China: Developmental State Or Dysfunctional Development?

ThirdWorld Quarterly, Vol. 17, No. 4, pp. 689± 706, 1996

China:developmental state or dysfunctionaldevelopment?

SHAUNGBRESLIN

Since1978 the Chinese Communist Party ( CCP)leadershiphas implementeda series ofreforms designedto restore the party’ s legitimacyby moving from a politicallyto a moreeconomically mobilised society. This strategy has entailed thegradual dismantling of the instruments of state planning, and their replace- mentwith market mechanisms, in anattempt to increaseeconomic ef® ciencyand productivity.With China’ s internalMaoist model of development and other socialistdevelopmental models now essentially moribund, some withinChina, andeven among the CCP itself,have looked East for inspirationÐ or for aspirations.Where reform in much of Eastern Europe had resulted in a decline inparty authority and the descent into economic chaos, China’ s EastAsian neighboursappeared to provideshining examples of howrapid economic growth couldbe compatible with continued authoritarian government. Whileit is truethat the Chinese reform experiment does contain some similaritieswith the capitalist developmental states ofEast Asia, we should not concludefrom this that China is followinga coherentstrategy of emulating the NICs. China’s developmentaltrajectory has beento a largeextent dysfunctional. Thereare threekey factors here which this paper will explore in detail.The ® rst is theway that political demandsÐ both from within the party± state structure itselfand from other societal groupsÐ have obstructed the formulation of a coherentand effective national economic development strategy. In the words of thetitle of Susan Shirk’ s excellentinterpretation of reform in China, we must assess `Thepolitical logic of economic reform’ . 1 Second,the reform process has resultedin a signi®cant redistribution of economicdecision-making power, primarily from central to local level authorities.The central authorities have granted local governments across China varyingdegrees of ®nancialautonomy. In addition, some haveexploited growing internationalcontacts to ® ndfurther sources ofextra-budgetary investment ®nance,or pressured theoretically (quasi) independent banks and enterprises to supportlocal development plans. As aresult,China’ s developmentaltrajectory owesat least as muchto the dysfunctional agglomeration of numerous local initiatives,as itdoes to the plans and strategies of the national level decision- makingelites. Thethird factor is theextent to whichChina’ s externaleconomic interests are nowin¯ uencing developmental strategies and processes. As localauthorities competewith each other (not to mentionother developmental states inEastAsia

ShaunG Breslinis atthe Department of Politics,University ofNewcastle uponTyne, Newcastle NE17RU, UK.

0143-6597/96/040689-18$6.00 Ó 1996Third World Quarterly 689 SHAUN GBRESLIN andelsewhere) for investment, there has beena tendencyto emphasise low-cost andlow-regulation advantages to attract labour intensive component assembly investments.As employmentand growth in some areas becomeincreasingly contingenton attracting and retaining such foreign investment, some local authoritieshave abandoned internally de® ned development strategies and pay moreheed to the requirements of Asian investors. While this investment has broughttangible bene® ts tothe Chinese economy, the long-term wisdom of this strategyis morequestionable. At the very least, the dynamic relationship betweenthe local and the international adds a furtherquestion mark over the centralelites’ ability to adopt and enforce coherent national development programmes. Thesethree factors are allinextricably linked. Indeed, there is almostan evolutionaryprocess here.The political constraints on economic strategy in the early1980s in large part contributed to the dysfunctional growth of local economicautonomy in the second half of the decade. This localism not only exacerbatedthe problem of political demands in¯ uencing economic strategy, but has alsoin turnbeen a majorfactor in explaining the importance of international economiccontacts in the 1990s. In short, inter-elite discord generated dysfunctionaldevelopment, which in turn generated further unexpected developments whichhave militated against the adoption of any coherent programme or model forreformÐ either internal or external.

TheChinese developmentalstate: economicsor politics in command? Inrejecting the Maoist model of development(and Hua Guofeng’ s variationon thetheme between 1976 and 1978), the Chinese leadership initially looked backwardsfor their models to theheyday of Leninist` orthodoxy’. The8th Party Congressof 1956 was takenas agoldenera ofcollective leadership pursuing pragmaticeconomic development strategies which placed economic develop- mentabove class con¯ict, and even allowed for a role(albeit a veryminor role) formarket forces inthe countryside. But1956 was obviouslynot 1978. On onelevel, the economic system thathad helpedgenerate rapid growth before 1956 had lost much of its appeal by 1978. Bothwithin China and in much of therest ofthesocialist world, state planning hadbecome more associated with stagnation than with growth. On anotherlevel, therewere considerable disagreements over how far the CCP couldor should go inabandoningthe past. All groups were ® rm intheir commitment to shoringup the CCP’sgripon power, but there was considerabledisagreement over the practicalitiesof rejecting the past, and on what point an increased role for the marketbecame incompatible with socialist principles. During1984 and 1985 consensus within the elites over the best development strategyhad become severely strained. The success oflimitedmarketisation and increasedinternational contacts through the Special Economic Zones ( SEZs) impressedkey leaders such as thethen Premier, Zhao Ziyang. For these more reformminded leaders, further price reform, economic liberalisation and interna- tionalisationwere the logical way forward. However, others argued that reform hadalready gone too far. In the Autumn of 1985,Chen Yun, one of theoriginal 690 CHINA: DYSFUNCTIONAL DEVELOPMENT? architectsof the post-Mao reforms, issued a strongpublic warning on the dangersof over-rapid development, neglect of state planning and grain production,and questioned the personal and socialist ethics of some partyof® cials. 2 Zhaowas unpreparedor unable(or both) to pushhis case intheface offairly widespreadand bitter opposition, particularly as heappears to have failed to persuadeDeng Xiaoping to supporthim. Zhao opted instead for a moremoderate andacceptable package of reforms, which included transferring power away fromthe more sceptically inclined central planning and ® nancialinstitutions to theprovinces. Despiteabandoning his formal positions of power,Deng Xiaoping has clearly remainedthe single most dominant ® gurein Chinese politics. Like Deng, Chen Yunand other members ofthe Long March revolutionary generation were also ableto exert considerable political in¯ uence from behind the scenes. This confusingdistribution of power has considerablyimpinged on the freedom of movementof formalpower-holders such as HuYaobang,Zhao Ziyang and Jiang Zemin.3 Theynot only have to consider how the party elders will respond to theirinitiatives, but how Deng Xiaoping will respond to anycriticism from these partyelders. Deng’ s skillat factional management has beena sourceof considerablesuccessÐ at least by his own criteria. He has keptthe party in powerand more or less togetherdespite considerable strains from both within theparty and without. Nevertheless, his political balancing game has impinged onthe smooth evolution of coherent and economically logical policies. If oppositionthreatens to disrupt unity and thus jeopardise the party’ s gripon power,Deng has beenquick to react and abandon (more often postpone until morefavourable times) offensive policies. Furthermore, Deng’ s patronage helpedboth Hu Yaobang and Zhao Ziyang assume theposition of party leader inthe 1980s, but both men also lost this position when Deng sacri® cedtheir politicalcareers forthe sake ofunity. SusanShirk has explainedhow the priorities of maintaining a semblanceof internalparty cohesion and unity has becomesomething of a de®ning character- isticof how the CCP has managedthe process ofreform. To prevent strong oppositionto reform, an ethos of fairness came todominate policy making, or whatShirk calls an ideology of `balancism’. 4 Inpractice, this meant that central decisionmakers chosepolicy options which ensured that no unit or actorlost too much,rather than those which were best for the country as awhole.In short, ratherthan choosing the best policy, decision makers chosethe satis® cing policyÐthe one that did just enough to satisfy and suf® ce. 5 Theimplications of thisapproach to policymaking are enormous.First, there is atendencyto make small incremental changes that are easier toget past obstructionswithin the policy-making elite than coherent and comprehensive reforms.It also means thatyou can back-track on speci® c policyif necessary withoutjeopardising the entire reform process. Second, Shirk notes that there was noreal coherence between reforms indifferent areas. Withindividual agenciesleft to develop their own reforms, the result of this fragmented decision-makingprocess was anumberof unconnected,incoherent, and at times contradictory,reforms. 6 Third,it has alsocreated something akin to a boomand bust cycleÐ or more 691 SHAUN GBRESLIN correctlyan expand and retrench cycle. You Ji suggeststhat Zhao Ziyang advocateda rapidpace of developmentto prove the wisdom of his policies and togain popular support for reform even though he knew that this policy would generatein¯ ation and be unsustainablein thelong run. 7 The¯ ipside of thiscoin is thatwhen even relatively minor problems emerge, opponents seize onthe opportunityto forcea rethinkof policy,usually resulting in retrenchment.In his analysisof industrialreform in China, Naughton paints the following picture of thegovernment’ s approachto reform: `Unableto dominate events, the govern- menthas hadto scramble repeatedly to ª putout ® resºand prevent disastrous outcomes’ .8 As aresultof this preoccupation with short-term problems and solutions,the CCP eliteswere unable to ` formulatea consistentprogram of economicchanges’ . Eventhough Naughton was speci®cally commenting on the periodup to1983,this analysis remains valid for much of the subsequent period.

Thecapitalist developmental state and ` relativestate autonomy’ Inhis analysis of the capitalist developmental states inEast Asia, Chalmers Johnsonargues that one of thekey, perhaps de® ning, features of theEast Asian NICsis `stablerule by a political±bureaucratic elite not acceding to political demandsthat would undermine economic growth’ . Aspoliticiansare motivated byseeking popular support in the short run, long-term economic strategy must, asfar aspossible,be insulatedfrom the vagaries of competingpolitical demands: Politicalleaders attempting to implement long term industrial policy must therefore havethe capacity to depoliticize in part their key economic decisions. This is normallydone by entrusting such decisions to a ªnonpoliticalelite,º sheltered to somedegree from directpolitical pressures. 9 Notwithstandingthe lack of substantialcompetitive elections, political demands havebeen a majorfactor in the Chinese economic decision-making process. Indeed,given the nature of the Chinese political system (notleast the fact that theentire central decision-making apparatus is staffedby party members), economicdecision making is perhapsmore responsive to politicaldemands from withinthan it is subjectto political demands from without. Party rule in China is theoreticallya means toan endÐ itis thevehicle for representing the wishes ofthe proletariat and establishing a communistsociety. In reality, the Chinese politicalsystem isnota `dictatorshipof theproletariat’ but a `dictatorshipof the party±state’ . Itistruethat the party devotes considerable time, effort (and more concretelymoney) to defending the interests of certain sections of the Chinese population,but the motivation behind these policies is toshore up the ruling elites’grip on powerÐ maintaining party rule has becomean end in itself. Reformistleaders thus faced the dilemma of squaring long-term economic requirementswith short-term political considerations. At therisk of over-simpli- fyingthe issues, wecanmake the following seven statements characterising the developmentalprocess inpost-Mao China which distinguish the Chinese case fromJohnson’ s capitalistdevelopmental state model: 1Economicreform was promotedto protect the political priorities of the Chineseruling elite. 692 CHINA: DYSFUNCTIONAL DEVELOPMENT?

2Therewas alackof knowledge and experience in utilising and controlling marketmechanisms. 3Therewere divergent, and at times bitter, con¯ icts within elites over the speci®cs ofreform policies. 4Therewere divergent, and at times bitter, con¯ icts within elites over the ultimatedestination of the reform process. 5Reformpolicies were de® ned to ensure that no groupwithin the elites lost too much,thus jeopardising political stability. 6Reformpolicies were adapted to ensure that no geographicarea losttoo much, thusjeopardising political stability. 7Reformpolicies were adapted to ensure that key societal groups (notably the ruralgrain producers and, more importantly the state employed working class) didnot lose too much, thus jeopardising political stability.

Theabove mentioned debate over reform in 1984 and 1985 provides a particu- larlygood example of how some ofthese forces havein¯ uenced Chinese development.Another good example is thefailure (over a longperiod) to fully reformthe pricing structure for key commoditiesÐ notably grain and energy inputs.State control in theseareas givesthe central elites the ability to retain at leasta semblanceof their ability to divert resources tokey areas andmaintain consumptionand production in areas whichmight fare badlyunder market competitionfor resources. It also re¯ ects theparty’ s uncertaintythat it can survivethe (at least) short-term social upheaval from the in¯ ation that price reformwould bring. For example, with in¯ ation rising in 1988, the then party leader,Zhao Ziyang, once again argued that comprehensive price reform would removethe structural causes ofin¯ ation and was thelogical way forward. However,abandoning low state-set prices would lead to considerableprice rises forkey goods (including food staples). For Premier Li Pengand others, growing socialdiscontent had to bedealtwith immediately, not in thelong term; instead theydecided to re-impose planning and price controlsÐ to treat the symptom ratherthan the disease itself. A®nalexample here is thedesire to maintain employment and urban living standards.Rising unemployment is abigenough problem in any society, but has particularsigni® cance in China. The work unit or the collective provides those welfareservices normallyprovided by the state (or the market) in the West. If youlose your job, you could also lose your accommodation, your health care and yourchildren’ s educationÐand all of thiswithout any unemployment bene® t to fallback on. This` negative’impact of the drive for economic ef® ciency® rst became apparentin the rural sector. The de-collectivisation of the Chinese countryside mayhave increased agricultural production, but at the cost of the declining provisionof health, education, welfare and employment. Social cohesion in some ruralareas is nowat breaking point, and the simultaneous rise inrural economicdislocation with the growth of gangsterism and the extension of triad activityinto the mainland is nomere coincidence.The CCP eliteshave primarily usedeconomic readjustment as ameans ofmaintaining social cohesion in the countryside.This is mostclearly seen inreadjustments to the purchasing price 693 SHAUN GBRESLIN forgrain, and the maintenance of comprehensive purchasing systems foragricul- turalproduce. For example, after the massive rapeharvest in 1996, the state agreedto purchasethe vast (and unusable) surplus. In an earlier ` rapecrisis’ the stateactually responded by aconcertedcampaign to persuadethe population that eatingrape was `patriotic’! Notwithstandingthe defence of rural prices and the fact that the rural populationvastly outnumbers the urban community, the CCP seems tohavedone muchmore to protect employment and conditions in the towns than in the countryside.The primary factor here is the CCP’sperceptionof the relevant importanceof different social groups. Rightly or wrongly, it perceived that disaffectionamong the urban working classes was agreaterthreat to the party thanthe more fragmented consequences of reform in the countryside. Theparty has goneto great expense to inoculate important sectors ofthe urbanpopulation against potential harmful side effects ofthe market. For example,the aforementioned semi-reform of pricing and allocation structure helpsensure an adequate and relatively equable supply of grain to all urban centres.The central state also pays a numberof othersubsidies and bonuses that allserve tomaintain urban purchasing power and to diminish urban income differentialsbetween different sectors andregions. Furthermore, notions of socialcohesion rather the logic of economic rationality provides the basis of employmentpolicy in manystate owned industries. Although some stateowned enterpriseshave gone to thewall, the state shores upthemajority of loss making enterprises(at least a thirdof the total by the CCP’sownreckoning), while over-staf®ng undermines the pro® tability of others. InDelfs’ words, ` thecitieshave been bought off’ . 10 Perhapswe shouldclarify thisstatement by distinguishing between different elements of the Chinese workingclass. Thereare anumberof sub-groups within the urban industrial sector,and the divisions correlate strongly with the non-wage privileges that accompanyoccupation in sectors distinguishedby type of ownership. The key grouphere, at least for the party leadership, are thoseemployed in the state sector.China’ s pre-reformsystem createda privilegedworking class withinthe urbanworking class, andit is thisgroup or class thatthe CCP has beenprimarily concernedin defending. Thisapproach to managing reform has maintainedsocial and economic stabilityand has alsohelped alleviate some ofthespatial problems of economic reform.But short-term gains may be costlyin the long run. For example, China isindangerof evolvinga `dualistic’economy. While the CCP has encouragedthe developmentof newcompetitive and ef® cienteconomic sectors, it has insulated vulnerablesectors ofthe economy from potentially fatal market competition. Thetwo sectors are nowso isolatedfrom each other that the ¯ ourishing cooperative,private and (particularly) foreign-owned sectors frequentlycannot use productsfrom the state sector. Components from the latter are oftoo low qualityand supplies too unreliable for other more vibrant and market rational sectors ofthe economy. Moreimportantly for this analysis of China’ s dysfunctionaldevelopment is theway the central elites become vulnerable to pleading for special treatment. Facedwith sometimes con¯ icting demands upon them, the central authorities 694 CHINA: DYSFUNCTIONAL DEVELOPMENT? frequentlyrespond by buying themselves out of problems.In some cases, itcan endup responding to one concession by paying for another. For example, when peasantdiscontent has forcedan increase in grain procurement prices, the state has frequentlysubsequently increased bonuses and subsidies to the urban sector tooffset the increase in grain prices. As thePolish government found in the 1980s,regimes of this sort can ® ndthemselves sucked ever deeper into a cycle ofspiralling subsidy payments. In the Chinese case, arounda quarterof the nationalbudget is devotedto subsidy payments of one kind or anotherÐ almost as muchas thecentre spends on development projects. Themore the centre accedes todemandsfor special treatment, the more it may becomevulnerable to new demands upon it. In the framework of Johnson’ s earlierwork on revolutionary potential, it risks becomingunable to cope with the paceand pressure forchange resulting in ` powerde¯ ation’ or a progressive reductionin system effectiveness. 11 Politicallobbying (notably, but not solely, fromprovincial and local interests) has beenan important component of the policy-makingprocess. Although competition between regions for preferential centraltreatment has alwaysoccurred in the PRC, the higher potential gains and losses thatmarket reforms unleashedenhanced this competition. An excellent examplehere is theway in which the Shanghai authorities successfully chal- lengedthe basis oftheir® nancialrelationship with the centre in response to the preferentialtreatment given to Guangdong and Jiangsu. 12 Thiscase is importantfor three main reasons. First, it is agoodexample of howpolitical pressures underminedthe logic behind a consciousstrategy. Shanghai’s gainsin terms of® nancialautonomy must be at somebody else’ s expense.Either the centre loses morein terms ofthe® nancialbalance of power betweencentre and province, or other provinces lose out from the removal of the extrafunds from the national budget. Second, Wang Huning notes that it re¯ ects anewprovincial con® dence that localities can exert pressure onthe central policy-makingprocess:

{Shanghai’s} actionin taking the initiative to ask for anewsystem to be put in practicediffered from theold model for reform inwhich the central government tookthe initiative in these matters. 13

Third,it had a knock-oneffect in thatit generated a waveof lobbyingfrom other provinceswith grievances about current policy. Immediately following Shang- hai’s successful approachto the centre, Inner Mongolia and Jilin formally asked (withoutsuccess) tomove to the Shanghai system. Hunan and Jiangsu of® cials similarlyasked for some ofthe preferential treatment afforded to Guangdong, andwe canonly guess atthe extent of theinformal lobbying that also occurred. A®nalproblem with this strategy is itsexpense. While the Polish regime in the1980s funded its subsidy payments through reliance on external debts, the Chineseregime has primarilyrelied on funding through budget de® cits. Econ- omicreform has generateda signi®cant growth in national wealth, but not enoughof thismoney has foundits way to the centre for it to meet its growing commitments.The key issue hereis thechanging relationship between centre andprovinces, which has notonly had a massive impacton theevolution of the 695 SHAUN GBRESLIN

Chinesereform process todate,but also remains a keydeterminant of thefuture trajectoryof the Chinese political economy.

Dysfunctionaldevelopment: the growthof provincial autonomy TheChinese reform process has entailedthe deliberate devolution of a signi®cant degree of economiccontrol to the provinces. As thiswas aconscious strategy,it may seem strangeto talk of the growth of provincialism as a dysfunctionalconsequence of reform. Indeed, responding to central complaints atthe loss of® nancialcontrol over the provinces, a 1990World Bank report arguedthat the loss ofcentral control in the 1980s was: amatterof deliberate choice and not inadvertent. The Centre cannot, therefore, chafeover the loss of ® scalresources. 14 However,this analysis somewhat misses thepoint. On one level, reformers adopteddecentralisation policies with little forethought or research onwhat this wouldactually mean. You Ji arguesthat the decentralisation policies of 1984 werein part designed to undermine the in¯ uence of conservative leaders whowere heavily (and disproportionately) represented in central ® nancialand planningorgans. 15 As such,the 1984 decentralisation was ameans toan end ratherthan an end in itselfÐ it is anexample of how an acceptable policy tookprecedence over what Zhao Ziyang thought was thebest and most logical policy. Onanother level, while the centre did indeed initiate the decentralisation process,provincial and lower level authorities did not use theirnew powers in themanner that central leaders envisaged. This was partlya consequenceof a lackof co-ordination between reforms indifferent areas, whichalmost forced provincesto adopt ¯ exibleresponses tonumerous and often con¯ icting new centralinitiatives. For example, changes in the® scal relationshipbetween centre andprovince were rapidly followed by the introduction of a newsystem which taxedenterprise pro® ts. As the® scal reforms werebased on the old system of allenterprise earnings being collected at the provincial level, many provinces nowfaced substantial reductions in revenuewithout any corresponding decrease intheir income collection targets. 16 Indeed,with revenue collection rates set to rise annually,many provinces decided to make short-term adjustments to their economicstructure. A popularresponse was tobuild small factories producing consumergoods with high prices that would guarantee a fast returnon investment. Onceestablished, these new factories needed considerable administrative supportto continue producing. Energy and raw material supplies were provided irrespectiveof thecomparative ef® ciencyof thenew factories, often by diverting suppliesout of the planned sector. In order to guarantee a marketfor their produce(and thus guarantee provincial income through taxing pro® ts), localities erectedlocal tariffs and quotas to keep out imports from other provinces. Thus, themore ef® cientfound their access toscarce commoditiesand markets squeezedwhile the inef® cient continued producing thanks to theirhand in glove relationshipwith local authorities. As aresult,the growth of provincial power 696 CHINA: DYSFUNCTIONAL DEVELOPMENT? has notonly generated signi® cant centre± province tension, but also a growthin province±province con¯ ict over access toraw materials. Butthe provinces were not just unwilling or reluctant players in the game. A furtherconsideration is theway in which local authorities distorted central reforms fortheir own end. One of the clearest examples here, and perhaps the beststarting point for this analysis, is thedysfunctional consequences of the managerialresponsibility reforms. In 1984, the CCP tookthe ® rst steps in separatingenterprise management from state control by extending reform into theindustrial sector. While these reforms didindeed extend management autonomy,state intervention (and even direct management at times) remains a factof life for most enterprises. Notwithstanding formal structural reforms, manyprovincial cadres’ haveproved reluctant to relinquish their control over localeconomic affairs. While it is relativelyeasy tomake legal± structural changes,changing cultures can be a muchharder and longer process. Many cadres andmanagers have simply not internalised the new ` politicalculture’ , andcontinue to relate to each other as theydid in the pre-reform era. Lack of reformto political structures also strengthens such ` organisationalimprint- ing’ .17 Wherechanges in the economy demanded administrative changes, reformers simplyadded new structures to existing ones. Where old and familiar structuresremain in place, it is notsurprising that they often take precedence overnew and strange ones. Asecondfactor was theway the semi-reformed pricing and distribution structureforced managers into a positionof dependence on provincial adminis- trativecontrol. To put it simply, reforms inother areas didnot support the logicof managerial autonomy. For example, if enterprises can’ t affordto pay highermarket prices for key industrial inputs (notably coal and other energy sources),they are essentiallydependent on administrative interference to guaranteeenergy supplies. Furthermore, with much of thetransportation system overloadedby coal shipments, it is dif®cult to move other commodities, even overrelatively short distances, without the active support of `sympathetic’local authorities. Finally,we must consider the retention of party control over important personneldecisions through the nomenklatura system,which Burns contends is themost signi® cant instrument of party control throughout the country. 18 While thisis true,it is importantto distinguish between party control and central control.From 1984, the central authorities gave up their right to vet appoint- mentsat sub-provincialleadership levels and devolved their vetting authority to provinciallevel party committees. Although this may have only made de jure whathad become de facto,19 itnevertheless represented a formalreduction in thecentre’ s powerat the provinces’ gain. It also meant that if local authorities impingedon the new rights of enterprise managers, the threat of dismissal was apowerfulcause ofmanagement acquiescence. Ratherthan granting autonomy to enterprises, these reforms insteadredis- tributedpower among bureaucratic agencies. The vertical dictatorship ( tiaotiao zhuanzheng)thatcharacterised the system ofstate planning where orders ¯oweddirectly down through ministerial and government agencies has byand largebeen removed. But this has onlybeen achieved at the expense of the 697 SHAUN GBRESLIN growthof horizontal dictatorship ( kuaikuaizhuanzheng ),adictatorshipwhich mayprove more dif® cult to dismantlethan the central state-planned dictatorship itreplaced.

Dysfunctionaldevelopment and provincial ® nancialautonomy Enterprisemanagement autonomy is notthe only areas whereprovincial authoritieshave usurped decentralised control. In 1985, bank loans to local govern- mentsreplaced central government grants as themain source of investment capital.Although banks were meant to use marketbased criteria in assessing the meritsof each loan application, provincial authorities have frequently applied pressure onprovincial bank branches to extend loans to support local economic strategies,even during periods when the central authorities were urging credit tightening.This feature was aresultof the dual control of local level organisa- tionsin China. A provincialbranch of a bankis verticallyresponsible to the bank’s centralof® ces, andultimately to theMinistry of Finance,but at the same time,it is alsoresponsible to the provincial ® nancebureau and the provincial government.The latter has theadvantage of hands-on contact with the branch, sinceit is indirect day-to-day contact with bank of® cials. Furthermore, it possesses considerablepower in terms ofallocatinggoods, services and,through the nomenklatura reforms,leading personnel to the banks. Theprocess ofdevolving power to banks in localities was alsoslightly misguided.In setting the investment quota for provincial specialised banks, the centreannounced that the 1985 investment quota would depend on the amount ofloansextended in 1984. In order to increasetheir quota, many banks decided tolendas muchas possibleduring 1984. The result was amassive expansionin credit,with some banksactively seeking enterprises and local authorities which mightaccept investment loans. Thetransition from central grants to bank credit to provideinvestment capital hada dramaticaffect on the economic structure. At the start of the reform process in1978, state planning agencies and ® nancialauthorities controlled the provisionof 76.6% of internal national investment capital. Within a yearof the introductionof banking reforms, this proportion had fallen to 33.2%. 20 Further- more,increased access todirect foreign investment and loans resulting from China’s openingto the international economy also enhanced the ® nancial autonomyof some provinces.While large foreign investment and loan projects remainedunder central control, provincial authorities and enterprises were free togooverseas and® ndinvestors and partners for smaller projects unhindered by centralcontrols. By the end of 1993, foreign direct investment ( FDI) accounted for12.3% of all investment in China, up from 7.7% in 1992and from less than 1%adecadeearlier. However, it is importantto notethat China’ s openingto the Westdid not bene® t allprovinces equally. For example, over 96% of the foreign exchangeearned through exports from joint ventures went to thecoast, and over athirdof the foreign currency earned from exports from companies utilising foreigninvestment went to the SEZs alone. Otherimaginative ways of raising extra revenue included levying illegal chargeson localhealth facilities, educational support, street cleaning, and so on. 698 CHINA: DYSFUNCTIONAL DEVELOPMENT?

Butperhaps the central element here is thereform of thecentre’ s ®scal relations withthe provinces, introduced between 1976 and 1980. The logistics of the revenue-sharingreform are quitecomplex, not least because different agree- mentswere made with different provinces. At the risk of over-simplifying the issue,we can identify three main departures from the old system whichhave contributedto the growth of provincial ® nancialautonomy. First, with the exceptionof Tianjin and Beijing, 21 a®ve-yearreadjustment cycle replaced annualrevisions to facilitate long-term planning. Second, rather than giving provincescomprehensive and detailed orders on how to spend and collect money,the provinces were granted autonomy to set theirown priorities as long astheystayed within broad expenditure and income guidelines. Third, and most importantlyfor this study, provincial authorities were allowed to retain a proportionof any income collected above the agreed ® gurefor remittances to the centre. Itis dif®cult to overstate the signi® cance of these changes for the dysfunc- tionalgrowth of local power. For example, the centre’ s failureto either step or indexlink provincial remittances proved to be a costlymistake. Arrangements withtwo of China’ s fastest growingprovinces, Guangdong and Fujian, meant thatthe provincial authorities retained all excess revenue.Thus the amazing economicgrowth of the coastal region did not generate a proportionategrowth incentral income. Furthermore, rapid in¯ ation (particularly during 1988) made iteasier forprovinces to meettheir remittance targets, and as in¯ation triggered increases incentral expenditure on compensation payments and subsidies, central® nancessoon came undersevere pressure.The 50% increase in the national® scal de®cit in 1993 (despite an of® cial government pledge to trim the shortfall)provides evidence of the size ofthe problem. Incombination with the banking reforms andnew found access toforeign ®nanceand partners, these revenue-sharing reforms gavemany provinces the wherewithalto advance their own local development priorities. Much Western analysishas focusedon the growing tensions between Guangzhou and Beijing asaresultof Guangdong’s evergrowing economic independence. It is certainly truethat Guangdong’ s localeconomic policies have been the cause ofconsider- ablecon¯ ict with the central authorities, but the growth of provincial power is notsimply con® ned to the newly rich coastal provinces, and the con¯ ict between nationaleconomic goals and provincial economic policies has becomealmost a de®ning characteristic of the Chinese economy in the recent era. Whileold style state planning is moreor less athingof the past, the state neverthelesscontinues to dominate the Chinese economy. On one level, state controlof the unreformed raw materials and infrastructural industries ensures thatthe state still exerts signi® cant control over the national economy. The state alsoremains an important player in the reformed sectors ofthe economy, but herewe must distinguish between the national state and the local state. As the latterremains a componentpart of theformer, it mayseem oddto refer tothem as separateentities. However, the diversity between national and local interests thathas emergedduring the reform process makes thisa validdistinction. Theattainment of national economic goals has beenobstructed by provincial authoritiesacross thecountry establishing and actively pursuing their own 699 SHAUN GBRESLIN economicstrategies. Inter-provincial competition for growth distorted the orig- inalnational strategy of replacing local self-suf® ciencywith a market-rational regionaldivision of labour and comparative advantage. Financial reforms have spawnedan explosion of local capital construction projects that the centre is unableto control. For example, investment in capital construction increased by 35%in the ® rst quarterof 1985,by 37%in April,and by 42%in Mayas adirect resultof banking reforms. Crucially, investment in state-planned projects recordeda mere 1.6%increase, whereas investment in unplanned local projects rose by 87%. Asthe1980s progressed, the central authorities found themselves increasingly unableto pull wayward local authorities back in line. For example, when they ordereda decrease ininvestmentspending in 1986,local authorities still invested atotalof 4.4billion Yuan in the ® rst seventhmonths alone. Similarly, during the of®cial retrenchment campaign of 1989, JingiiRibao (EconomicsDaily) re- portedthat the policy was meetinga `greatdeal of resistance’ in ` afew’ provinces.Many areas wereopenly setting plans for industrial growth above the 4±5% prescribed by the state plan, and an unnamed province in Eastern China (probablyJiangsu) had gone as far astoset aplannedindustrial growth rate of 12±25%.22 Underthe pre-reform system, the centre’ s maintool of control over the provincescame througheconomic mechanismsÐ primarily the restrictions on localautonomy provided by mandatory planning, through control of industrial enterprises,and through control of investment capital. Economic reform saw mandatoryplanning replaced by guidance planning, saw thedevolution of enterprisemanagement away from the centre, and vastly increased locally controlledsources ofinvestment capital. Yet, for much of the 1980s, the centre continuedto act as ifit retained these powerful levers over the provinces. With theultimate sanction of removing the provincial leader a dangerousgame to play,this left the centre with little more than exhortations to unite and comply withcentral priorities, and these pleas often fell on deaf ears. Tomake matters worse, provincial leaders exploited disunity within the centralleadership itself, particularly during the 1989 retrenchment campaign. It was nosecret thatsome centralleaders, notably Zhao Ziyang, disapproved of retrenchmentpolicies, and that the tide might sweep back in favour of the reformers.What was thepoint of making short term cutbacks if the period of retrenchmentwas goingto be short-lived? Indeed, the lesson of the 1980s was thata furtherwave of reform soon followed a periodof retrenchment (usually soonerrather than later). By the end of the decade there was noexpectation or con®dence that the retrenchment policy would last, and that: ` havingexperi- encedshort-lived bouts of contractionover the preceding three years, enterprises attempted,initially, to ` waitout’ credit restraint’ . 23 Thetransition from plan-rational economic control levers to new approaches morein keeping with emerging quasi-market structures is aslowand ongoing process.Furthermore, those mechanisms that might once have served to pullthe provincesback in line now have to operate from the reality of increased provincialautonomy. For example, the introduction of a newincome tax system in1994has gonesome wayto redress thebalance between central and provincial 700 CHINA: DYSFUNCTIONAL DEVELOPMENT? controlof revenues.However, as theprovinces remain the main conduit of ®scal informationto the centre, there is considerablescope for local authorities to exercise` creativeaccounting’ to support local priorities. Surveys from within Chinashow that faking de® cits to avoidtaxation is nowcommonplace (if not the norm).As such,the best efforts of ® nancialsupremo, Zhu Rongji, to create a morerational macroeconomic structure remain in large part contingent on encouragingprovinces (and lower level authorities) to subvert their own devel- opmentalstrategies to national objectives. Thegrowth of provincial autarky is aconsequenceof a numberof factors. First,as notedabove, developing light industrial bases was arelativelyquick wayof ensuring that provinces could meet their ® nancialobligations. Second, infrastructuredefects have obstructed the free ¯owof goodsacross thecountry, andmean that local production can be theonly way of ensuringsupplies. Third, theprovinces, like the centre, are keento maintain employment and social stabilityin urban centres. Fourth,provincial autarky is aresponseto the uneven central treatment of provincesduring the reform process. There is considerableresentment at the specialtreatment granted to Guangdong in particular (but the coastal region in general)to facilitaterapid economic development. Furthermore, despite repeated assurance tothe contrary, many in the interior remain unconvinced of themerits oftrickle-downtheories of development.Indeed, Ji Chongweipaints a pictureof localeconomic strategy built around emulating any successful projectelsewhere:

Wheneverone chunk { ofthe country} starts to produce something, every other chunkwants to get into the act, fearing it willbe left out. Latex medical gloves, for example,have recently been selling well on theAmerican market, so in a ¯ash200 plusproduction lines went up in China. 24 Tomake matters worse, 10 projects were under construction in the USA which wouldprovide all that country’ s domesticneeds on completion. So even those Chinesemanufacturers which had a footholdin the US marketfaced the contraction(and ultimately the removal) of its demand. The desperate bid to makemoney in theshort run undermined long-term prospectsÐ as Ji complains, `thepoorer we are, it seems themore we squander investment’ . Inaddition, Li Xianguoargues that this duplication of regionaleconomic activity ` has destroyed regionalspecialisation and division of labour’ . Anyshort-term gains may have beenmade at the expense of long-term economic ef® ciency,and have also stretchedthe energy, raw material and transport sectors. As such,` theresult of indiscriminatecopying is actuallya non-developmentpolicy’ . 25

Dysfunctionaldevelopment: expanding international economicrelations Themore that the scale of goods and assets produced, exchanged, and/ orused in aparticulareconomic sector or activity diverges from thestructural scale of the nationalstateÐ both from above(the global scale) and from below(the local scale)¼then the more the authority, legitimacy, policymaking capacity, and policy implementingeffectiveness of states will be challenged from bothwithin and without.26 701 SHAUN GBRESLIN

Perhapsthe area whereChina’ s developmentalprocess appearsclosest to the capitalistdevelopmental states is inthe absorption of foreign capital to ® nance economicdevelopment. While the initial intention of the party elites was to concentrateand therefore control international economic contacts, this proved to bean elusive task. Indeed, the development of quasi-autonomous economic policyin Guangdong in particular has beenmuch facilitated by the in¯ ux of foreigncapital and has longsince contributed to thedysfunctional spatial pattern ofdevelopment. These trends are nowbeing intensi® ed in Guangdong and emulatedelsewhere, to theextent that the relationship between the local and the internationalis nowa signi®cant driving force behind China’ s developmental process.It is alsogoing a longway to undermine the central state’ s abilityto control(not to mention dictate) China’ s developmentaltrajectory. Usingforeign capital to aid the developmental process has becomea feature ofdevelopment in East Asia. Indeed, the experience of other East Asian states has clearlyin¯ uenced Chinese decision makers (notleast Deng Xiaoping). As Johnsonnotes: theKoreans and Taiwanese have given virtuoso performances in howto useforeign andmultinational capital without at the same time becoming subservient to it. 27 Bernard’s &Ravenhill’s analysisof East Asian development suggests that Johnsonunderestimated the importance of ` structuraldependence’ on Japanese technologyin Korea and particularly Taiwan. 28 Nevertheless,the ` subservience’ toforeign capital is moremarked in parts of China than at similar stages of developmentin the NICs. Whilemany in the West perceive China as amajorpotential market, other investors,particularly from the rest ofEast Asia, are moreinterested in using Chinaas alow-cost,low-regulation production base forcomponents assembly. Itis thislatter form of investmentwhich dominates in manyparts of China,and is themain focus of attention here. Where investors are keento access the Chinesemarket, the Chinese side have something that the investors want (ie access tothe market) and can negotiate accordingly. But in the case ofthese export-basedinvestments, the Chinese side is competingwith other Asian states (andEuropean regions) for investment, so theforeign side can exert considerable pressure toensure the best possible dealÐ it has somethingthat the competing hosteconomies want. Thisinternational competition is intensi®ed by the competition within China fromdifferent provinces and localities keen to attract new investments. Locali- tiesare notonly striving to offer bigger and better incentives to gain new investments,but also to lure existing investments away from other parts of China.This inter-local competition is manifestin intense competition to gain moreinvestment, irrespective of the quality of that investment. In too many cases, thelocal authorities concerned set targetsfor attracting x numberof investments,or xamountof contracted investment capital, and don’ t payenough attentionto how these investments will affect their industrial structure in the long term. Tomakematters worse, many foreign investors have not ful® lled the promises theymade during the negotiation process. For example, a surveyin Beijing of 702 CHINA: DYSFUNCTIONAL DEVELOPMENT?

Taiwaneseinvestments found that the average actual investment was as littleas 29%of thecontracted amount. 29 Insome cases, theprivileges extended to attract investmentoutweigh the ® nancialgains to thelocality. Of 30000joint ventures auditedby the National Administrative Bureau of State-owned Assets, the Chineseparty had lost 90% of itscapital in 80% of casesÐatotal of 10 billion Yuan.30 Thedevelopment of off-shore production centres is nowa keyfeature of China’s development.Asian investors, many of them overseas Chinesefrom Taiwan,Hong Kong, Singapore and the Chinese diaspora, 31 are establishing large-scaleproduction sites inChina which are bringingemployment and revenueto the country, and facilitating greater links with the international economy.Nevertheless, there are anumberof question marks thatremain over thelong-term impact of this type of investmentfor China’ s futuredevelopmental trajectory.On onelevel we face thedanger of aneverincreasing cycle of higher incentives(and neglect of regulatorystandards) to attractinvestment. In Palan’ s &Abbot’s terms, 32 thereis adangerthat provincial competition will bring about adysfunctional` downwardlymobile’ state strategy which the central state cannotcontrol. Theisolation of the foreign investment sector from the domestic Chinese economyalso raises anumberof concerns. Much investment from the rest of Asiainto China is componentassembly (screwdriver assembly). This investment uses lowskills and low technology, and is generatingremarkably few forward andbackward linkages into the host economy. 33 Afurtherarea forconsideration is China’s growingdependence on foreign investments for export activity. Around38% of Chinesetrade is theimport± export activities of foreign investors, butthis national ® gureunderplays the importance of the foreign direct investment (FDI)-tradelink in China’ s coastalregion. Guangdong, for example, has received40% of allforeign investment since 1979 and now provides 41% of the entirecountry’ s exports,of which 68% ` re-exports’. Withthe expansion of foreigninvestments further north in recent years, this heavy investment± trade linkageis increasinglyimportant in other coastal areas as well. Itis perhapsstretching the point too far tosay thatChina’ s globaltrade is now contingenton foreign investmentsÐ but not too far. At the very least, both externalobservers and the Chinese government itself has tocome to terms with thecomplexities of developing coherent national state strategies, given the complexitiesof lifein an interdependent international economy. And this brings metothesecond point here. Of® cial policy towards China’ s integrationinto the internationaleconomy has longbeen one of ensuring that the SEZs ` meet the needsof the international community’ . 34 Withnumerous localities outside the SEZsnowalso gearing their strategies to the interests of the international economy,the aggregate impact of these local strategies is underminingthe logic (andef® cacy)of coherent national strategy. Partlyas aconsequenceof lobbying from ` unfavoured’provinces, the Ninth FiveYear Plan carried promises of moving the focus of investment towards northernand central China. These may well turn out to be empty promises, given thecombined priorities of China’s majorexternal investors and local authorities furtherEast. Perhaps more indicative of the underlying dynamic here was Liu 703 SHAUN GBRESLIN

Ji’s `forecast’that the Bohai Rim was tobecome the next centre of foreign investment. 35 This` forecast’is aprettysound bet because it is alreadya reality, notleast because of the aggressive and proactive campaign by the Dalian authoritiesto encourage foreign investment. As such,Liu Ji’ s statementwas morea reactiverecognition of reality, than a prescriptiveplan for future developments. However,it isnotonly the national state that is facingproblems in staying in controlof events. In their analysis of Taiwanese investment in the Xiamen SEZ, QiLou and Howe show how the Xiamen authorities originally concentrated on attractingelectronics manufacturers and designed their local development strat- egyaccordingly. However, Taiwanese investors had different priorities, and insteadbrought in more and more chemicals producers. Faced with this diver- gence,the Xiamen authorities apparently abandoned their original goal and declaredpetrochemicals to be their new ` investmentemphasis’ . 36 A similar process occurredin Shenzhen where the local leadership tried to restructure the localeconomy by imposing disincentives for processing industries and componentassembly. While these industries did indeed decline, the high-tech and ®nanceinvestments that Shenzhen was hopingto attract were not forthcoming. As aresult,the local authority reversed its policy, and re-introduced a number ofincentivesto lure back the processing and component assembly investments. 37 Itseems, then,that the interaction with the regional economy which did so much tofacilitatethe development of dynamic quasi-autonomous economic policies in some partsof Chinamay now be underminingthese localities’ ability to control anddictate strategy within their own territories.

Conclusions Thispaper has emphasisedthe obstacles to effective long-term economic policy makingwhich have in¯ uenced (and continue to in¯ uence) the Chinese reform process.While it has concentratedon impediments to central control and effectivenational policy coordination, this does not mean that the central state isnolonger in¯ uential or has nocontrol.For example, the growth of investment after1992 was inmany ways a responseto Deng Xiaoping’ s tourto southern Chinawhich indicated a newcentral commitment to expanding international contacts.Nevertheless, the central state has beenunable to control the conse- quencesof its initiatives. It has set newprocesses inmotion, but has thentoo frequentlybeen placed in aresponsiveand reactive position as otheragents (both internaland external) have taken the initiative and imposed their own agendas. Indeed,one of the features of the reform process inthe 1980s was thecentre’ s increasinginability to force provincial authorities to accept periodic economic retrenchmentcampaigns (particularly Li Peng’ s campaignof 1988). Thequestion of the transferability of any model of development cannot be answeredby reference to one case studyÐIwill leave that task to others. What thisexample shows is thatit is impossibleto understand the dynamics of the Chinesereform process withoutacknowledging the speci® c historical,political andsocial context that has informedthe reform process. Notwithstanding the growinge importance of the international economic environment for China’ s 704 CHINA: DYSFUNCTIONAL DEVELOPMENT? development,perhaps the central determinant of China’s developmentstrategy is theuncertain and unclear relationship between central and local authorities. Until andunless this relationship is stabilised,it will be extraordinarily dif® cult to developa coherentand cohesive national strategy for development: in Johnson’ s wordsto facilitate ` the national mobilizationof a united peoplefor economic goals’ .38

Notes 1 S Shirk ThePolitical Logic of Economic Reform in China ,Berkeley,CA: Universityof California Press, 1993. 2 Xinhua,` ChenYun’ s speech tothe conference’ sclosingsession’ , BBCSummary of World Broadcasts , 25 September1985, FE/ 8065/C/1. 3 Theyhave also, at onetime oranother, faced theadded dilemma ofbuilding a suf®ciently large coalition towin the struggle to succeed Deng,but in a way thatdoes not threaten Deng’ s ownpower bases and interests. 4 SShirk` TheChinese political system andthe political strategy of economic reform’ inK Lieberthal& D Lampton,eds, Bureaucracy,Politics and Decision Making in Post-Mao China ,Berkeley,CA: Universityof CaliforniaPress, 1992,p 77. 5 Aterm usedby Lindblom in his incremental modelof policy making. 6 Shirk` TheChinese political system andthe political strategy of economic reform’ . 7 YouJi, ` ZhaoZiyang and the politics of in¯ ation’ , TheAustralian Journal of ChineseAffairs ,No25, 1991, pp 69±91. 8 BNaughton,` False starts andsecond wind: ® nancialreforms inChina’ s industrialsystem’ ,inE Perry& C Wong, eds, ThePolitical Economy of Reform in Post-Mao China ,Cambridge,MA: Harvard University Press, 1985,p 224. 9 CJohnson,` Politicalinstitutions and economic performance: thegovernment± business relationship in Japan, SouthKorea andTaiwan’ ,inF Deyo,ed, ThePolitical Economy of theNew East Asian Industrialism , New York:Cornell University Press, 1987,pp 151± 152. 10 RDelfs, `Lopsided growth’ , FarEastern Economic Review ,4April1991, p 24. 11 C Johnson, RevolutionaryChange ,Boston,MA: Little, Brown, 1966. 12 See ZhangZhongli, ` ShanghaiHe ShanghaiJingjiqu Zai ZhongguoJingji Xiandaihua Zhong De Diwei He Zuoyong’(The position and role of Shanghai and its economic zones in the modernisation of China’ s economy) ShehuiKexue (SocialSciences), Vol1, 1988, pp 18± 22. 13 WangHuning ` ZhongguoBianhuazhong De ZhongyangHe DifangZhengfu De Guanxi:Zhengzhi De Hanyi’(Rami® cationsof changing relationship between central andlocal government in China) Fudan Xuebao (FudanUniversity Journal), Vol 5, pp 1± 8, 30. 14 World Bank, China:Macroeconomic Instability and Industrial Growth Under Decentralized Socialism , WashingtonDC: WorldBank, 1990, p 82. 15 YouJi, ` ZhaoZiyang and the politics of in¯ ation’ , TheAustralian Journal of ChineseAffairs ,No25, 1991, pp 69±91. 16 PFerdinand, Centre± ProvincialRelations in the PRC Sincethe Death of Mao: Financial and Political Dimensions,WorkingPaper No 47, Department ofPolitics, University of Warwick, 1987, pp 7± 9. 17 JChild,` Theprocess wherebyorganizations acquire structures, programmes, staf® ngpro® les andcultures whichre¯ ect characteristics at thetime oftheir founding and which they tend to retain subsequently’ , Managementin China During the Age of Reform ,Cambridge:Cambridge University Press, 1994,p 308. 18 J` Burns, TheChinese Communist Party’ s NomenklaturaSystem ,Armonk,NY: MESharpe,1989, p xxxi. 19 Thesheer numberof appointmentsfalling under the Central Committee’ sremit hadgrown so large thatthey simplyapproved provincial nominations in most cases. Tosave time, some localauthorities had stopped seekinghigher approval before the 1984 reforms. JBurns,` China’s Nomenklaturasystem’ , Problems of Communism,Vol36, No 5, 1987, pp 36± 51. 20 ZhuLi, ` ZijinFengpei De ZhuyaoGaibian’ (Major Changes in the Distribution of Fundsin China), Jingji Guanli (EconomicReview), No9, 1987. 21 Themunicipalities were consideredtoo important for national ® nances tobe granted the same degree of ®scal autonomyas otherprovinces. Although Shanghai was initiallytreated inthe same way,it mounted an impressive lobbyingcampaign and forced a reorganisationof its® nancialobligations to thecentre in1985. 22 Yan Kalin,` Prices rise? Prices fall?’ , JingjiRibao (EconomicsDaily), 1 February1989. 705 SHAUN GBRESLIN

23 World Bank, China:Macroeconomic Instability and Industrial Growth Under Decentralized Socialism , p 9. 24 Ji Chongweicited in ` Doregional responsibility contracts point the way inwhich reform shouldhead, or are theyjust an expedient?’ , Gaige (Reform),No 3, translated in JointPublication Research Services , 7 September1988. 25 LiXianguo, ` QuyuFazhan Zhanlue De NeiyongJi ZhidingFangfa’ (The contents and formulation methods fora regionaldevelopment strategy), KeyanGuanli (.Science Research Management),No 2, April 1988, pp 14±19. 26 PCerny,` Globalizationand the changing logic of collectiveaction’ , InternationalOrganization , Vol 49, No 4,1995, p 597. 27 CJohnson,` Politicalinstitutions and economic performance’ ,p163. 28 MBernard& JRavenhill,` Beyondproduct cycles and¯ yinggeese: regionalization,hierarchy, and the industrializationof East Asia’ , WorldPolitics ,1995,Vol 47, p 194. 29 ChinaNews Service,17 July 1995. 30 ChinaNews Service,24 November 1995. 31 Somethingin the region of 80% of all foreigndirect investment in China is `Chinese’in one context or another.However, it is dif®cult to assess thereal extentof this Chinese investment given the number of externalagents who use HongKong companies as theconduit for their Chinese portfolios. 32 RPalan& JAbbott, StateStrategies in a GlobalWorld ,London:Pinter, 1996. 33 Notethat this is partlybecause thestagnating state sector is unableto ensure the reliable supplyof quality components. 34 LiBoxi cited in China News Service,3 July1995. 35 ChinaNews Service,15 November 1995. 36 Qi Lou& CHowe,` Direct investmentand economic integration in theAsia Paci® c: thecase ofTaiwanese investmentin Xiamen’ ,inD Shambaugh,ed, Greater China:The Next Superpower? ,Oxford:Oxford UniversityPress, 1995,pp 94± 117. 37 ChinaNews Service,6 December 1995. 38 CJohnson,` Politicalinstitutions and economic performance, p139(emphasis added).

706