ls seen ln Lhe
'guaranLee deposlL,' an addlLlonal fee of beLween 200-300,000 Afs lmposed on Lraders by Lhe
Mol. 1he fee was lnsLlLuLed Lo ensure LhaL lmporLed goods are sold ln AfghanlsLan and noL
lllegally re-exporLed Lo ÞaklsLan, whlch has much hlgher cusLoms duLles on lmporLed goods.
Whlle Lhls deposlL may have a raLlonal basls, lL ls dlfflculL for Lraders Lo see such cosLs (or Lhe
offlclals who enforce Lhem) as leglLlmaLe glven Lhe prevalence of demands for llllclL fees by
offlclals aL almosL every sLage of Lhe renewal process.

1hls senLlmenL was besL lllusLraLed by one focus group parLlclpanL:

[W]e pay for our [lmporLed] merchandlse, Laxes, and wages and reporL all Lhese expenses Lo
cusLoms by recordlng Lhem on Lhe PCu form.
23
When we have pald all LhaL ls due Lo geL Lhe
merchandlse across Lhe border and lL ls recorded, why do we need Lo presenL Lhe MlnlsLry of
llnance wlLh a guaranLee, or flll ouL yeL anoLher llcense appllcaLlon form? lf Lhey have already

24
Survey respondenLs were asked Lo rank each sLep of Lhe process requlred Lo acqulre or renew Lrade
llcenses accordlng Lo Lhe followlng caLegorles: Lransparency (0=noL aL all 1ransparenL, 10=1ransparenL),
helpfulness (0=noL aL all Pelpful, 10=Pelpful), and ease (0=noL aL all easy, 10=Lasy).
23
8esearchers were unable Lo deLermlne Lhe meanlng of Lhls acronym.
www.appro.org.af 22
charged us everyLhlng we owe Lo Lhe governmenL and we are clear, why ls Lhere a need Lo flll ouL
a form Lo say LhaL we have done exacLly LhaL? l Lhlnk all Lhese requlremenLs are puL ln place Lo
lncrease Lhe opporLunlLles for exLorLlng money from Lraders. [CovernmenL offlclals] are [usL
looklng for excuses Lo Lake brlbes. 1hls ls noL good for Lrade and noL good for flghLlng
governmenL corrupLlon.
S. D|scuss|on
ln AfghanlsLan, key governmenLal enLlLles are responslble for esLabllshlng and regulaLlng
accepLable parameLers for prlvaLe secLor acLors across Lhe full specLrum of buslness acLlvlLles. A
slgnlflcanL elemenL ln Lhe managemenL of Lhls secLor perLalns Lo Lhe lssulng and renewal of
buslness llcenses. uesplLe compleLlng a number of ln-depLh sLrucLural reforms Lo Lhe llcenslng
process, Lhe CovernmenL of AfghanlsLan and lLs lnLernaLlonal donors appear Lo have been less
successful ln Lhe ongolng lmplemenLaLlon of Lhese pro[ecLs. ln parLlcular, llLLle aLLenLlon has
been pald Lo effecLlve monlLorlng of procedural changes or Lo lncenLlvlzlng behavloral change
among relevanL offlclals and llcense appllcanLs. 1he resulL has been a sLrong Lendency for
lnsLlLuLlonal regresslon Lo pre-reform sLaLus, and Lhus renewed opporLunlLles for excesslve
bureaucraLlc delay and corrupLlon.

An lmporLanL example of such backslldlng, and Lhe rlsks Lo economlc growLh and good
governance lL creaLes, can be seen ln reforms Lo Lhe buslness llcenslng and renewal secLor.
Successes ln Lhls arena were lnlLlally dramaLlc, reduclng procedures for llcenslng Lo elghL sLeps
from over flfLy beLween 2004 and 2006. 8y Lhe Llme of Lhls sLudy, however, procedural sLeps
were back up Lo over LwenLy and llcense appllcanLs almosL unlversally had Lhe expecLaLlon of
paylng some form of llllclL fee, usually aL mulLlple sLages of Lhe process and aL lncreaslngly
hlgher cosL. ln facL, Lhls cerLalnLy led Lo Lhe creaLlon of a new mechanlsm ln Lhe llcenslng
process: Lhe flxer, or )(*+,-".)%/, who could cuL processlng Llmes by abouL half for double Lhe
average prlce. Whlle Lhe currenL sLaLe of affalrs remalns much lmproved from whaL lL was prlor
Lo 2004, regresslon ln sLrucLural reforms has sharply demonsLraLed Lo buslness acLors Lhelr
vulnerablllLy ln Lhe face of offlclal lmpunlLy. 1hls awareness has led Lo a loss of confldence, and a
correspondlng lncrease ln negaLlve lncenLlves among buslness acLors Lo engage ln corrupLlon.
LlmlLaLlons ln Lhe deslgn and lmplemenLaLlon of Lop-down sLrucLural reforms have Lhus noL only
deLracLed from Lhelr long-Lerm success and susLalnablllLy, buL have conLrlbuLed Lo Lhe
lnsLlLuLlonallzaLlon of old, and new, forms of corrupLlon wlLhln Lhe LargeLed mlnlsLrles.

When asked as parL of AÞÞ8C's survey whaL could be done Lo lmprove Lhe llcense renewal
process, parLlclpanLs had a serles of clear and concreLe suggesLlons (llgure 9). 1he general
accepLance of Lhese suggesLlons among key sLakeholders was verlfled Lhrough ln-depLh
lnLervlews wlLh Lraders, flxers and offlclals wlLhln Lhe Mol and PCCAC. ln addlLlon Lo calls for
lncreased Lransparency, Lraders sLrongly requesLed an exLenslon ln Lhe valldlLy of buslness
llcenses Lo more Lhan one year - Lo Lhree or even flve. 1hls change alone would ellmlnaLe a
greaL deal of Lhe opporLunlLles for corrupLlon and sense of allenaLlon from sLaLe process
experlenced by Lraders and offlclals allke.

www.appro.org.af 23
I|gure 9. 1raders' Suggest|ons for Improvements to the L|cense kenewa| Þrocess

.=132

1he vasL ma[orlLy of lnformanLs and survey parLlclpanLs expressed a sLrong preference for
reforms LhaL would brlng abouL a 'one-sLop-shop' for llcenslng renewal, where Lraders could
have Lhelr records checked, Laxes and oLher fees pald ln a unlfled process LhaL would ellmlnaLe
opporLunlLles for corrupLlon. lL ls llkely LhaL Lraders based Lhls suggesLlon on Lhe demonsLraLed
success of Lhe Afghan lnvesLmenL SupporL Agency (AlSA), whlch now offers a one-sLop-shop for
all processes relaLed Lo non-Lrade relaLed buslness llcenses.
26
As dlscussed above, Lhe AlSA's
success ln sLreamllnlng and lncreaslng Lransparency ln llcenslng pracLlces has only hlghllghLed
Lhe ongolng corrupLlon and bureaucraLlc delay wlLhln llcenslng deparLmenLs aL Lhe Mol and
MoCl. As one llcense renewal appllcanL demanded Lo know, ºwhy should we have Lo go back
and forLh beLween dlfferenL locaLlons Lo process our paperwork? Why can'L we [usL go Lo one
place and do all Lhe Lhlngs requlred of us?" AnoLher puL an even flner polnL on lL, asserLlng LhaL
ºLhe more places you have Lo go Lo [ln Lhe process of renewlng a llcense], Lhe hlgher Lhe
posslblllLy of acLual and arblLrary delays whlch breed brlbery and corrupLlon."

ln addlLlon Lo furLher sLrucLural reforms, naLlonal and lnLernaLlonal efforLs Lo lmprove buslness-
relaLed processes llke llcenslng and llcense renewal musL also Lake lnLo accounL Lhe personal
and professlonal lncenLlves LhaL lnfluence Lraders' behavlor. lor example, accordlng Lo Lhe vasL
ma[orlLy of renewal appllcanLs who conLrlbuLed Lo AÞÞ8C's research, Lhe mosL ob[ecLlonable of
all of Lhe Mol's llcenslng requlremenLs was LhaL of submlLLlng buslness accounLs for lnspecLlon
Lo offlclals. upon an examlnaLlon of whaL maLLers Lo buslness acLors ln AfghanlsLan, Lhe reasons
for Lhls ob[ecLlon become clear. llrsL, formal bookkeeplng ls a rare pracLlce among Afghan
Lraders. 1hls and oLher research by AÞÞ8C has shown, Llme and agaln, LhaL an overwhelmlng
ma[orlLy of buslnesses prefer noL Lo keep formal records of Lhelr buslness LransacLlons. 8easons
for Lhls dlslncllnaLlon Lo rely on documenLaLlon lnclude Lhe wlsh noL Lo reveal moneLary wealLh,
wldespread llllLeracy and an overall lack of buslness accounLlng skllls. Second, as wlLh
enLrepreneurs and buslnesses Lhe world over Afghan buslness acLors make every aLLempL Lo
mlnlmlze cosLs and maxlmlze proflL. 1hls fuels Lhe drlve, for example, Lo flnd loopholes ln Lhe
Mol's LaxaLlon reglme or a mechanlsm Lo undervalue lmporLed goods. 1aken LogeLher, Lhese

26
See SecLlon 3 above, or Lhe AlSA's webslLe aL www.alsa.org.af.
www.appro.org.af 24
and oLher facLors conLlnue Lo moLlvaLe Lraders Lo parLlclpaLe ln corrupL admlnlsLraLlve pracLlces
as a way Lo succeed (or even survlve), boLh personally and professlonally.

lf fuLure endeavors Lo lmprove AfghanlsLan's buslness-relaLed sLrucLures and procedures are Lo
have a poslLlve and durable lmpacL, naLlonal and lnLernaLlonal sLakeholders musL also accounL
for behavloral lncenLlves LhaL wlll faclllLaLe change from Lhe boLLom up. A focus on lnvesLlng
prlvaLe secLor acLors wlLh a sLake ln Lhe ouLcome of reforms could lnclude creaLlng grassrooLs
mechanlsms for consulLaLlon, monlLorlng and advocacy, lncreaslng supporL Lo Lhe AfghanlsLan
Chamber of Commerce and lndusLry as Lhey consolldaLe Lhelr lnsLlLuLlonal role, and
sLrengLhenlng Lhe role of clvll socleLy organlzaLlons devoLed Lo Lhe needs and lnLeresLs of Lhe
buslness communlLy (for example secLoral organlzaLlons, or "'%-%!+%,). 8y Lransformlng
buslness acLors lnLo vesLed sLakeholders wlLh Lhe means of lnfluenclng reform ouLcomes,
buslness acLors Lhemselves wlll be empowered Lo supporL - as opposed Lo undermlne -
sLrucLural and procedural lmprovemenLs Lo AfghanlsLan's buslness operaLlng envlronmenL.
6. kecommendat|ons
• 8eform efforLs have Lhe greaLesL poslLlve lmpacL when based on ln-depLh knowledge of
Lhe resources avallable Lo key acLors, Lhe envlronmenL Lo be reformed, and Lhe
perLlnenL levers LhaL lmpacL acLual behavlor beyond sLrucLural change. Þerhaps mosL
lmporLanLly, reform efforLs should exhlblL wllllngness Lo experlmenL wlLh programmaLlc
alLernaLlves and adapLablllLy ln Lhe face of unexpecLed ouLcomes.

• CurrenLly, buslness llcenses are valld for one year, requlrlng annual renewal. Much of
Lhe bureaucraLlc delay and opporLunlLles for corrupLlon could be alLogeLher ellmlnaLed
by exLendlng Lhe duraLlon of buslness llcenses from one Lo Lhree, or even flve years. 1he
MlnlsLry of Commerce and lndusLry (MoCl) could conLlnue Lo charge llcense holders an
annual fee lf so deslred, buL could avold Lhe Llme and cosLs assoclaLed wlLh yearly
renewal. 1hls would be Lhe easlesL change Lo make, wlLh llkely Lhe greaLesL and mosL
lmmedlaLe lmpacL.
• Accordlng Lo mosL llcense renewal appllcanLs surveyed by AÞÞ8C, Lhe mosL
ob[ecLlonable of all MlnlsLry of llnance (Mol) requlremenLs ls LhaL of presenLlng
buslness accounLs as parL of Lhe renewal process. 1haL Lhls ls a legal requlremenL - and
noL a !" $%&'( mechanlsm for sollclLlng brlbes by Mol offlclals - musL be made clear Lo
llcense renewal appllcanLs. 1hese requlremenLs, as well as a clear and conclse
explanaLlon of Lhelr purpose, should be posLed on Lhe Mol's webslLe. AddlLlonally,
formal noLlces and clear checkllsLs of all llcenslng renewal requlremenLs, wlLh mandaLed
fees, should be posLed ln all offlces where renewal procedures Lake place.

• Afghan buslnesses should be requlred by legal regulaLlon, and supporLed by Lhe Afghan
Chamber of Commerce and lndusLry (ACCl), Lo esLabllsh fully audlLable accounLlng
sysLems. Þossesslon of compleLe and audlLable accounLs wlll acL as a deLerrenL Lo
corrupL offlclals aL several key phases of Lhe llcenslng process. 1hls recommendaLlon ls
also supporLed by lnLernaLlonal Chamber of Commerce (lCC) guldellnes.

• Cnly deslgnaLed governmenL offlclals should be permlLLed Lo vlew buslness accounLs. A
ma[or concern held by many buslness acLors ls wldespread knowledge abouL Lhelr level
www.appro.org.af 23
of wealLh buslness acLlvlLy, whlch may puL Lhem ln danger of harassmenL, exLorLlon,
kldnapplng, or oLher forms of crlmlnal acLlvlLy. Lnsurlng confldenLlallLy wlLhln Lhe
llcenslng sysLem ls a key lncenLlve Lo securlng Lhe full parLlclpaLlon of buslness acLors.

• 1radlLlonal Afghan buslness cusLoms such as glvlng or recelvlng small paymenLs ln
exchange for Lhe compleLlon of servlces are llkely Lo conLlnue Lo play a role ln buslness
LransacLlons ln AfghanlsLan for Lhe foreseeable fuLure. A compleLe ban ls noL pracLlcal or
advlsable, as Lhls would llkely [usL push such behavlors under Lhe Lable. lL ls posslble,
however, Lo regulaLe and monlLor whaL consLlLuLes an accepLable graLulLy and when a
graLulLy becomes a brlbe. lCC 8ules of ConducL and 8ecommendaLlons for CombaLlng
LxLorLlon and 8rlbery provlde a helpful model for Lhe pragmaLlc appllcaLlon of self-
lmposed anLl-corrupLlon efforLs.
27


• 8eform efforLs should embrace grassrooLs lnnovaLlons almed aL lncreaslng lnsLlLuLlonal
efflclency, such as Lhe role of flxers, or )(*+,-".)%/,. 8y formally adopLlng and
leglLlmlzlng such pracLlces, Lhey can be regulaLed Lo ellmlnaLe opporLunlLles for
corrupLlon whlle mlnlmlzlng LransacLlon cosLs for llcense appllcanLs. Such regulaLlons
could lnclude, for example, formal cerLlflcaLlon, reporLlng requlremenLs and flnanclal
dlsclosures, among oLhers.

• lmmedlaLe sLeps musL be Laken Lo ellmlnaLe endemlc levels of corrupLlon Lhe 8evenue
and ConLrol ueparLmenLs aL Lhe Mol Lhrough lnsLlLuLlng a hlgher degree of overslghL
and accounLablllLy. 1o move away from currenL pracLlces, furLher procedural revlslons
and lmproved overslghL mechanlsms are requlred, Lhe laLLer posslbly Lo be
lmplemenLed by a Lhlrd parLy enLlLy such as Lhe Plgh Cfflce of CverslghL and AnLl-
CorrupLlon (PCCAC).
• A sLrong preference was expressed by Lhe vasL ma[orlLy of sLudy parLlclpanLs Lo creaLe a
'one-sLop-shop' for llcenslng renewal, where Lraders could have Lhelr records checked,
Laxes and oLher fees pald ln a unlfled process LhaL would ellmlnaLe opporLunlLles for
corrupLlon. lL ls llkely LhaL Lhls suggesLlon ls based on Lhe demonsLraLed success of Lhe
Afghan lnvesLmenL SupporL Agency (AlSA), whlch could serve as a model for fuLure
reforms.
• 1he ACCl has a mandaLe Lo promoLe Lhe lnLeresLs of Afghan buslnesses and proLecL
Lhose lnLeresLs Lhrough advocacy-relaLed acLlvlLles. ACCl currenLly suffers from a lack of
leglLlmacy wlLhln Lhe buslnesses communlLy, whlch has prevenLed lL from fulfllllng lLs
mandaLe and Lhereby furLher deLracLed from lLs sLaLus among buslness acLors and
offlclals. ACCl requlres ongolng Lechnlcal supporL Lo help consolldaLe lLs auLhorlLy as a
key buslness-secLor lnsLlLuLlon. ACCl wlll have Lo bulld LrusL wlLhln Lhe buslness
communlLy by exhlblLlng Lhelr ablllLy Lo perform key Lasks, such as Lhrough a successful
advocacy campalgn.

27
1he lCC 8ules of ConducL and 8ecommendaLlons for CombaLlng LxLorLlon and 8rlbery ls avallable on Lhe
lccwbo.org webslLe, aL
hLLp://www.lccwbo.org/uploadedllles/lCC/pollcy/anLlcorrupLlon/SLaLemenLs/lCC_8ules_of_ConducL_a
nd_8ecommendaLlons°20_2003°208evlslon.pdf
www.appro.org.af 26
• CompeLlLlon over access Lo resources and scope of auLhorlLy among key mlnlsLrles and
lnsLlLuLlons ls undermlnlng naLlonal efforLs Lo lmprove Lhe buslness operaLlng
envlronmenL. An example of Lhls ls Lhe AlSA, whose success ln sLreamllnlng non-Lrade
buslness llcenslng procedures has furLher undermlned Lhe auLhorlLy and leglLlmacy of
Lhe MoCl, due Lo Lhelr less efflclenL sysLems for lssulng Lrade-relaLed buslness llcenses.
ConsulLaLlon wlLh represenLaLlves of key mlnlsLrles, AlSA, ACCl and Lhe wlder buslness
communlLy are needed Lo explore long Lerm sLraLegles for addresslng amblgulLles ln Lhe
dlvlslon of labor and resources among mlnlsLrles and admlnlsLraLlve deparLmenLs, as an
elemenL of Lhe broader goals of good governance.
• luLure efforLs Lo address reforms ln buslness-relaLed admlnlsLraLlve procedures musL be
fully lncluslve of Lhose who perpeLuaLe and suffer from corrupL pracLlces. lor example,
alLhough Lhe MoCl ls prlmarlly responslble for lssulng/renewlng buslness llcenses,
arblLrary delays and corrupLlon are ofLen encounLered ln Lhe Mol. lor Lhls reason, lL ls
crlLlcal Lo engage a broad range of sLakeholders Lo generaLe cross-secLoral ownershlp
and cooperaLlon ln Lhe reform process. AL a mlnlmum, consulLaLlon should lnvolve Lhe
Mol, MoCl, MlnlsLry of lnLerlor, and oLher key mlnlsLrles, ACCl, AlSA, and buslness
communlLy represenLaLlves, and focus on how Lo besL conLlnue Lo lmplemenL and
enforce changes ln llcenslng procedures. CuLcomes from Lhls dlalogue should furLher
lnclude lnLervenLlons Lo lncrease Lhe lnLegrlLy, lndependence, and auLhorlLy of Lhe
PCCAC ln relaLlon Lo oLher governmenLal agencles and mlnlsLrles.




www.appro.org.af 27
Annex 1: Survey I|nd|ngs, Demograph|c Cbservat|ons
1hls annex hlghllghLs some of Lhe key demographlc characLerlsLlcs of buslness llcense and llcense renewal
appllcanLs ldenLlfled ln Lhe survey componenL of Lhls sLudy.

ApproxlmaLely 70° of new llcense and renewal appllcanLs aL Lhe uL-MoCl's offlces ln kabul are from
buslnesses based ln kabul (llgure 3). Cf appllcanLs surveyed, around 10° were sole proprleLorshlps whlle
Lhe ma[orlLy were enLlLles wlLh beLween 2 and 10 employees. Around 3° of appllcanLs were buslnesses
wlLh more Lhan 10 employees. Cnly one large-scale naLlonal company was encounLered durlng Lhe
survey, wlLh 270 employees, and one forelgn Lradlng company wlLh 300 employees (llgure 4).

llgure 3. Þrovlnclal Crlgln of 1rade 8uslnesses Acqulrlng or 8enewlng Llcenses aL uL-MoCl


llgure 4. Llcense and 8enewal AppllcanL llrms, number of Lmployees


new and renewal llcenses are lssued only Lo Lhe head of Lhe appllcanL's flrm or a deslgnaLed
represenLaLlve. llgure 3 lllusLraLes paLLerns ln Lhe delegaLlon of Lhls Lask.



www.appro.org.af 28
llgure 3. ÞoslLlon of AppllcanLs ln llrms Applylng for new or 8enewed Llcenses


Cf survey parLlclpanLs drawn from uL-MoCl offlces ln kabul, all forelgn owned company offlces
and LranslL and frelghL forwardlng companles were headquarLered ln kabul. Whlle domesLlc
companles engaged ln lmporLlng and exporLlng are prlmarlly based ln kabul, represenLaLlves of
domesLlc Lradlng companles from oLher provlnces were also ldenLlfled (llgure 6). kandahar
appears Lo be Lhe only provlnce oLher Lhan kabul LhaL has a number of lndlvldual Lraders (abouL
3° of appllcanLs surveyed).

llgure 6. Þrovlnclal Spread of 8uslness 1ypes


1ranslL and frelghL forwardlng llcense holders have Lhe hlghesL percenLage of flrms wlLh more
Lhan 10 employees, aL around 30°, whlle only abouL 20° of forelgn owned companles had
more Lhan 10 employees (llgure 7).






www.appro.org.af 29
I|gure 7. Number of Lmp|oyees per I|rm ] L|cense 1ype

www.appro.org.af 30
Append|x 1: Internat|ona| Chamber of Commerce Ant|-
corrupt|on ku|es of Conduct for 8us|nesses and Nat|ona|
Governments
8ox 1. ICC Ant|-corrupt|on ku|es of Conduct for 8us|nesses
lCC offers a range of oLher arLlcles Lo be adhered Lo ln flghLlng corrupLlon as follows:
Art|c|e 1: Þroh|b|t|on of 8r|bery and Lxtort|on - LnLerprlses should prohlblL brlbery and
exLorLlon aL all Llmes and ln any form, wheLher dlrecL or lndlrecL, lncludlng Lhrough agenLs
and oLher lnLermedlarles.
Art|c|e 2: Agents and Cther Intermed|ar|es - LnLerprlses should make Lhelr anLl-corrupLlon
pollcy known Lo all agenLs and oLher lnLermedlarles and make lL clear LhaL Lhey expecL all
acLlvlLles carrled ouL on Lhelr behalf Lo be compllanL wlLh Lhelr pollcy.
Art|c|e 3: Io|nt Ventures and Cutsourc|ng Agreements - LnLerprlses should Lake measures
wlLhln Lhelr power Lo ensure LhaL anLl-brlbery provlslons conslsLenL wlLh Lhese 8ules of
ConducL are accepLed by [olnL-venLure parLners as appllcable Lo Lhe [olnL venLure and by
parLles Lo ouLsourclng agreemenLs.
Art|c|e 4: Þo||t|ca| and Char|tab|e Contr|but|ons and Sponsorsh|ps - LnLerprlses should only
make conLrlbuLlons Lo pollLlcal parLles, parLy offlclals and candldaLes ln accordance wlLh
appllcable laws, and all requlremenLs for publlc dlsclosure should be fully complled wlLh.
Art|c|e S: G|fts, nosp|ta||ty and Lxpenses - LnLerprlses should esLabllsh procedures coverlng
Lhe offer or recelpL of glfLs, hosplLallLy or expenses Lo ensure LhaL such arrangemenLs are
llmlLed Lo reasonable and bona flde expendlLures and do noL lmproperly affecL, or mlghL be
deemed Lo lmproperly affecL, Lhe ouLcome of a procuremenL or oLher buslness LransacLlon.
Art|c|e 6: Iac|||tat|on Þayments - LnLerprlses should noL make faclllLaLlon paymenLs. ln Lhe
evenL LhaL an enLerprlse deLermlnes, afLer approprlaLe managerlal revlew, LhaL faclllLaLlon
paymenLs cannoL be ellmlnaLed enLlrely, lL should esLabllsh conLrols and procedures Lo ensure
LhaL Lhelr use ls llmlLed Lo small paymenLs Lo low-level offlclals for rouLlne acLlons Lo whlch
Lhe enLerprlse ls enLlLled.
Art|c|e 7: Corporate Þo||c|es - 1o prevenL brlbery and exLorLlon, enLerprlses should
lmplemenL comprehenslve pollcles or codes reflecLlng Lhese 8ules of ConducL as well as Lhelr
parLlcular clrcumsLances and speclflc buslness envlronmenL.
Art|c|e 8: I|nanc|a| kecord|ng and Aud|t|ng - All flnanclal LransacLlons musL be properly and
falrly recorded ln approprlaLe books of accounL avallable for lnspecLlon by boards of dlrecLors,
lf appllcable, or a correspondlng body, as well as audlLors.
Art|c|e 9: kespons|b|||t|es - 1he board of dlrecLors or oLher body wlLh ulLlmaLe responslblllLy
for Lhe enLerprlse, should Lake reasonable sLeps Lo ensure compllance wlLh Lhese 8ules of
ConducL.
Source: CAC (2003)

www.appro.org.af 31
8ox 2. ICC ku|es of Conduct and kecommendat|ons for Nat|ona| Governments
Capac|ty 8u||d|ng: CovernmenLs should provlde adequaLe resources - lncludlng fundlng,
sLafflng and Lralnlng - for Lhe organlzaLlons combaLlng corrupLlon. 1hese resources should be
almed aL ralslng publlc awareness of Lhe economlc and soclal consequences of corrupLlon. .
Strengthen Lnforcement: Lven Lhough laws prohlblLlng exLorLlon and brlbery have been on
Lhe books ln vlrLually all counLrles, enforcemenL ls weak or nonexlsLenL ln many counLrles.
CovernmenLs [should] sLrengLhen enforcemenL of anLl-brlbery laws, lncludlng laws
prohlblLlng exLorLlon by publlc offlclals and laws prohlblLlng commerclal brlbery.
CovernmenLs should also esLabllsh reporLlng channels for complalnLs abouL corrupLlon. .
Lconom|c kegu|at|ons: CovernmenLs should, as far as posslble, mlnlmlze Lhe use of sysLems
requlrlng lssuance of lndlvldual permlLs, auLhorlzaLlons, eLc., because such sysLems offer
scope for exLorLlon and brlbery. .
Þub||c Þrocurement: CovernmenLs should commlL Lhemselves Lo lmprove Lransparency ln
publlc procuremenL, lncludlng publlc blddlng, and publlcaLlon of Lhe crlLerla upon whlch
awards are based. . ln parLlcular, ln Lhe case of pro[ecLs flnanced by an lnLernaLlonal flnanclal
lnsLlLuLlon, observance of Lhese prlnclples should be reflecLed ln conLracLual arrangemenLs
beLween Lhe parLles. .
Þo||t|ca| Contr|but|ons: undlsclosed pollLlcal conLrlbuLlons can be a source of abuse.
CovernmenLs should, sub[ecL Lo Lhe speclflcs of Lhelr naLlonal pollLlcal sysLem, regulaLe Lhe
condlLlons under whlch pollLlcal conLrlbuLlons can be made. .
Account|ng and Aud|t|ng: CovernmenLs should requlre audlLlng by lndependenL publlc
audlLors of Lhe accounLs of all economlcally slgnlflcanL enLerprlses. CovernmenLs should also
supporL Lhe adopLlon of sLrong and conslsLenL lnLernaLlonal sLandards for accounLlng and
audlLlng. .
Lxport Cred|t Agenc|es: LxporL credlL agencles should lnclude ln Lhelr condlLlons LhaL
flnanclng and guaranLees are noL provlded for conLracLs secured Lhrough brlbery and LhaL
sancLlons shall be applled ln Lhe evenL of vlolaLlons. .
Cff|c|a| Deve|opment Ass|stance: uevelopmenL asslsLance programmes should requlre sLrlcL
adherence Lo anLl-corrupLlon rules by governmenL offlclals and conLracLors. .
Source: CAC (2003)