Qian, Yingyi
Wo king Pape
The Ins i u ional Founda ions o Ma ke T ansi ion in he
People's Republic o China
ADBI Resea ch Pape Se ies, No. 9
P o ided in Coope a ion wi h:
Asian De elopmen Bank Ins i u e (ADBI), Tokyo
Sugges ed Ci a ion: Qian, Yingyi (2000) : The Ins i u ional Founda ions o Ma ke T ansi ion in he
People's Republic o China, ADBI Resea ch Pape Se ies, No. 9, Asian De elopmen Bank Ins i u e
(ADBI), Tokyo,
h ps://hdl.handle.ne /11540/4116
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/111100
S anda d-Nu zungsbedingungen:
Die Dokumen e au EconS o dü en zu eigenen wissenscha lichen
Zwecken und zum P i a geb auch gespeiche und kopie we den.
Sie dü en die Dokumen e nich ü ö en liche ode komme zielle
Zwecke e iel äl igen, ö en lich auss ellen, ö en lich zugänglich
machen, e eiben ode ande wei ig nu zen.
So e n die Ve asse die Dokumen e un e Open-Con en -Lizenzen
(insbesonde e CC-Lizenzen) zu Ve ügung ges ell haben soll en,
gel en abweichend on diesen Nu zungsbedingungen die in de do
genann en Lizenz gewäh en Nu zungs ech e.
Te ms o use:
Documen s in EconS o may be sa ed and copied o you pe sonal
and schola ly pu poses.
You a e no o copy documen s o public o comme cial pu poses, o
exhibi he documen s publicly, o make hem publicly a ailable on he
in e ne , o o dis ibu e o o he wise use he documen s in public.
I he documen s ha e been made a ailable unde an Open Con en
Licence (especially C ea i e Commons Licences), you may exe cise
u he usage igh s as speci ied in he indica ed licence.
h ps://c ea i ecommons.o g/licenses/by/3.0/igo/
ADB I
NSTITUTE
W
ORKING
P
APER 9
The Ins i u ional Founda ions
o Ma ke T ansi ion in he
People’s Republic o China
Yingyi Qian
May 2000
ADB INSTITUTE
TOKYO
ASIAN DEVELOPMENT BANK INSTITUTE
ASIAN DEVELOPMENT BANK INSTITUTE
Fo he pas 20 yea s, PRC’s economic e o m has succeeded
wi hou comple e ma ke libe aliza ion, wi hou p i a iza ion
and secu e p ope y igh s, and wi hou democ acy. This
sugges s ha conside able g ow h is possible wi h sensible
bu no pe ec ins i u ions, app op ia e bu no op imal
sequencing and ha some “ ansi ional ins i u ions” can be
mo e e ec i e han “bes p ac ice ins i u ions” a ce ain
s ages due o second-bes p inciples.
This impo an pape challenges con en ional wisdom by
showing ha (i) planned economies can be e o med, (ii) such
e o m can be deepened in o a ull-scale ansi ion o ma ke ,
and (iii) he e may be di e se pa hs o success ul ansi ion
depending on coun y-speci ic condi ions. The au ho
imp o es ou unde s anding o PRC’s unique e o m pa h and
ad ances a new pa adigm o he s udy o e o m and
ins i u ional changes in gene al.
ADB Ins i u e
Wo king Pape Se ies
No. 9
May 2000
The Ins i u ional Founda ions
o Ma ke T ansi ion in
he People’s Republic o China
Yingyi Qian
II
ADB INSTITUTE WORKING PAPER 9
This pape is pa o he Ins i u e’s esea ch p ojec on de elopmen pa adigms. Addi ional copies o he pape a e
a ailable ee om he Asian De elopmen Bank Ins i u e, 8 h Floo , Kasumigaseki Building, 3-2-5 Kasumigaseki,
Chiyoda-ku, Tokyo 100-6008, Japan. A en ion: Publica ions.
The Wo king Pape Se ies p ima ily dissemina es selec ed wo k in p og ess o acili a e an exchange o
ideas wi hin he Ins i u e's cons i uencies and he wide academic and policy communi ies. An objec i e o
he se ies is o ci cula e p ima y indings p omp ly, ega dless o he deg ee o inish. The indings,
in e p e a ions, and conclusions a e he au ho 's own and a e no necessa ily endo sed by he Asian
De elopmen Bank Ins i u e. They should no be a ibu ed o he Asian De elopmen Bank, i s Boa ds, o
any o i s membe coun ies. They a e published unde he esponsibili y o he Dean o he Asian
De elopmen Bank Ins i u e. The Ins i u e does no gua an ee he accu acy o easonableness o he
con en s he ein and accep s no esponsibili y wha soe e o any consequences o i s use. The e m
"coun y", as used in he con ex o he ADB, e e s o a membe o he ADB and does no imply any iew on
he pa o he Ins i u e as o so e eign y o independen s a us. Names o coun ies o economies men ioned
in his se ies a e chosen by he au ho s, in he exe cise o hei academic eedom, and he Ins i u e is in no
way esponsible o such usage.
Copy igh ©2000 Asian De elopmen Bank Ins i u e and he au ho . All igh s ese ed.
P oduced by ADBI Publishing.
ABOUT THE AUTHOR
Yingyi Qian is cu en ly P o esso o Economics a Ma yland Uni e si y. His esea ch in e es s
include compa a i e ins i u ional economics, economies in ansi ion, and PRC’s economic
e o m and de elopmen . A e ecei ing his doc o a e om Ha a d, P o esso Qian augh
economics a S an o d Uni e si y h oughou he 1990s. Du ing his pe iod he has been awa ded
a numbe o ellowships, including he McNama a Facul y and Hoo e Na ional. He also holds a
Mas e s in s a is ics om Columbia and a M. Phil. in managemen science and ope a ions
esea ch om he Yale School o Managemen .
III
PREFACE
The ADB Ins i u e aims o explo e he mos app op ia e de elopmen pa adigms o Asia
composed o well-balanced combina ions o he oles o ma ke s, ins i u ions, and go e nmen s in he
pos -c isis pe iod.
Unde his b oad esea ch p ojec on de elopmen pa adigms, he ADB Ins i u e Wo king
Pape Se ies will con ibu e o dissemina ing wo ks-in-p og ess as a building block o he p ojec and
will in i e commen s and ques ions.
I us ha his se ies will p o oke cons uc i e discussions among policymake s as well as
esea che s abou whe e Asian economies should go om he las c isis and cu en eco e y.
The con e ence e sion o his pape was p esen ed on 10 Decembe 1999 a he High-
Le el Dialogue on De elopmen Pa adigms, on he occasion o he second anni e sa y o he
es ablishmen o he ADB Ins i u e.
Masa u Yoshi omi
Dean
ADB Ins i u e

IV
ABSTRACT
This pape in ends o p ope ly accoun o he People’s Republic o China’s (PRC) wo decades o
ma ke ansi ion by examining i s ins i u ional ounda ions. The jou ney o ansi ion is analyzed as a
wo-s age p ocess. In he i s s age (1978-93), he sys em was e o med o unleash he s anda d ma ke
o ces o incen i es, ha d budge cons ain s, and compe i ion. The unde lying ins i u ional o ms and
mechanisms, howe e , we e a om con en ional: e o ming go e nmen h ough egional
decen aliza ion; en y and expansion o non-s a e (mos ly local go e nmen ) en e p ises; inancial
s abili y h ough “ inancial dualism;” and a dual- ack app oach o ma ke libe aliza ion.
In he second s age, PRC aimed o build a ule-based ma ke sys em inco po a ing in e na ional
bes p ac ice ins i u ions bu p oceeded in i s own way. Majo p og ess was made in he i s i e yea s
(1994-98) on he uni ica ion o exchange a es and con e ibili y o he cu en accoun ; he o e haul o
he ax and iscal sys ems; eo ganiza ion o he cen al bank; downsizing o he go e nmen
bu eauc acy; and p i a iza ion and es uc u ing o s a e-owned en e p ises (SOEs). To comple e i s
ansi ion o ma ke s, howe e , PRC s ill aces se ious challenges, especially in ans o ming i s
inancial sys em and SOEs and in es ablishing he ule o law.
The pape concludes by e lec ing on he economics o e o m and ins i u ional change om he
PRC expe ience. The main lesson lea ned is ha conside able g ow h is possible wi h sensible bu no
pe ec ins i u ions, and ha some uncon en ional “ ansi ional ins i u ions” can be mo e e ec i e han
he “bes p ac ice” ins i u ions o a pe iod o ime because o he second-bes p inciple. Speci ic lessons
include: incen i es, ha d budge cons ain s, and compe i ion should apply no only o i ms bu also o
go e nmen s; e o ms can be implemen ed wi hou c ea ing many o big lose s; and success ul e o ms
equi e app op ia e, bu no necessa ily op imal, sequencing.
V
TABLE OF CONTENTS
P e ace iii
Abs ac i
Table o Con en s
1. In e p e ing PRC’s T ansi ion o Ma ke s: The Ins i u ional Pe spec i e 1
2. Re o ming he Sys em: 1979-93 5
A. Regional Decen aliza ion o he Go e nmen 5
B. En y and Expansion o Non-S a e (Mos ly Local Go e nmen ) Fi ms 7
C. “Financial Dualism”10
D. Ma ke Libe aliza ion h ough he Dual-T ack App oach 12
E. An Assessmen 13
3. Replacing he Sys em: Since 1994 14
A. S a egic Mo e: Se ing he Goal o a Ma ke Sys em 15
B. Majo Accomplishmen s in he Fi s Fi e Yea s (1994-98) 18
C. Poli ical Economy o Re o m and he Dynamics o T ansi ion 21
4. Comple ing PRC’s T ansi ion: Challenges Ahead and P io i y Resea ch Agenda 23
A. The Financial Sys em 23
B. S a e-Owned En e p ises and Co po a e Go e nance 25
C. The Rule o Law 27
5. Re lec ions on he Economics o Re o m and Ins i u ional Change:
Lessons om PRC 29
A. Re lec ions on he P inciples o Sys em Change 29
B. Re lec ions on he P ocess o Re o m 30
C. Re lec ions on he Theo y and P ac ice o T ansi ion 31
Re e ences 33
1
The Ins i u ional Founda ions o Ma ke T ansi ion
in he People’s Republic o China
Yingyi Qian†
1. In e p e ing PRC’s T ansi ion o Ma ke s: The Ins i u ional Pe spec i e
In he wo decades be ween 1978 and 1998, he People’s Republic o China (PRC) has
ans o med i sel om a cen ally planned economy o an eme ging ma ke economy and a
he same ime has achie ed nea ly a 10 pe cen a e age g ow h a e. Du ing his pe iod, PRC’s
pe capi a GDP has mo e han quad upled and he li ing s anda d o o dina y Chinese people
has imp o ed signi ican ly. Fo ins ance, pe capi a consump ion has inc eased ou imes o
eggs and eigh imes o poul y, he pe pe son li ing space has mo e han doubled in he u ban
a eas and nea ly ipled in he u al a eas, and o al household bank deposi s, measu ed agains
he GDP, inc eased om less han 6 pe cen in 1978 o mo e han 60 pe cen in 1998.
The bene i s o he e o m we e also sha ed by people on a b oad basis. The numbe o
people li ing in absolu e po e y has been subs an ially educed om o e 250 million o abou
50 million in wo decades, a decline om one- hi d o a wen y- i h o PRC’s popula ion. Li e
expec ancy on he o he hand has inc eased om 64.37 yea s in he 1970s o 70.80 yea s in
1996 (68.71 o men and 73.04 o women), wi h in an mo ali y alling om o e 50 pe
housand in he 1970s o less han 30 pe housand in he 1990s (China S a is ical Yea book,
1997; Almanac o China’s Popula ion, 1997). In 1998, he Wo ld Bank mo ed PRC’s anking
up om a low-income o a lowe -middle-income coun y.1
Such a pe o mance appea s mo e imp essi e when compa ed wi h he a e age
pe o mance o he ansi ion economies in Eas e n Eu ope and he o me So ie Union. By
1998, wi h only a ew excep ions, he g ea majo i y o hese coun ies s ill ha e no eco e ed
o hei 1989 ou pu le els acco ding o o icial s a is ics. The PRC pe o mance looks e en
mo e imp essi e when conside ing he ac ha ans o ming la ge coun ies is much mo e
complica ed han ans o ming smalle ones; concei ably, he asks o ans o ming Russia o
PRC a e mo e challenging han hose o ans o ming Poland o Vie Nam. A he ou se ,
PRC’s e o m wen agains all odds. Coming ou o he disas ous decade o he Cul u al
Re olu ion, i was poo , o e -popula ed, lacked human capi al and na u al esou ces, and was
cons ained by ad e se ideology and poli ical opposi ion. Two decades ago ew economis s
would ha e be on oday’s ou come o e o m in PRC.
† I app ecia e he commen s o Masahiko Aoki, Nicholas Hope, Law ence Lau, Bo is Plesko ic, Gé a d Roland,
And ew Walde , and h ee e iewe s. O iginal pape i s p esen ed a he Wo ld Bank’s Annual Con e ence on
De elopmen Economics, Washing on, D.C., 28-30 Ap il 1999. The con e ence e sion o his pape was p epa ed
o he ADB Ins i u e’s High-Le el Dialogue on De elopmen Pa adigms, Tokyo, 10 Decembe 1999. The indings,
in e p e a ions and conclusions exp essed in his pape a e en i ely hose o he au ho .
1 Cen ally planned economies also had a high g ow h pe iod (such as he So ie Union in he 1930s and 1950s,
Eas e n Eu opean coun ies in he 1950s and 1960s, and PRC in he 1950s). Howe e , i is well known ha such
a g ow h a e was based on hea y indus y expansion a he sac i ice o consume indus y and hus he people’s
li ing s anda ds, and i was always associa ed wi h ch onic sho ages (Ko nai, 1980). PRC’s high g ow h in he
pas wo decades was di e en : consume and expo indus ies boomed, he people’s li ing s anda ds imp o ed,
and ch onic sho ages disappea ed.
2
E en so, PRC’s e o m expe ience has been always iewed as an anomaly in e ms o
ansi ion o a ma ke economy, and i has no been p ope ly accoun ed o by mains eam
economics and hus app ecia ed by mains eam economis s. Fo example, F om Plan o
Ma ke : Wo ld De elopmen Repo 1996 on ansi ion economies (Wo ld Bank, 1996) ga e
PRC sho sh i because i could no igu e ou whe e o pu PRC on he a ious measu emen
pa ame e s, and ins ead illus a ed he PRC expe ience mainly in boxes a he han in he ex .
PRC simply does no i he gene al desc ip ion o he epo . Howe e , he da a poin o PRC
is oo impo an o igno e: i has been one o he mos success ul ansi ion economies, i
p oduced mo e han all o he ansi ion economies combined in 1998 in e ms o GDP, and,
mo eo e , i s pe capi a GDP is e y likely o su pass ha o he 15 o me So ie Union
coun ies in he nex decade, which was un hinkable a decade ago.2
S ill, economis s end o unde es ima e he signi icance o PRC’s e o m expe ience. The
mos popula a gumen is ha PRC was a poo ag icul u al coun y and hus e o m was easy.
O cou se PRC was much less de eloped han Eas e n Eu ope and he o me So ie Union a
he ou se o e o m and he la e aced some di icul ies ha PRC did no ha e, such as
p oblems o excess indus ial capaci y and comp ehensi e wel a e co e age. Howe e , his
a gumen does no explain how and why PRC’s e o m was success ul, especially conside ing
ha i aced double di icul ies. As a planned economy, PRC aced many p oblems simila ,
al hough no iden ical, o Eas e n Eu ope and he o me So ie Union, such as a lack o
p ope y igh s and ma ke s, pe sis ence o a p eda o y go e nmen , and he di icul y o
main aining inancial s abili y. In addi ion, as an unde de eloped coun y, PRC also aced
many p oblems ha do no exis in Eas e n Eu ope and he o me So ie Union, such as
eno mous popula ion p essu e, se e e sho ages o human capi al and na u al esou ces, e y
poo indus ial and in as uc u e bases, and a lack o democ acy. The ac ha PRC aced he
double p oblem o ansi ion and de elopmen p esen ed a bigge challenge and i is a om
clea how PRC managed o succeed. A e all, he e a e many poo coun ies ha do no g ow.
The easons why PRC’s e o ms a e no p ope ly unde s ood and hus app ecia ed by
mains eam economis s a e p o ound. The e a e s ong p io belie s, based on he exis ing
knowledge o economics, abou he o mula ion ha a ansi ion should use. Fu he mo e, such
belie s a e suppo ed by he s ong e idence om he ailed economic e o m in Eas e n Eu ope
and he o me So ie Union p io o 1990 which did no ollow ha o mula ion. The heo y
and e idence oge he o med a powe ul “con en ional wisdom” abou a se o necessa y and
su icien condi ions o a success ul ansi ion; ha is, s abiliza ion, libe aliza ion, p i a iza ion,
and democ a iza ion. Lea ing aside he issue o whe he hey a e su icien o he expe s on
Eas e n Eu ope and he o me So ie Union, he PRC pa h o e o m and i s associa ed apid
g ow h seemed o de y he necessa y pa o con en ional wisdom: al hough PRC has adop ed
many o he policies ad oca ed by economis s, such as being open o ade and o eign
in es men , and en o cing mac oeconomic s abili y, iola ions o he s anda d policy
p esc ip ions a e also s iking. Fo mos o he pas wo decades, PRC’s e o m succeeded
wi hou comple e ma ke libe aliza ion, wi hou p i a iza ion and secu e p i a e p ope y igh s,
and wi hou democ acy. One migh ha e hough ha in he absence o hese “essen ial” ac o s
e o m would ail.
2 Acco ding o Maddison’s (1998) calcula ion based on pu chasing powe pa i y, wi hou aking in o accoun
he 1998 Russian economic c isis, PRC’s pe capi a GDP will su pass ha o he 15 o me So ie Union
coun ies by 2010.
9
communi y go e nmen can make use o i s poli ical connec ions wi h he s a e banks o
channel loans o TVEs, and he s a e banks a e also mo e willing o lend o TVEs because
disc imina ion agains p i a e en e p ises makes lending o he la e poli ically mo e isky. On
economic g ounds, he communi y go e nmen is able o educe he isks bo ne by he banks
h ough c oss-subsidiza ion among i s many di e si ied en e p ises (By d, 1990), o i can use
accumula ed collec i e asse s as colla e al o as co-in es men unds o educe po en ial haza ds
in he lending-bo owing ela ionship (Che and Qian, 1998a). Also, he communi y go e nmen
can educe in o ma ion asymme y in ol ed in ma ke ansac ions by in eg a ing a numbe o
in es men s, since ma ke obse a ions d awn om hese ansac ions a e much mo e
in o ma i e han hey a e when d awn om ansac ions esul ing om uno ganized p i a e
in es men s (Che, 1998).
Insecu e p ope y igh s and impe ec capi al ma ke s a e also he common ea u es o
o he ansi ion and de eloping economies, bu why do we no see TVEs o simila ypes o
i ms elsewhe e mo e o en? The egional decen aliza ion in PRC desc ibed abo e appea s o
ha e played a cen al ole, because local go e nmen s a ownship and illage le els a e
empowe ed wi h comp ehensi e au ho i y o local economic de elopmen , and hey a e also
p o ided wi h he incen i es o do so since hey can keep he e enues gene a ed (By d and
Gelb, 1990; Oi, 1992). Elsewhe e, he asks o go e nmen bu eauc a s a e simply collec ing
axes and passing hem up o he highe le el go e nmen s. Bu a deepe ques ion is: why
canno he local go e nmen ge mo e e enue by axing p i a e i ms han by de eloping hei
own i ms? Che and Qian (1998b), using he incomple e con ac ing amewo k, a gue ha
owne ship igh s gi e he go e nmen con ol o e he i ms’ inancial accoun s and hus make
i less cos ly o ex ac e enues om hem han axing p i a e i ms. Fo he same eason,
when local go e nmen s con ol i ms, i is also ha de o he cen al go e nmen o ex ac
e enue om hem, and hus e enue is mo e likely o s ay in he local a eas. The e o e,
owne ship and con ol make he di e ence.
Econome ic s udies on he da a om PRC’s u al indus y p o ides some e idence o
suppo he abo e heo e ical a gumen s. Using panel da a om 28 p o inces in PRC be ween
1986 and 1993, Jin and Qian (1998) ound ha he sha e o TVEs ela i e o p i a e
en e p ises in u al indus y in a p o ince is highe i he ini ial collec i e asse s unde he
con ol o communi y go e nmen is la ge , o i he local poli ical s eng h o esis p essu e
om highe le el go e nmen (app op ia ely measu ed) is highe . This e idence is consis en
wi h he heo y ha local go e nmen owne ship o i ms is ela ed o he ins i u ional
en i onmen . They also examined he consequence o owne ship o i ms on he e enue
dis ibu ion among he na ional go e nmen , communi y go e nmen , and households. They
ound ha a one pe cen inc ease in he sha e o TVEs ela i e o p i a e en e p ises is
associa ed wi h a 0.11 pe cen inc ease in he sha es o e enue acc ued o he na ional
go e nmen and a 0.24 pe cen inc ease in he sha es o e enue acc ued o he
ownship/ illage go e nmen s. These esul s con i m he iscal incen i es o he local
go e nmen s in de eloping TVEs, and also show ha go e nmen con ol o i ms plays an
impo an ole o subs i u ing o axa ion ins i u ions.
C. “Financial Dualism”
PRC has gene ally managed mac oeconomic s abili y well, excep o he pe iods o 1988-89
and 1993-94. PRC, no less han Russia, expe ienced a sha p go e nmen ax e enue decline.
Wha , hen, a e he mic oeconomic and ins i u ional ounda ions o i s inancial s abili y?
The hi d ins i u ional pilla o e o m is “ inancial dualism” (McKinnon, 1993; Bai e al.,
1999).

10
The e a e wo aspec s o inancial dualism. One aspec conce ns go e nmen e enue:
al hough ax e enue sha ply declined, i was accompanied, and hus pa ially compensa ed, by
an inc ease o “quasi- iscal” e enue om imp essi e inancial deepening. This p o ides a
basis o PRC’s mac oeconomic s abili y and a oids a inancial c isis like Russia’s (McKinnon,
1993). In PRC, consolida ed go e nmen budge a y e enue as a sha e o GDP declined om
31 pe cen in 1978 o 13 pe cen in 1993. Taking in o accoun he ex a-budge a y and o -
budge e enues, o al ax e enue also declined d ama ically, om 40 pe cen o GDP in 1978
o abou 19 pe cen in 1993 (Bai e al., 1999). On he o he hand, cash in ci cula ion as a
pe cen age o GDP was less han 6 pe cen in 1978, and inc eased o 16 pe cen in 1993. To al
household bank deposi s we e less han 6 pe cen o he GDP in 1978, and hey inc eased o
abou 50 pe cen in 1993 and u he o 62 pe cen in 1997. The M2 o GDP a io con inued o
climb, om less han 50 pe cen be o e he e o m o mo e han 100 pe cen in he 1990s
(Almanac o China’s Finance and Banking, 1996). Wha seems o be su p ising is ha he
inancial buildup las ed much longe han mos economis s expec ed. The esul has been ha
he go e nmen eaped subs an ial bene i s. By one es ima ion, be ween 1986 and 1994, he
go e nmen collec ed quasi- iscal e enue om he banking sec o , a e aging abou 9 pe cen
o he GDP, o mo e han one hal o he budge a y e enue (Bai e al., 1999).
Bai e al. (1999) go one s ep beyond he mac oeconomic issue o go e nmen e enue
and s udy mo e undamen al mic oeconomic implica ions o iscal-decline-cum- inancial-
deepening. They highligh he coexis ence o wo ins i u ional a angemen s in PRC’s inancial
sys em. The i s is he well-known one o inancial ep ession, ha is, a combina ion o
go e nmen con ol on in e na ional capi al lows wi h es ic ion on domes ic in e es a es and
p i a e inancial ac i i ies. The second is wha hey called “anonymous banking,” which is a
combina ion o he go e nmen ’s elaxed egula ion on he use o cash o ansac ions and
pe mission o use anonymous household sa ings deposi s. They a gue ha anonymous banking
oge he wi h inancial ep ession implies some majo ad an ages o e di ec axa ion in he
ins i u ional en i onmen o PRC.
The con en ional wisdom holds ha axa ion is less dis o iona y han e enue ex ac ion
om inancial ep ession. Bu his iew igno es he go e nmen beha io on axa ion. In PRC,
as in many de eloping and ansi ion economies, because he e is a lack o ule o law, he
go e nmen has di icul ies commi ing i sel o ixed ax a es, le alone o low ax a es. Such
a commi men p oblem unde mines p i a e incen i es and is o en ega ded as a majo obs acle
o economic de elopmen as well as e o m (No h, 1997; Williamson, 1994). Bai e al. (1999)
a gue ha anonymous banking p o ides a simple and e ec i e commi men de ice o limi he
go e nmen ’s p eda o y beha io and c ea e p i a e incen i es. When ansac ions a e made
h ough cash a he han bank ans e s, i is di icul o he go e nmen o moni o business
ansac ions and hus o ax away he gene a ed e enue. When bank deposi s a e anonymous,
he go e nmen does no know he iden i y o deposi o s and hus is unable o a ge a pa icula
pe son and con isca e his inancial weal h. Al hough he go e nmen can s ill “ ax” inancial
sa ings h ough in la ion o egula ing he in e es a e, his me hod o e enue ex ac ion is
indisc imina e. In hei model, Bai e al. show ha e enue ex ac ion om he inancial sys em
can be cos ly o he go e nmen and hus i imposes a limi . The e o e, h ough anonymous
banking, he go e nmen is able o achie e a c edible commi men o c ea ing p i a e
incen i es. Th ough inancial ep ession, al hough he go e nmen can acqui e some quasi-
iscal e enue, i is mo e limi ed han disc e iona y axa ion.5 They conclude ha indi ec
5 As a esul , inancial ep ession in PRC was “mild” because he eal in e es a e was no oo nega i e. Indeed,
du ing he pas wo decades, in la ion was gene ally below 10 pe cen . In wo pe iods o 1988-89 and 1993-94
11
e enue ex ac ion h ough he banking sys em has no jus p e en ed e enue collapses, bu
mo e impo an ly, i has bound he go e nmen ’s hands and limi ed i s abili y o e enue
ex ac ion and hus is conduci e o p i a e incen i es.
The o he aspec o inancial dualism conce ns he lending side o he inancial sys em.
The e was an asymme y: s a e en e p ises ecei ed he mos c edi om he s a e banks and
aced so budge cons ain s, while non-s a e en e p ises ecei ed only limi ed c edi and aced
much ha de budge cons ain s. Fo example, he o al size o he SOE indus ial ou pu was
abou wice ha o TVEs in he la e 1980s and ea ly 1990s. Howe e , loans o u al en e p ises
(mainly TVEs) accoun ed o only abou 8 pe cen o all non-ag icul u al loans, while loans o
SOEs accoun ed o abou 86 pe cen (Almanac o China’s Finance and Banking, 1993). While
c edi disc imina ion agains non-s a e i ms was o en c i icized, such disc imina ion became a
blessing because when he non-s a e sec o was subjec o a ha d budge cons ain s, i was
mo e disciplined and pe o med be e .
An in iguing ques ion is why he s a e sec o is subjec o so budge cons ain s and he
non-s a e sec o is subjec o ha d budge cons ain s. One main eason conce ns he poli ical
bene i s o he go e nmen . His o ically, he go e nmen has been commi ed o he wel a e o
wo ke s in he s a e sec o in exchange o hei poli ical suppo , bu no o hose in he non-
s a e sec o . The e o e, he go e nmen will bail ou s a e en e p ises when pe o ming badly
since hey alue wo ke s’ employmen . Ano he main eason conce ns he go e nmen ’s
inancial objec i e. In Bai e al. (1999) he go e nmen equi es s a e i ms o conduc
ansac ions h ough s a e banks so i can con enien ly obse e hem and ax hem acco dingly.
In compa ison, non-s a e i ms o en use cash o ansac ions and he go e nmen inds i ha d
o moni o and ax hem. Gi en his di e ence, he go e nmen would p e e o p o ide c edi s
o s a e i ms a he han o p i a e i ms because o i s own e enue conce ns. The downside o
p o iding c edi o s a e i ms is hei p oduc i i y, which implies a lowe po en ial ax e enue.
In Che and Qian (1998a), he di e ence comes om he con ol s uc u e. In he case o an
SOE, he go e nmen con ols bo h he en e p ise and he s a e bank. I he e o e ecei es all
he bene i s om e inancing a p ojec a e sunk in es men s. This is ex pos e icien , bu
makes less c edible an ex an e commi men o no u he inancing. In he case o TVEs, he
communi y go e nmen con ols only he TVEs bu no s a e banks. This sepa a ion o con ol
o e i ms and unds induces in o ma ion asymme y be ween he c edi o ( he s a e bank) and
bo owe ( he communi y go e nmen on behal o TVEs). Following he logic o Dewa ipon
and Maskin (1995), his in o ma ion asymme y educes he bene i s he s a e bank ecei es
om e inancing a p ojec a e sunk in es men . This in u n inc eases he ex an e c edibili y o
no e inancing. The e o e, limi ed au ho i y o a communi y go e nmen becomes a good hing
and i se es as a commi men de ice o ha den budge cons ain s.
D. Ma ke Libe aliza ion h ough he Dual-T ack App oach
By he mid-1990s, p ices o mos p oduc s in PRC we e comple ely libe alized. Bu he way
he Chinese achie ed his esul is qui e di e en om he way i was done by he Eas e n
Eu opeans. In Hunga y, o example, a e he 1968 e o m, al hough all manda o y planning
was abolished, p ices we e s ill de e mined adminis a i ely by bu eauc a s subjec o
poli ical ba gaining. The e was no eal ma ke , only a simula ed ma ke (Ko nai, 1986). A e
1990, on he o he hand, p ices we e swi ly libe alized in one s oke. PRC ook a “dual-
when in la ion ose o mo e han 20 pe cen pe yea , he go e nmen quickly indexed ime deposi s (o e a
h ee-yea ma u i y) o ensu e a non-nega i e eal in e es a e.
12
ack” app oach o p ice libe aliza ion unde which he ma ke was i s libe alized a he
ma gin while (in ama ginal) planned p ices and quo as we e main ained and hen phased ou
la e . Unde he plan ack, economic agen s we e assigned igh s o and obliga ions o ixed
quan i ies o goods a ixed plan p ices as speci ied in he p e-exis ing plan. In addi ion, a
ma ke ack was in oduced unde which economic agen s pa icipa ed in he ma ke a ee
ma ke p ices, p o ided ha hey ul illed hei obliga ions unde he p e-exis ing plan. Wi h
his app oach, eal ma ke p ices and ma ke s as a esou ce alloca ion ins i u ion we e c ea ed
in PRC in he e y ea ly s ages o e o m, which clea ly di e ed om he Eas e n Eu opean
expe ience p io o 1990.
Again he dual- ack app oach was o en c i icized by he economis s who iew i as a
pa ial e o m lacking he comple eness o libe aliza ion. Bu Lau, Qian, and Roland (1997,
1999) a gue ha such a pe cep ion is no co ec , and ha he dual- ack app oach o ma ke
libe aliza ion in ac has wo ad an ages: i can libe alize ma ke s wi hou c ea ing lose s and
hus is poli ically appealing, and i can also achie e e iciency unde ce ain condi ions. They
show, in bo h pa ial and gene al equilib ium models, ha independen o he ini ial condi ions
conce ning supply and demand (such as whe he planned p ices o quan i ies a e abo e o
below he ma ke equilib ium), as long as he p e-exis ing easible plan con inues o be
en o ced app op ia ely, he dual- ack app oach o ma ke libe aliza ion is always Pa e o-
imp o ing. In addi ion, i also achie es e iciency unde usual condi ions such as p o i
maximiza ion and pe ec compe i ion, p o ided ma ke esales and ma ke pu chases o
edeli e y a e all allowed.
E iciency-enhancing economic e o m should po en ially allow winne s o compensa e
lose s, he eby making he e o m, Pa e o-imp o ing. Howe e , in p ac ice, i seems e y
di icul o ind mechanisms ha make economic e o m Pa e o-imp o ing, and e en mo e
di icul o e o m o be simul aneously Pa e o-imp o ing and e icien , because o he
dis o iona y cos s o compensa ion o a lack o c edibili y in i s implemen a ion. The dual-
ack app oach p o ides one mechanism o he implemen a ion o e icien Pa e o-imp o ing
e o m. The in oduc ion o he ma ke ack p o ides he oppo uni y o economic agen s who
pa icipa e in i o be be e o , whe eas he main enance o he plan ack p o ides implici
ans e s o compensa e po en ial lose s om ma ke libe aliza ion by p o ec ing he s a us quo
en s unde he p e-exis ing plan. Thus, he dual- ack app oach is, by design, Pa e o-imp o ing.
Mo eo e , as he compensa o y ans e s a e in ama ginal and hus lump sum in na u e, he
dual- ack app oach can be e icien oo. While single- ack (o “big-bang”) ma ke
libe aliza ion will lead o e iciency unde he usual condi ions (such as p o i maximiza ion
and pe ec compe i ion), Pa e o-imp o emen canno be assu ed. Fu he mo e, wi h he
p esence o some ma ke impe ec ion (such as sea ch ic ions o impe ec compe i ion),
single- ack libe aliza ion may cause a decline in agg ega e ou pu , as shown in he models o
Blancha d and K eme (1997), Roland and Ve die (1999), and Li (1999), bu he dual- ack
app oach can a oid his.
I is clea ha en o cemen o he plan ack is c ucial o p ese ing he p e-exis ing en s.
Howe e , su icien s a e en o cemen powe is needed no o implemen an unpopula e o m,
bu o ca y ou one ha c ea es no lose s, only winne s. One desi able ea u e o he dual- ack
app oach is i s minimal, addi ional in o ma ional and ins i u ional equi emen s: i u ilizes he
exis ing in o ma ion con ained in he o iginal plan and en o ces he plan h ough exis ing
planning ins i u ions. No new in o ma ion and no new ins i u ions a e necessa y.
Ag icul u al ma ke libe aliza ion illus a es how he dual- ack app oach can be bo h
Pa e o-imp o ing and e icien . The commune (and la e he households) was assigned he
obliga ion o sell a ixed quan i y o ou pu o he s a e p ocu emen agency as p e iously
13
manda ed unde he plan a p ede e mined plan p ices and o pay a ixed ax o he go e nmen .
I also had he igh o ecei e a ixed quan i y o inpu s, p incipally chemical e ilize s, om
s a e-owned supplie s a p ede e mined plan p ices. Subjec o ul illing hese condi ions, he
commune was ee o p oduce and sell wha e e i conside ed p o i able, and e ain any p o i .
Mo eo e , he commune and households could pu chase g ain (o o he ) ou pu s om he
ma ke o esale o he s a e o ul ill i s esponsibili y. Unde he dual- ack, he s a e
p ocu emen o domes ically p oduced g ains be ween 1978 and 1988 emained essen ially
ixed, while he e was almos a one- hi d inc ease in g ain ou pu .
Indus ial libe aliza ion also shows how ma ke s could g ow ou o plans (By d, 1991;
Naugh on, 1995). Fo coal, PRC’s p incipal ene gy sou ce, he planned deli e y was inc eased
somewha om 329 million ons in 1981 o 427 million ons in 1989 (mainly because new s a e
coal mines we e opened), bu he ma ke ack inc eased d ama ically om 293 million ons o
628 million ons in he same pe iod. The inc emen s came mainly om small u al coal mines
un by indi iduals and TVEs. Fo s eel, ano he o PRC’s majo indus ial ma e ial, he plan
ack in absolu e e ms was qui e s able, bu he sha e o plan alloca ion ell om 52 pe cen in
1981 o 30 pe cen in 1990. Unlike coal, he supply esponse in s eel came mainly om la ge
SOEs a he han small non-s a e i ms. In he cases o bo h coal and s eel, because he plan
ack was basically “ ozen,” he economy was able o g ow ou o he plan on he basis o he
ma ke ack expansion by s a e o non-s a e i ms.
E. An Assessmen
PRC’s i s i een yea s o e o m was a ema kable success, judging by he esul s o as
GDP g ow h, elimina ion o sho ages, and d ama ic imp o emen in he li ing s anda ds o
o dina y people. By he end o 1993, e o m was suppo ed by people in all walks o li e
simply because e e yone bene i ed om i . This was in sha p con as wi h he us a ion o
Eas e n Eu opean e o me s in he la e 1980s, when hey saw only a dead end o hei e o m
e o s o decades.
Cen al planning has long been c i icized as an ine icien esou ce alloca ion mechanism
compa ed wi h he ma ke , bu i s ailu e has deepe ins i u ional p oblems han simply
planning mis akes o ushed de elopmen s a egies. Fundamen ally, he sys em is unable, in a
c edible way, o p o ide posi i e incen i es o economic agen s and o impose inancial
discipline upon hem, and i also ails o c ea e genuine compe i ion. This had emained he
case in Eas e n Eu ope e en du ing he economic e o m in he 1970s and 1980s (Ko nai,
1986).
PRC was able o a oid he a e o Eas e n Eu opean e o m p io o 1990 because o i s
deepe ins i u ional changes. Each o he ou pilla s o ins i u ional change analyzed abo e
con ibu ed o PRC’s e o m success since hey changed he unc ioning o he go e nmen ,
i ms, he inancial sys em, and ma ke s o unleash he o ces o posi i e incen i es, ha d
budge cons ain s, and compe i ion, bu in no el ways. Fo example, posi i e incen i es we e
p o ided h ough a a ie y o means, such as iscal con ac ing unde egional decen aliza ion,
local go e nmen owne ship, and in o ma ion decen aliza ion h ough he elaxa ion o he
egula ion on cash and pe mission o anonymous bank accoun s. Ha de budge cons ain s
we e imposed in ci cums ances whe e he e a e cons ain s imposed on he go e nmen , o
example h ough con ol igh s and in o ma ion s uc u es. Finally, compe i ion was c ea ed no
only among i ms (bo h s a e and non-s a e) bu also among local go e nmen s a p o incial,
ci y, coun y, ownship, and illage le els. The PRC expe ience has demons a ed ha e o ming
he go e nmen and p o iding i wi h incen i es is as c ucial as e o ming he economy; non-
14
p i a e and non-s a e owne ship can be an engine o g ow h; inancial s abili y can be ob ained
o an ex ended pe iod h ough quasi- iscal e enues om he banking sys em; and dual ack
libe aliza ion p o ides one mechanism o minimize he numbe o lose s om e o m.
In e es ingly, none o hese me hods we e ecommended by economis s; o he con a y, all o
hem we e c i icized.
Despi e hese imp essi e achie emen s, he e we e many se ious p oblems du ing his
s age o e o m. Fi s , some mis akes we e made. Fo example, decen aliza ion o go e nmen
wen oo a in he a eas o mone a y policy: local go e nmen s gained subs an ial con ol o e
he c edi supply, and a one poin , he bonuses o he cen al bank’s local b anches we e e en
linked o he amoun o c edi hey ex ended. This was a sou ce o so budge cons ain s o
local go e nmen s and SOEs and o in la ion as well (Qian and Roland, 1998).
Second, many di icul e o ms we e delayed. Fo example, no single s a e en e p ise was
p i a ized and almos none wen bank up . This is in u n because o he delay o es ablishing a
social sa e y ne . The e was also no e o made in es ablishing p ope y igh s p o ec ed by he
ule o law and con ac en o cemen mechanisms. Al hough es ablishing a ule-based ax
sys em was a emp ed, i was delayed and ailed (Shi k, 1993).
Thi d, and mos impo an , he achie emen s up o 1993 we e made basically h ough
ins i u ional inno a ions which we e ei he ad hoc esponses o pa icula cons ain s in he
planning sys em o ook ad an age o he loopholes in i . The a ie ies o con ac ing p ac ices
be ween di e en le els o go e nmen and be ween go e nmen and en e p ises we e good
examples. Al hough hey we e e ec i e in b eaking he cen al command, hey we e ad hoc and
subjec o equen enego ia ion and change. In hei empi ical in es iga ion, Jin, Qian, and
Weingas (1999) ound e idence o he “ a che e ec ” o e ime unde he iscal con ac ing
sys em: du ing 1982 and 1992 and on a e age, a 1 yuan inc ease o e enue collec ion his yea
leads o a 0.24 yuan inc ease o e enue emi ance (o dec ease o subsidies) in he nex yea .
They also ound ha a majo i y o p o inces ecei ed ex a, hough limi ed, subsidies beyond
con ac ual s ipula ion, an indica ion o some so budge cons ain s. While he ad e se
incen i e e ec o hese p oblems in he case o in e -go e nmen al ela ionships migh be
modes , i could be much mo e se ious in he case o he manage ial con ac esponsibili y
schemes in he SOE e o m (Wu, 1995; Qian, 1996).
3. Replacing he Sys em: Since 1994
Since 1994, PRC’s ansi ion has mo ed in o he second s age which aimed o eplace he
planned sys em wi h a ma ke sys em. I will add ess h ee issues in his sec ion. Fi s , I will
show ha PRC se a goal, becoming clea e o e ime, o es ablish a ule-based ma ke
economy inco po a ing in e na ional bes p ac ice ins i u ions as we know hem. In ac , PRC
has become he i s and he only coun y in which he uling Communis Pa y olun a ily
shi ed i s o icial ideology o emb ace a ma ke sys em and p i a e owne ship. Because
many people ou side o PRC a e s ill unawa e o i , I will p o ide some de ailed e idence o
show ha his is indeed he case. Second, I will examine ad ancemen in se e al majo a eas
owa d his goal in he i s i e yea s o he second s age (be ween 1994 and 1998). The
e idence shows ha signi ican p og ess has been made, which should cla i y many doub s
o e whe he PRC’s e o m is con inuing. Finally, I will p o ide an analysis o he poli ical
economy o e o m and he dynamics o ansi ion, ocusing on how he ea ly e o ms c ea ed
cons i uencies and momen um o u he e o m wi hou gene a ing obs acles o block i .

15
A. S a egic Mo e: Se ing he Goal o a Ma ke Sys em
A he ou se o e o m, PRC desi ed e o m in o de o inc ease p oduc i i y and imp o e
li ing s anda ds, bu a no ime did he leade ship hink ha i was going o a ull ma ke
sys em (Pe kins, 1994). This s a ed o change in he ea ly 1990s. Howe e , es ablishing ha
goal in PRC was an e olu iona y p ocess, and in ha aspec , PRC is unlike some Eas e n
Eu opean coun ies whose goal was e y speci ic a he beginning o ansi ion, ha is, o
build a ma ke economy like hei neighbo s and join he Eu opean Union. I will highligh
ou consecu i e miles one e en s in Sep embe 1992, No embe 1993, Sep embe 1997, and
Ma ch 1999, espec i ely, o show how his s a egic mo e has e ol ed p og essi ely o e
ime. The Fou een h Pa y Cong ess o Sep embe 1992. Du ing much o he i s s age o
e o m, he o icial ideology was he one o “combining plan and ma ke oge he .” A mo e
conse a i e e sion o i was “planning supplemen ed by ma ke ” and a mo e libe al e sion
was “planned commodi y economy.” Clea ly, he goal o e o m was no a ma ke sys em. In
he sp ing o 1992 Deng Xiaoping made his amous Sou he n ou o mobilize local suppo o
u he and mo e adical e o m. The big ideological b eak h ough occu ed a e wa ds a he
Fou een h Pa y Cong ess in Sep embe 1992 when he Pa y, o he i s ime, endo sed he
“socialis ma ke economy” as PRC’s goal o e o m. I is impo an o dis inguish he PRC
socialis ma ke economy om “ma ke socialism” as ad oca ed by some Eas e n Eu opean
e o me s in he 1970s and 1980s. In ma ke socialism, ma ke is a simula ed one o se e he
pu pose o socialism based on public owne ship (Ko nai, 1992). In con as , in a socialis
ma ke economy, “ma ke economy” is he goal and he wo d “socialis ” in an adjec i e.
The e o e, a socialis ma ke economy di e s om a ma ke socialism in a undamen al way
while i is close o he “social ma ke economy” o Ge many.
The Decision o No embe 1993. The con en s o ansi ion o “socialis ma ke
economy” only became clea e one yea la e . In 1993, he Communis Pa y’s Economics and
Finance Leading G oup, headed by Pa y Sec e a y Gene al Jiang Zemin, wo ked oge he wi h
economis s o p epa e a g and s a egy o ansi ion o a ma ke sys em. Se e al esea ch eams
we e o med o s udy a ious aspec s o ansi ion, anging om axa ion, he iscal sys em, he
inancial sys em, and en e p ises, o o eign ade. The inal ou pu was he “Decision on Issues
Conce ning he Es ablishmen o a Socialis Ma ke Economic S uc u e” adop ed by he Thi d
Plenum o he Fou een h Pa y Cong ess in No embe 1993.6 Wi h he objec i e o a ma ke
sys em in mind, his landma k documen made ou majo ad ances in he a eas o e o m
s a egy, a ule-based sys em, building ma ke -suppo ing ins i u ions, and p ope y igh s and
owne ship espec i ely. I was he u ning poin on PRC’s oad o ma ke s.
Fi s , unlike he p e ious s a egy o “g oping o s ones o c oss he i e ,” he
“Decision” emphasized he impo ance o coo dina ion among a ious aspec s o e o ms. I
ad oca ed a cohe en package and an app op ia e sequencing o e o ms, known as “combining
package e o m wi h b eak h ough in key a eas.”
Second, in he i s s age o e o m, pa icula is ic con ac ing played a dominan ole,
such as iscal con ac ing, manage ial con ac ing, and household con ac ing. Fo he i s ime,
he “Decision” called o a ule-based ma ke sys em o c ea e a le el playing ield. This
included he decision o uni y he o eign exchange a e and ax a es among all en e p ises
6 Fo he ull ex o he “Decision,” see China Daily, Supplemen , No embe 17, 1993.
16
ega dless o owne ship. Sho ly be o e he “Decision,” PRC had al eady adop ed Wes e n
accoun ing ules in i s en e p ises.
Thi d, he “Decision” ocused on he building o ma ke -suppo ing ins i u ions, such as
o mal iscal ede alism, a cen alized mone a y sys em, and a social sa e y ne . Fo example,
sepa a ion o cen al and local axes and hei adminis a ion was a c i ical s ep in mo ing
owa d o mal iscal ede alism. Re enue ans e s be ween he cen al and p o incial
go e nmen s we e o be based on a ixed o mula a he han ba gaining. Ano he example was
cen aliza ion o he ope a ion o he cen al bank o minimize he local go e nmen ’s in luence
on mone a y policies.
Finally, he “Decision” add essed he en e p ise e o m issue in e ms o p ope y igh s
and owne ship, a he han, as be o e, one o “expanding en e p ise au onomy.” I in ended o
ans o m SOEs in o “mode n en e p ises” wi h “cla i ied p ope y igh s, clea ly de ined
esponsibili y and au ho i y, sepa a ion o en e p ises om he go e nmen , and scien i ic
in e nal managemen .” I has also, o he i s ime, le he doo open ega ding he
p i a iza ion o SOEs: “As o he small s a e owned en e p ises, he managemen o some can
be con ac ed ou o leased; o he s can be shi ed o he pa ne ship sys em in he o m o s ock
sha ing, o sold o collec i es and indi iduals.” Bu he majo b eak h ough on owne ship
issues had o wai a while longe .
The Fi een h Pa y Cong ess o Sep embe 1997. In he “Decision” o No embe 1993,
s a e owne ship was s ill ega ded as a “p incipal componen o he economy” while p i a e
owne ship was a “supplemen a y componen o he economy.” The Fi een h Pa y Cong ess
held in Sep embe 1997 made a majo b eak h ough on owne ship issues: S a e owne ship was
downg aded o a “pilla o he economy” and p i a e owne ship was ele a ed o an “impo an
componen o he economy.” In PRC poli ics, hese sub le changes o he o ic mean a big
change in ideology. The documen ecognized ha “ a ie ies o owne ship should de elop
oge he ,” bu because p i a e owne ship was disc imina ed agains o decades, he majo new
in o ma ion he e was ha p i a e owne ship had gained legi imacy. Fu he mo e, al hough he
he o ic o public owne ship was main ained, i s meaning was ede ined, because public
owne ship may ha e many “di e en ealiza ion o ms,” such as join s ock co po a ions wi h
in es men by many owne s. A his ime, he o icial ideology owa d p i a e owne ship
inally became “ iendly.”
The second majo b eak h ough o he Fi een h Pa y Cong ess was i s emphasis on he
ule o law. The ule o law is no he same as democ acy. Fo example, he wo mos ee
ma ke economies, Hong Kong, China; and Singapo e, ha e he ule o law bu a e no
democ acies, by Wes e n s anda ds. PRC seemed o decide o gi e p io i y o he ule o law
a he han democ acy in i s sequencing o poli ical e o ms and o conside he ule o law
c ucial o a mode n ma ke economy o wo k well.7 As always in PRC, he con en o he
ule o law will e ol e o e ime.
The Cons i u ional Amendmen s o Ma ch 1999. P i a e owne ship and he ule o law
we e inco po a ed in o he PRC Cons i u ion in Ma ch 1999.8 An amendmen o A icle 11 o
he Cons i u ion places p i a e businesses on an equal oo ing wi h he public sec o by
changing he o iginal clause “ he p i a e economy is a supplemen o public owne ship” o “ he
non-public sec o , including indi idual and p i a e businesses, is an impo an componen o
7 Ko nai (1998) has emphasized ha democ acy is no a necessa y basis o a ma ke sys em bu a poli ical
powe ha is iendly o p i a e p ope y and he ma ke is.
8 “Top lawmake s yes e day o e whelmingly endo sed China’s [sic] landma k cons i u ional amendmen s
which ensh ine he ‘ ule o law’ and bols e he s a us o p i a e businesses” (China Daily, Ma ch 16, 1999).
17
he socialis ma ke economy” (China Daily, Ma ch 16, 1999). Immedia ely a e he
amendmen , local go e nmen s s a ed o elax local es ic ions on p i a e en e p ises. Fo
ins ance, he Jiangsu p o incial go e nmen adop ed a new policy o gi e p i a e en e p ises
equal ea men as s a e-owned and collec i e en e p ises in he a eas o g an ing business
scopes and c edi access (People’s Daily, Ap il 9, 1999). Fu he mo e, A icle 5 o he
Cons i u ion was amended o include he p inciple o “go e ning he coun y acco ding o law”
(China Daily, Ma ch 16, 1999). These Cons i u ional amendmen s ha e demons a ed PRC’s
commi men o a ull ma ke sys em based on he ule o law.
The signi icance o he s a egic mo e. The ailu e o Eas e n Eu opean e o m p io o
1990 has made a pe suasi e a gumen o he necessi y o ha ing democ a ic e o m p ecede
economic ansi ion (Ko nai, 1992). The Communis Pa ies he e we e unwilling o change
hei ideology. The collapse o he Communis Pa ies in Eas e n Eu ope was he logical
consequence. PRC p o ided a case ha p o ed impossible in Eas e n Eu ope and elsewhe e:
he Communis Pa y o PRC i sel made he s a egic shi olun a ily. I appea s ha PRC is
he i s and only coun y unde he Communis Pa y sys em o emb ace p i a e owne ship
and he ule o law in i s Cons i u ion.
I is qui e ema kable o PRC o ha e o e come ideological and poli ical opposi ion o
emb ace he ma ke sys em and p i a e owne ship wi hou a poli ical e olu ion. The e we e
bo h in e nal and ex e nal easons. The in e nal eason was ha by 1993 e o m in PRC had
bene i ed almos e e ybody, who en isioned bigge bene i s om u he e o m. Solid
poli ical suppo o u he e o m came om bo h he op leade s and he o dina y people,
which p o ided a poli ical basis o ideological change (see mo e on his in subsec ion C
below). The ex e nal eason was ha he ou side p essu e om Eas Asian coun ies was
eno mous; hei ex ao dina y pe o mance p essed PRC o mo e o wa d. A he same ime,
compe i i e p essu e om he Eas e n Eu opean ansi ion o ma ke s was also signi ican . The
PRC go e nmen ea ed ha hese ansi ion economies would soon ca ch up wi h PRC, which
migh unde mine i s legi imacy ha was buil en i ely upon he economic pe o mance o he
coun y.
The popula sen imen in PRC o a ansi ion o a con en ional ma ke economy can be
sensed om he wo ph ases which ha e been popula in he 1990s: yu guoji jiegui, which
means connec o (o join) he in e na ional ack (o p ac ice, o s anda d); and gui an hua,
which means s anda diza ion. In No embe 1998, majo newspape s in PRC published Joseph
S igli z’s (1998) a icle “Second-Gene a ion S a egies o Re o m o China,” which con ains
policy ecommenda ions o PRC o deepe ins i u ional es uc u ing o comple e i s ansi ion.
This can be in e p e ed as a s ong de e mina ion om he op leade ship o comple e PRC’s
mo e o ma ke s.
B. Majo Accomplishmen s in he Fi s Fi e Yea s (1994-98)
Following he “Decision” o No embe 1993, a se ies o adical e o ms we e launched
s a ing in Janua y 1994. In building ma ke ins i u ions, PRC s a ed a li le bi la e compa ed
o Eas e n Eu ope o e en Russia. Today, i s ill lags behind he s a coun ies in Eas e n
Eu ope such as Poland, Hunga y, and he Bal ic coun ies. Bu his g oup o coun ies may be
excep ional because o hei aspi a ions o join he Eu opean Union. So a , PRC’s p og ess
has been s eady, and i has been able o a oid bad mis akes made by some o he ansi ion
economies.
In building a ma ke sys em, PRC con inued i s p e ious p ac ices ha we e use ul, such
as go e nmen decen aliza ion, inancial s abili y, and compensa ion o lose s in he e o m
18
p ocess. I began o add ess he h ee p oblems le om he i s s age. Fi s , i co ec ed some
ea ly mis akes, such as mone a y decen aliza ion and so budge cons ain s on local
go e nmen s and SOEs. Second, i began delayed e o ms, such as p i a iza ion o s a e
en e p ises and comme cializa ion o s a e banks. Thi d, and mos impo an , PRC s a ed o
inco po a e in e na ional bes p ac ice ma ke ins i u ions based on uni o m ules, bu in doing
so, i i s c ea ed new ins i u ions be o e des oying he old ones. Below, I examine some
majo accomplishmen s in hose i s i e yea s. Many o he e o ms will con inue in o he
nex decade, bu he e idence so a shows ha PRC is mo ing in he igh di ec ion and
p oceeding a a s eady pace.
Uni ica ion o o eign exchange a es and con e ibili y o he cu en accoun . Be o e
1994, libe aliza ion o o eign exchange ma ke s, like many o he ma ke s, ollowed a dual-
ack app oach and he e exis ed an o icial a e and a “swap a e” (i.e., he ma ke a e).
Because o he d ama ic g ow h o he ma ke ack, by 1993 he sha e o he plan alloca ed
o eign exchange had allen o less han 20 pe cen o he o al. On Janua y 1, 1994, plan
alloca ion o o eign exchange was comple ely abolished, and he wo acks we e me ged in o
a single ma ke ack. Howe e , o hose o ganiza ions which we e used o ecei ing cheap
o eign exchange, annual lump-sum subsidies in he domes ic cu ency su icien o enable he
pu chase o he p e ious alloca ion o o eign exchange we e o e ed o a pe iod o h ee yea s
o compensa e o hei losses. In Decembe o 1996, PRC wen one s ep u he o announce
cu en accoun con e ibili y o i s cu ency. Howe e , i did no mo e o capi al accoun
con e ibili y and ye s ill main ained capi al con ol. This is one impo an eason ha PRC
wea he ed he Asian inancial c isis a he well.
Be ween 1994 and 1998, he exchange a e emained s able and e en app ecia ed sligh ly
om 8.7 yuan pe U.S. dolla o 8.3 yuan pe U.S. dolla . Bo h expo s and o eign di ec
in es men inc eased d ama ically, and he coun y’s o eign ese es inc eased om US$21
billion o US$145 billion. Despi e he Asian inancial c isis, PRC con inued o a ac o eign
di ec in es men o abou US$45 billion annually in 1997 and 1998.
The o e haul o he ax and iscal sys ems. Be o e 1994, he iscal con ac ing sys em had
played a posi i e ole o p o iding badly needed incen i es o local go e nmen s. Bu he
iscal con ac ing was ad hoc and no ule-based. Also, PRC did no ha e a na ional ax bu eau,
and all axes we e collec ed by local go e nmen s, which o en educed o exemp ed axes ha
we e supposed o be paid o he cen al go e nmen .
On Janua y 1, 1994, PRC in oduced majo ax and iscal e o ms mo e aligned wi h
in e na ional p ac ices. This e o m in oduced a clea dis inc ion be ween na ional and local
axes and es ablished a na ional ax bu eau and local ax bu eaus, each esponsible o i s own
ax collec ion. This ax e o m has made i e y di icul o local go e nmen s o educe
na ional axes as hey did in he pas (Dong, 1997). Re o m also es ablished ixed ax ules
be ween he na ional and local go e nmen s. Fo example, unde he new sys em, he alue
added ax (VAT) became he majo indi ec ax sha ed by he na ional and local go e nmen s a
a ixed a io o 75:25. Bu local go e nmen s we e compensa ed o hei e enue losses o
h ee yea s.
In 1995, he new “Budge Law” ook e ec . I p ohibi ed he cen al go e nmen om
bo owing om he cen al bank and om de ici inancing i s cu en accoun , bu he cen al
go e nmen could ha e de ici inancing in i s capi al accoun al hough i had o inance he
de ici wi h go e nmen bonds. I also imposed mo e s ingen es ic ions on local
go e nmen s. Local go e nmen s a all le els we e equi ed o ha e hei budge s balanced (as
be o e), and u he mo e, he law s ic ly con olled hei bond issuance and es ic ed hei
bo owing in he inancial ma ke (a change om he pas ). To ensu e en o cemen o he
25
comme cial banks, was ecen ly selec ed by he cen al bank o expe imen wi h ways o
cleaning up i s bad deb s h ough a newly-es ablished “ inancial asse s managemen company”
(Sing ao Daily, Ma ch 6, 1999).
The inancial e o m equi es ca e ul esea ch on bo h he low and s ock p oblems. To
add ess he low p oblem in a undamen al way, PRC needs o in oduce a bank up cy
p ocedu e which clea ly speci ies he igh s and obliga ions o c edi o s (a p esen ime mainly
s a e banks) in he e en o bank up cy o bo h liquida ion and eo ganiza ion ypes. In
pa icula , banks, as majo c edi o s, should ha e he igh s o ecei e ship and he
esponsibili y o eo ganiza ion o he bank up i ms. Because he cu en PRC law does no
allow banks o hold any equi y o i ms, esea ch is needed o in oduce new egula ions which
would pe mi banks o empo a ily hold equi y du ing he eo ganiza ion pe iod. On he issue o
cleaning up he bad deb s o s a e banks, one needs o ind a way o ensu e ha such a clean up
will no esul in a mo al haza d p oblem om he s a e banks. To ha end, he go e nmen
mus make a c edible commi men o no mo e bailou s in he u u e. Resea ch is also needed o
iden i y me hods o achie e his goal.
Ano he esea ch opic conce ns app op ia e sequencing o inancial libe aliza ion. In he
case o Japan, Hoshi and Kashyap (1999) a gue ha i is no he Japanese main bank sys em pe
se bu he p oblema ic sequencing o inancial de egula ion ha explains he banking c isis
he e. De egula ion allowed la ge co po a ions o quickly swi ch o he capi al ma ke o di ec
inancing, and a he same ime, he banks’ new lending p ima ily lowed o high isk small
businesses and became ied mo e s ongly o he bubble economy. As he bubble bu s , a
banking c isis eme ged. PRC may ace simila p oblems when i s s a e banks become mo e
independen om he go e nmen and shi hei lending po olios o smalle i ms. To educe
inancial isks a he ime o inancial libe aliza ion, PRC needs o lea n he lessons om he
ecen Asian expe iences.
B. S a e-Owned En e p ises and Co po a e Go e nance
Re o ms o s a e-owned en e p ises ha e been disappoin ing. The manage ial con ac
esponsibili y sys em p omo ed in he 1980s had only limi ed success and SOE pe o mance
in he 1990s con inued o decline (Wu, 1995). Fo ins ance, he e we e mo e han one- hi d o
SOEs making losses, and on a e age, p o i s and axes pe uni o ne capi al s ock and
wo king capi al in s a e indus ial en e p ises ell om 24.2 pe cen in 1978 o 12.4 pe cen in
1990 and u he down o 6.5 pe cen in 1996 (China S a is ical Yea book, 1997). E en
hough new p i a e i ms a e de eloping and small SOEs a e being p i a ized, his is no a
subs i u e o ans o ming la ge SOEs, since hey s ill cons i u e he backbone o he
economy, a e he main e enue sou ce and inancial bu den o he go e nmen , and a e
ul ima ely esponsible o he inancial sec o p oblem.
Al hough he excessi e “social bu dens” o SOEs (i.e., wo ke s’ wel a e obliga ions) a e
o en blamed o hei poo pe o mance, he main p oblems a e ins i u ional, conce ning
inance ( he so budge cons ain p oblem) and pe sonnel appoin men s ( he Pa y con ol
p oblem). The so budge cons ain p oblem has been discussed in Sec ions 2 and 3, and in
ecen yea s he budge cons ain s o SOEs ha e been ha dening. Bu Pa y con ol o e
pe sonnel has emained basically unchanged o he pas 20 yea s. The Pa y has exe cised
con ol o e he selec ion and dismissal o SOE manage s h ough i s O ganiza ion
Depa men s a di e en le els. Fo example, he Cen al Pa y O ganiza ion Depa men has
he au ho i y o e appoin men s o he op manage s o e y la ge SOEs (minis e o depu y
minis e le el), as does he P o incial (o Municipali y) Pa y O ganiza ion Depa men o

26
mos la ge and medium-sized SOEs (bu eau le el). This au ho i y applies o join -s ock
companies as long as he s a e has he majo i y sha e, e en i hey a e lis ed on he s ock ma ke
o a e loca ed in he special economic zones. The appoin men and dismissal p ocess ep esen s
he mos impo an channel o poli ical in luence o e en e p ises by he Pa y appa a us.
Unde he Pa y con ol pe sonnel sys em, SOE manage s, like mayo s, minis e s, and
Poli bu o membe s, a e poli ical appoin ees o he Pa y. This poli ical p ocess o manage ial
appoin men s has se e al se ious p oblems. Fi s , he appoin men p ocess is poli icized,
sec e i e, and complica ed. When he Pa y selec s bo h manage s and poli icians a he same
ime, i may no choose he igh people as manage s. Second, he selec ion and e alua ion
me hods a e based on in o ma ion h ough bu eauc a ic a he han ma ke channels ( he la e
includes he s ock ma ke , a ing companies, and in es men banks). Thi d, he Pa y
bu eauc a s ha e nei he he abili y no he incen i es o make he igh decisions on manage ial
selec ion acco ding o business c i e ia because hey a e mainly poli ically mo i a ed.11
Th ough “expanding en e p ise au onomy” and “inc easing en e p ise e ained p o i s”
( angquan angli), he pas e o ms imp o ed manage ial incen i es somewha bu he e ec
was a he limi ed and sho -li ed. Why? The main di icul y o e o ming SOEs wi hou
p i a izing hem is e lec ed in he ollowing ins i u ional dilemma (Qian, 1996): On he one
hand, delega ing mo e e ec i e con ol igh s o manage s p o ides hem wi h incen i es o
inc ease cu en p oduc ion bu also enables hem o plunde s a e asse s, which esul s in high
agency cos s. On he o he hand, main aining Pa y con ol o e he selec ion and dismissal o
manage s se es o check manage ial asse s ipping somewha bu is also he ul ima e sou ce
o poli ical in e e ence, esul ing in high poli ical cos s. The SOE p oblem in PRC is he
in e wining p oblem o bo h high agency cos s and poli ical cos s.
Be o e being p i a ized, la ge SOEs (including s a e banks) should aim o educe bo h
agency and poli ical cos s, which can only be achie ed by a combina ion o es ablishmen o
co po a e go e nance and depoli iciza ion (Qian, 1996). Co po a e go e nance is a se o
ins i u ional a angemen s go e ning he ela ionships among in es o s (sha eholde s and
c edi o s), manage s, and wo ke s. The s uc u e o co po a e go e nance conce ns (1) how
con ol igh s a e alloca ed and exe cised; (2) how boa ds o di ec o s and op manage s a e
selec ed and moni o ed; and (3) how incen i es a e designed and en o ced. Co po a iza ion, a
p ocess o es ablishing co po a e go e nance, is a use ul s ep in en e p ise e o m e en wi hou
p i a iza ion, because i helps hold di ec o s esponsible o he asse s o he company and
p e en u he asse he ; p o ides a mechanism o in o ma ion exchange; se s he s age o
selling sha es; and sepa a es he s a e om en e p ise. In Eas e n Eu ope and in o he
de eloping coun ies, co po a iza ion was ca ied ou be o e p i a iza ion and imp o ed
en e p ise pe o mance.
In de eloped economies, majo issues o co po a e go e nance conce n legal ules
limi ing he agency p oblems, p o ec ing sha eholde s and c edi o s, and p o iding oom o
manage ial ini ia i es. The same p oblems a ise in PRC, bu wi h a special conce n abou he
ole o he s a e as a la ge s akeholde . In addi ion o co po a e go e nance, depoli iciza ion is
also necessa y, which means ha he Pa y’s di ec con ol o e manage s should be limi ed, i
no elimina ed. Unless he issue o he Pa y’s ole is add essed, he goal o “sepa a ion o
go e nmen and en e p ise” canno be ma e ialized, and unless he s a e, ins i u ional in es o s,
11 I is in e es ing o compa e SOEs wi h TVEs in his ega d. Al hough TVE manage s a e appoin ed by
ownship o illage go e nmen s, hey do no go h ough he highe le el Pa y appa a us and hus hey a e no
subjec o he same poli ical p ocess as SOE manage s. A e all, mos TVE manage s a e no “s a e cad es.”
27
and indi idual in es o s a e pu on an equal oo ing, poli ical in e en ion by he go e nmen
will con inue o plague he pe o mance o hese la ge i ms.
Co po a iza ion and depoli iciza ion a e di icul asks o PRC because hey necessi a e
some poli ical e o m. Fu u e esea ch needs o ind a way o limi he Pa y’s ole in he
appoin men and dismissal p ocess o SOE manage s wi hou elimina ing i s leade ship ole.
The e we e sugges ions abou he c ea ion o he S a e Asse s Managemen Commi ee o
o e see s a e asse s in en e p ises. In such a scheme, only he Commi ee ep esen a i es ha e
au ho i y, h ough he boa d o di ec o s and depending on he numbe o sha es held by he
s a e, o e he appoin men and dismissal o op manage s (Wo ld Bank, 1997). Many
ques ions emain. Fo example, who will appoin he membe s o he S a e Asse s Managemen
Commi ee? The Pa y, a go e nmen agency, he People’s Cong ess, o o he s? Will such a
pe son o ins i u ion s op in e e ing in he appoin men and dismissal p ocess o he boa d o
di ec o s? Mo e esea ch is needed o make he scheme poli ically accep able and economically
sensible.
C. The Rule o Law
The bigges challenge o PRC o comple e i s ma ke ansi ion is he es ablishmen o he
ule o law. The ecen Cons i u ional amendmen on he ule o law is a good s a . The
economic ad an ages o he ule o law o e ad hoc a angemen s a e anspa ency,
p edic abili y, and uni o mi y, ha educe idiosync a ic isks, en -seeking, and co up ion,
which in u n educe ansac ion cos s. Bu he ule o law is mo e han pu ing he
go e nmen ’s wo ds in o public codes; i undamen ally conce ns a ela ionship be ween he
go e nmen and ma ke s ha is app op ia e o making a c edible commi men .
The e a e wo economic oles o he ule o law. The i s is ha he law should be
applied o he go e nmen — he go e nmen needs o be cons ained by law is-a- is o he
economic agen s in he ma ke . Th ough he ule o law, he go e nmen binds i sel and hus
makes a c edible commi men o he p o ision o p i a e incen i es, which a e he ul ima e
o ce o economic de elopmen . This ole o he ule o law p o ides a ounda ion o secu e
p i a e p ope y igh s agains go e nmen in usion, and i e lec s he c ucial di e ence
be ween he “ ule o law” and he “ ule by law.”
One se ious p oblem unde mining p i a e incen i es in PRC has been insecu e p i a e
p ope y igh s. Fo example, p i a e en e p ises and a me s a e equen ly o ced o pay
a ie y o “unau ho ized ees” o local go e nmen s, and hei businesses a e also ha assed by
local go e nmen s. Re enue hiding is a esponse o go e nmen p eda ion, bu is cos ly.
Recen ly, p i a e businessmen and a me s began o use legal means o p o ec hemsel es
agains go e nmen ’s in usion. Fo example, The Wall S ee Jou nal (Ma ch 25, 1999)
epo ed a case o Peijiawan illage in Shaaxi P o ince, whe e 12,000 a me s in 1996 iled a
class-ac ion lawsui agains he local go e nmen o le ying excess ees o $75,000. In he all
o 1998, he local cou made an ini ial uling in a o o he a me s, he local go e nmen
appealed, and now he case is be o e he Shaaxi P o incial Sup eme Cou . This is an
unp eceden ed e en in PRC.
The second economic ole o he ule o law is ha he go e nmen needs o p o ec
p i a e p ope y igh s and en o ce con ac s and o c ea e a le el ield o ma ke compe i ion.
This will help achie e c edible commi men among economic agen s. To achie e his goal, he
go e nmen needs o become a neu al hi d pa y, a egula o a he han a manage . Du ing
he ecen es uc u ing o he go e nmen bu eauc acy, many go e nmen agencies we e
elimina ed and he ole o he newly es ablished agencies has been eo ien ed om
28
managemen o egula ion. One example is he Minis y o In o ma ion Indus y. I used o
di ec ly adminis e China Telecom, he elephone monopoly in PRC. Now, China Telecom is
being b oken up in o h ee companies, while o eign compe i ion is being in oduced, and he
Minis y becomes he egula o y agency o he en i e elecommunica ions indus y (Sing ao
Daily, Ma ch 6, 1999).
To p ese e ma ke compe i ion, he na ional go e nmen needs o igh agains egional
p o ec ionism. In he pas , ma ke compe i ion in PRC was la gely an acciden o egional
decen aliza ion unde Mao, combined wi h he mo e-o -less spon aneous eme gence o small-
scale indus ies h oughou he coun y. F om ime o ime, he e ha e been many epo s ha
local go e nmen s e ec ed ade ba ie s o block goods and ac o mo emen in o de o p o ec
local in e es s. PRC needs a ee in e s a e comme ce clause as in he U.S. Cons i u ion and he
cen al go e nmen should ha e he esponsibili y o police he common ma ke agains
egional ma ke p o ec ionism.
Fu u e esea ch is needed o s udy speci ic ways o implemen ing he ule o law. Fo
example, an impo an opic conce ns how o c ea e a subs an ially independen judicia y
sys em, s a ing a he g ass oo s le el, and hen g adually mo ing upwa d. The ule o law
equi es an independen and unco up ed judicia y sys em o unc ion. A comple ely
independen judicia y sys em seems impossible unde a one-Pa y sys em, bu i may no be
necessa y o he ime being because a subs an ially independen judicia y sys em can achie e
qui e a lo economically o he ule o law o wo k. The la e can be done because PRC is such
a la ge coun y and mos o he economic decisions a e made a he local le el. Building a
judicia y sys em ha is independen o he local go e nmen ’s in luence is easible a he
p esen ime, and he abo e example o 12,000 a me s in Peijiawan illage suing he local
go e nmen illus a es his possibili y.
Two decades ago ew economis s p edic ed ha PRC would go his a and accomplish
so much. PRC has su p ised he wo ld. Al hough pas pe o mance is no gua an ee o u u e
esul s, he e a e good easons o being op imis ic. Today, he ideology agains ma ke s and
p i a e owne ship has subsided, he goals o ansi ion ha e been se , many ma ke -suppo ing
ins i u ions a e being buil , and he Chinese people a e mo e willing han e e o in eg a e
PRC’s economy wi h he global one. Fo PRC o success ully comple e i s ansi ion o ma ke s,
p io i y on he esea ch agenda should be gi en o he h ee a eas ou lined abo e.
5. Re lec ions on he Economics o Re o m and Ins i u ional Change:
Lessons om PRC
Economis s o en o e ad ice o ansi ion economies based on hei knowledge abou
economics. Bu economics has as much o lea n om he expe iences o he ansi ion
economies as i has o each hem (Walde , 1995). This can be done in wo ways: One is by
using he da a gene a ed om he ansi ion economies o es exis ing economic heo ies.
Ano he is h ough inspi a ion om he expe iences o he ansi ion economies o ex end he
exis ing heo ies o e en de elop new ones. S udying PRC’s e o m is use ul o bo h pu poses.
Much esea ch has been done using da a om PRC o es exis ing heo ies. Bu s udying
PRC’s expe ience is e en mo e use ul o gene a ing new heo ies, especially in he a ea o
ins i u ional changes. This is because PRC’s ansi ion pa h is so unusual ha i cas s doub on
much o he con en ional hinking on undamen al issues conce ning sys em changes and he
p ocess o e o m (Chow, 1997).
29
A. Re lec ions on he P inciples o Sys em Change
PRC di e s om Eas e n Eu ope and Russia in many impo an ways including hei ini ial
condi ions. PRC had a la ge ag icul u e sec o which was he sp ingboa d o i s e o m; PRC’s
mo e decen alized planning sys em made i na u al o pu sue e o m in a mo e decen alized
way based on egional expe imen s; and he en-yea u moil o he Cul u al Re olu ion
disc edi ed cen al planning and subs an ially weakened he go e nmen (including he Pa y)
bu eauc acy and i s es ed in e es s.12 Hence, PRC is unlikely o p o ide a model o Eas e n
Eu ope o Russia o ollow. Howe e , because ou p e ious hinking on e o m and ansi ion
was almos exclusi ely based on he expe iences o Eas e n Eu ope, he con en ional wisdom
is biased. I is he e o e use ul now o inco po a e PRC — he la ge hal o all ansi ion
economies — in o he analysis.
I is commonplace o compa e he PRC e o m expe ience since 1978 (and o en, up o
1994) wi h he ansi ion expe iences o Eas e n Eu ope and Russia a e 1990 and o con as
he di e ence be ween he wo as “g adualism” s. “big bang.” This is an inapp op ia e
compa ison. In Eas e n Eu ope, e o ms s a ed as ea ly as 1968 in Hunga y, 1980 in Poland,
and 1985 in he So ie Union. PRC, Hunga y, Poland, and he So ie Union all wen h ough a
wo-s age p ocess: i s e o ming he planned sys em and hen building he ma ke sys em.
When compa ing PRC wi h Eas e n Eu ope a a compa able s age, I d aw h ee conclusions.
The i s conclusion, based on PRC’s achie emen s du ing he i s i een yea s o
e o m, is ha e o ming a planned economy, bo h as a single e en and as a s ep along he
longe pa h, can be success ul. The e o ms in Eas e n Eu ope p io o 1990 we e a ailu e.
Ko nai (1992) made a con incing analysis o hem and eached he ollowing conclusion: “In
spi e o gene a ing a whole se ies o a o able changes, e o m is doomed o ail: he socialis
sys em is unable o enew i sel in e nally so as o p o e iable in he long un.” In e ospec ,
he e o ms unde aken in PRC be ween 1978 and 1993 we e much mo e comp ehensi e and
adical han hose in Eas e n Eu opean p io o 1990, he la e ha ing no se ious e o m o he
go e nmen , no dynamic de elopmen o non-s a e en e p ises, no inancial deepening, and no
eal ma ke s. Tha is why PRC’s ea ly e o m was mo e success ul han hose in Eas e n
Eu ope. The e idence om PRC is compelling, demons a ing ha he Eas e n Eu opean
expe ience is no uni e sal.
The second conclusion, based on he PRC expe ience o he las i e yea s, is ha sys em
change om a planned sys em o a ma ke sys em can occu wi hou a poli ical e olu ion. The
Eas e n Eu opean e o m in he 1970s and 1980s led o a dead end because “ he sys em is
incapable o s epping away om i s own shadow” (Ko nai, 1992). I e en ually esul ed in a
poli ical e olu ion ha jump-s a ed he ansi ion in he 1990s. Al hough i is s ill oo ea ly o
p edic ha PRC can success ully comple e i s ansi ion o a ma ke sys em, he e is enough
e idence o show ha he goal is se , he jou ney is well unde way, and he chance o a aining
he goal is good. The e o e, he p e ious conclusion based on he Eas e n Eu opean expe ience
ha planned economies a e unable o eplace hei sys ems wi h ma ke sys ems needs o be
econside ed. I PRC comple es i s ansi ion, i will be he i s coun y unde he Communis
Pa y o do so.
The hi d conclusion, based on PRC’s wo decades o e o m, is ha he e a e di e se
pa hs o a success ul ansi ion. A good pa h o ansi ion, especially conce ning ins i u ional
changes, has o conside coun y-speci ic condi ions. PRC p o ides a case o a easible and
12 Due o space limi a ion, his pape will no discuss how ini ial condi ions a ec he ansi ion pa h. See Qian
and Xu (1993), and Sachs and Woo (1994) o discussions.
30
success ul al e na i e pa h o ansi ion o ma ke s, in which no poli ical e olu ion p eda ed he
change o economic sys ems as was he case in Eas e n Eu ope. Hence, he assump ion ha all
planned economies we e alike and ha hey should ollow he same pa h o ansi ion needs o
be modi ied.
PRC’s case shows ha e o ming a socialis economy can be success ul; he planned
sys em can e ol e smoo hly o a ma ke sys em; and he e a e di e se pa hs o making a good
ansi ion. I may be a gued ha PRC is special because i was a poo ag icul u al coun y. I
may also be a gued ha Hunga y and Poland a e special because hei ea ly e o ms we e
cons ained by no being uly independen om he So ie Union, and hei la e ansi ions
we e acili a ed by hei aspi a ions o join he Eu opean Union. In any e en , he p inciples o
sys em changes a e mo e complica ed han one used o belie e.
B. Re lec ions on he P ocess o Re o m
A e wen y yea s o expe ience, PRC’s e o m can con ibu e se e al lessons abou he
p ocess o e o m in pa icula and ins i u ional change in gene al. The main lesson is ha
conside able g ow h is possible wi h sensible bu no pe ec ins i u ions, and some “ ansi ional
ins i u ions” can be mo e e ec i e han he “bes p ac ice ins i u ions” o a pe iod o ime
because o he second-bes p inciple: i.e. emo ing one dis o ion may be coun e -p oduc i e in
he p esence o ano he dis o ion.
In ansi ion economies basic ma ke -suppo ing ins i u ions (such as he ule o law) a e
lacking, as well as he people and human capi al o ope a e hem (such as law en o cemen ).
Bo h usually ake yea s o de elop, excep in he case o Eas Ge many, which was abso bed
in o Wes Ge many immedia ely a e uni ica ion. This o en means ha he in e na ional bes
p ac ice ins i u ions, e en i we know wha hey a e, may no wo k well o he ime being. I
also means ha some exis ing ins i u ions can be use ul o ma ke -o ien ed e o m e en hough
hey may e en ually anish. Fiscal con ac ing, TVEs, and anonymous banking a e all
ins i u ional inno a ions ha wo ked qui e well o a ce ain ime pe iod in he PRC
en i onmen .
All ansi ion economies ha e g ea po en ial o imp o emen because o eno mous
alloca i e dis o ions and he lack o incen i es o he planning sys em. The e o e, in heo y,
e en i ins i u ions a e highly impe ec , imp essi e esul s a e po en ially possible. Howe e ,
many ailu es in he ea ly e o ms and he ecen ansi ion sugges ha no all changes can
p oduce good esul s. Wha kinds o changes can p oduce good esul s? PRC’s expe ience
o e s h ee speci ic lessons, bu hey a e be e iewed as gene al p inciples han a o mula o
ansi ion.
Fi s , ins i u ional changes ha c ea e incen i es, impose ha d budge cons ain s, and
in oduce compe i ion should no only apply o i ms bu also o go e nmen s. Indeed,
e o ming go e nmen is an impo an componen o economic e o m. When he go e nmen
is p o ided wi h posi i e incen i es and is subjec ed o ha d budge cons ain s and compe i ion
(such as he ownship and illage go e nmen ), he ou come o he e o m is p oduc i e; i he
go e nmen is subjec ed o so e budge cons ain s (such as highe le el local go e nmen s
we e in he 1980s), he ou come will be p oblema ic.
Second, success ul e o m elies on poli ical suppo , which in u n depends on deli e ing
angible bene i s o a la ge majo i y o he popula ion. Economis s usually blame “s upid
poli ics” when hei beau i ully designed e o m p og ams a e no implemen ed. Bu poli ical
scien is Ba y Weingas has his o say: “An i onic aspec o he economis s’ posi ion is ha
hey wan indi iduals o pu sue sel -in e es in ma ke s bu no in poli ics.” Compensa ing

31
po en ial lose s in he e o m is bo h a poli ical and economic issue, and a e o m ha does no
c ea e many o big lose s can be poli ically accep able ex an e and sus ainable ex pos . The
dual- ack app oach o ma ke libe aliza ion, when app op ia ely implemen ed, is one example
o good poli ics and economics.
Thi d, success ul ins i u ional change equi es app op ia e, bu no necessa ily op imal,
sequencing. Unlike mac oeconomic s abiliza ion policy, ins i u ional change is an una oidably
long p ocess. Judging om he ou come, he p ocess o PRC’s ansi ion o ma ke s has been
ai ly as o e all. I ook Hunga y 30 yea s and Poland 20 yea s o achie e hei cu en esul s
and will ce ainly ake Russia e en longe . The PRC expe ience shows ha whene e
poli ically easible, i is be e o disman le he exis ing ins i u ions a e he new ones a e pu in
place, o allow he new ones o eme ge om he old, o a oid an ins i u ional acuum. I also
shows ha he implemen a ion o e o m can be compa ible wi h he ins i u ional capaci y o
he economy. Examples o app op ia e e o m sequencing in PRC include: empowe ing local
go e nmen s o de elopmen o local go e nmen en e p ises be o e p i a iza ion can be
ins umen al o g ow h; pu suing indus ial libe aliza ion i s while delaying inancial
libe aliza ion can compensa e o iscal decline, a oid inancial c ashes, and, a he same ime,
c ea e be e incen i es; en o cing a plan ack and phasing i ou la e can minimize he
opposi ion o e o m while i imp o es economic e iciency.
On he o he hand, he PRC expe ience shows ha one does no need op imal sequencing
o achie e an imp essi e pe o mance, and wha is mos impo an is o a oid a al mis akes
a he han o make ine uning. Indeed, many a gumen s can be made ha PRC’s e o m is no
op imal. Bo h in e ms o e o m sequencing and o any speci ic e o m, a guably, PRC could
ha e done a be e job and achie ed e en g ea e esul s i be e al e na i e e o ms had been
done. The e o s made wi h he “c ossing he i e by ouching he s ones” s yle can be de ec ed.
Fo example, as I ha e shown abo e, PRC’s e o m ell sho in he a eas o e o ming he
inancial sys em and la ge s a e-owned en e p ise and es ablishing o he ule o law. Howe e ,
o his da e, hese e o s ha e no unde mined PRC’s e o m in a a al way.
C. Re lec ions on he Theo y and P ac ice o T ansi ion
T ansi ion om a planned o a ma ke economy is one o he mos signi ican economic e en s
o he wen ie h cen u y. The co e o he ansi ion conce ns ins i u ional ans o ma ion, which
is complica ed and di icul . Fo mos ansi ion coun ies, wi h he excep ion o a ew,
ansi ion will ake mo e ime o comple e. Du ing he pas decade he e we e wo big su p ises.
One big su p ise was he sha p ini ial decline in ou pu ollowed by eco e y in mos coun ies
o Eas e n Eu ope and he o me So ie Union — he phenomenon known as he U-shaped
ou pu esponse (Blancha d, 1997). Ano he big su p ise was PRC’s ema kable success on i s
di e en pa h o ansi ion.
The gap be ween con en ional economic hinking and he eali ies o he ansi ion shows
ha ou knowledge abou ins i u ional change in gene al and ansi ion in pa icula is qui e
limi ed. This limi a ion has wo ela ed componen s. Fi s , ou knowledge abou how a
capi alis ma ke economy wo ks is s ill limi ed. Al hough neoclassical economics is a powe ul
ool o explaining how p ice mechanisms wo k, i does no p o ide comple e heo ies abou
he ole o ins i u ions, his o y, and poli ical economy ye . Ronald Coase (1992), e e ing o
ansi ion economies, says, “wi hou he app op ia e ins i u ions no ma ke economy o any
signi icance is possible. I we knew mo e abou ou own economy, we would be in a be e
posi ion o ad ise hem.” Un o una ely, we do no . Douglass No h (1997) ag eed: “While neo-
classical heo y is ocused on he ope a ion o e icien ac o and p oduc ma ke s, ew wes e n
32
economis s unde s and he ins i u ional equi emen s essen ial o he c ea ion o such ma ke s
since hey simply ake hem o g an ed. A se o poli ical and economic ins i u ions ha
p o ides low-cos ansac ing and c edible commi men makes possible he e icien ac o and
p oduc ma ke s unde lying economic g ow h.” To build a ma ke sys em om sc a ch, no hing
can be aken o g an ed.
Second, ou knowledge abou he ansi ion p ocess is e en mo e limi ed. The e is a
di e ence be ween he inal des ina ion and he p ocess o ansi ion, and e en i we ha e
pe ec knowledge abou he des ina ion, how o ge he e is a sepa a e issue. As Robe Solow
said, “The e is no some glo ious heo e ical syn hesis o capi alism ha you can w i e down in
a book and ollow. You ha e o g ope you way” (The New Yo k Times, Sep embe 29, 1992, p.
E1). Richa d F eeman (1994) echoed ha iew, “[E]conomics does no ha e su icien ly
compelling heo y o empi ical knowledge o answe ques ions abou he ins i u ional design o
ad anced capi alis economies, much less economies in ansi ion.”
Because he ansi ion om plan o ma ke is unp eceden ed, he p ac ice o ansi ion is
o iginal and he heo y o ansi ion is in i s in ancy. Two excellen su ey pape s on he
economics o ansi ion (Dewa ipon and Roland, 1996; and McMillan, 1996) ecognized he
limi s o ou knowledge on ansi ion. Mo e and mo e people, bo h inside and ou side he
ansi ion economies, ha e ealized his oo. E en Ana oly Chubais, one o he leading
a chi ec s o he Russian e o m, acknowledged a e he Russian c isis in 1998: “We mus
ecognize ha we did no ully unde s and he scale o he p ocess which we had unde aken.
We hough he e would be a e y di icul h ee yea s, i e yea s, eigh yea s. Now,
un o una ely, i is clea ha e o m will ake decades” (Reu e s, Sep embe 24, 1998).
PRC’s e o m pa h was no designed a he ou se , bu e ol ed. Ra he han being nai e
o cynical abou e o m, Deng Xiaoping and he PRC e o me s we e p agma ic, ecognizing
om he beginning ha hey did no know wha would make he e o ms succeed. S udying
PRC’s expe ience should add a g ea deal o economics as a social science o human beings.
A e all, his is a coun y ha has nea ly one- i h o he wo ld’s popula ion and he possibili y
o becoming, once again, he la ges economy in he wo ld in less han wo decades (Maddison,
1998).
33
Re e ences
A S a is ical Su ey o China. Beijing: China S a is ical Publishing House, a ious yea s.
Almanac o China’s Finance and Banking. Beijing: China Finance Publishing House, a ious
yea s.
Almanac o China’s Popula ion. Beijing: China Popula ion Publishing House, a ious yea s.
Bai, Chong-En, Da id D. Li, and Yingyi Qian, Yijiang Wang. “Anonymous Banking and
Financial Rep ession: How Does China’s Re o m Limi he Go e nmen ’s P eda ion
wi hou Reducing I s Re enue?” Mimeo, S an o d Uni e si y, 1999.
Blancha d, Oli ie . Economics o Pos -Communism T ansi ion, Ox o d Uni e si y P ess, 1997.
Blancha d, Oli ie ; and S anley Fische . Edi o ial, NBER Mac oeconomics Annual 1993,
Na ional Bu eau o Economic Resea ch, 1993.
Blancha d, Oli ie , and Michael K eme . “Diso ganiza ion,” Qua e ly Jou nal o Economics,
No embe 1997, 112(4), pp. 1091-1126.
By d, William. “En ep eneu ship, Capi al, and Owne ship,” in William By d and Qingsong
Lin (eds.), China’s Ru al Indus y: S uc u e, De elopmen , and Re o m, Ox o d Uni e si y
P ess, 1990.
By d, William. The Ma ke Mechanism and Economic Re o ms in China. New Yo k: M.E.
Sha pe, 1991.
By d, William; and Alan Gelb. “Why Indus ialize? The Incen i es o Ru al Communi y
Go e nmen s,” in William By d and Qingsong Lin (eds.), China’s Ru al Indus y:
S uc u e, De elopmen , and Re o m, Ox o d Uni e si y P ess, 1990.
Cao, Yuanzheng; Yingyi Qian; and Ba y R. Weingas . “F om Fede alism, Chinese S yle, o
P i a iza ion, Chinese S yle.” Economics o T ansi ion, Ma ch 1999, 7(1), pp. 103-131.
Chang, Chun; and Yijiang Wang. “The Na u e o he Township En e p ises.” Jou nal o
Compa a i e Economics, 1994, 19, pp. 434-452.
Che, Jiahua. “A Theo y o O ganized Ma ke T ansac ions.” Mimeo, Uni e si y o No e Dame,
1998.
Che, Jiahua; and Yingyi Qian. “Ins i u ional En i onmen , Communi y Go e nmen , and
Co po a e Go e nance: Unde s anding China’s Township-Village En e p ises.” Jou nal o
Law, Economics, and O ganiza ion, Ap il 1998a, 14(1), pp. 1-23.
34
Che, Jiahua; and Yingyi Qian. “Insecu e P ope y Righ s and Go e nmen Owne ship o
Fi ms.” Qua e ly Jou nal o Economics, May 1998b, 113(2), pp. 467-496.
China S a is ical Yea book. Beijing: China S a is ical Publishing House, a ious yea s.
China Township En e p ises S a is ical Yea book. Beijing: China S a is ical Publishing House,
1994.
Chow, G ego y. “Challenges o China’s Economic Sys em o Economic Theo y.” Ame ican
Economic Re iew, May 1997, 87(2), pp. 321-327.
Coase, Ronald. “The Ins i u ional S uc u e o P oduc ion.” Ame ican Economic Re iew,
Sep embe 1992, 82, pp. 713-719.
Dewa ipon , Ma hias; and E ic Maskin. “C edi and E iciency in Cen alized and
Decen alized Economies.” Re iew o Economic S udies, 1995.
Dewa ipon , Ma hias; and Gé a d Roland. “Economic Re o m and Dynamic Poli ical
Cons ain s,” Re iew o Economic S udies, 1992, 59, pp. 703-730.
Dewa ipon , Ma hias; and Gé a d Roland. “The Design o Re o m Packages unde
Unce ain y,” Ame ican Economic Re iew. Decembe 1995, 83(5), pp. 107-123.
Dewa ipon , Ma hias; and Gé a d Roland. “T ansi ion as a P ocess o La ge Scale Ins i u ional
Change,” in Da id K eps and Kenne h Wallis (eds.), Ad ances in Economics and
Econome ics: Theo y and Applica ions, Camb idge: Camb idge Uni e si y P ess, 1996.
Dong, Fu en. “The ‘Budge Law’ and Ha dening Go e nmen s’ Budge Cons ain s,” in
Tianqing Xu and Jinyan Li (eds.), China’s Tax Re o m, Beijing: China Economics
Publishing House, 1997.
F ye, Timo hy; and And ei Shlei e . “The In isible Hand and he G abbing Hand.” Ame ican
Economic Re iew, May 1997, 87, pp. 354-358.
G o es, Theodo e; Yongmiao Hong; John McMillan; and Ba y Naugh on. “Au onomy and
Incen i es in Chinese S a e En e p ises.” Qua e ly Jou nal o Economics, 1994, 109(1).
Ha , Oli e . Fi ms, Con ac s, and Financial S uc u e. Ox o d Uni e si y P ess, 1996.
on Hayek, F ied ich A. “The Use o Knowledge in Socie y.” Ame ican Economic Re iew,
1945, 35, pp. 519-30.
Hellman, Joel S. “Winne s Take All: The Poli ics o Pa ial Re o m in Pos communis
T ansi ions.” Wo ld Poli ics, Janua y 1998, 50, pp. 203-34.
Hoshi, Takeo; and Anil, Kashyap. “The Japanese Banking C isis: Whe e Did I Come F om
and How Will I End?” NBER Mac oeconomics Annual 1999, o hcoming.