indigenous innovation for sustainable...

341

Indigenous innovation for sustainable growth16

Yanrui Wu

Introduction

After three decades of rapid growth, the Chinese economy is now at thecrossroadsheadingtothenextphaseofdevelopment.WhileChina’seconomicgrowthhas indeedbeenphenomenal, ithasalsobeenresource intensiveandenvironmentallydamaging.Tosustainhighgrowthinthecomingdecades,theroleoftechnologicalprogresshastobeboosted.Technologicalprogresswithinacountrycanbeduetotechnologytransferfromabroadorindigenousinnovation.TheformerhasbeenwidelydiscussedintheliteratureontheChineseeconomy.Forexample,WeiandLiu(2006)examineproductivityspill-oversfromexportsandforeigndirectinvestment(FDI)intheChinesemanufacturingsector,Tian(2007) and Liu et al. (2009) investigate technology spill-overs from FDI andmultinationalcorporationsandKuoandYang(2008)analyseknowledgespill-oversandregionaleconomicgrowth.TheinnovationcapacityandachievementsofindigenousfirmsinChinaare,however,under-documented.1Theobjectiveof this chapter is to examine China’s indigenous innovation capacity and toexplore the potential for innovation to provide a key source of sustainablegrowthinthefuture.ThechapterbeginswithareviewofChina’sinnovationcapacityandachievements.Thisisfollowedbyananalysisofinnovationatthefirmlevel.Subsequently,China’sinnovationisexaminedfromaninternationalperspective, before discussing the implications for sustainable growth in thecountry.

China’s innovation capacity and achievements

Chinahasadoptedanactivescienceandtechnologydevelopmentprogramsincethe foundationof thePeople’sRepublic in1949.Theprogramhas longbeenbiased towards technologicaladvancement indefence-relatedsectors.Science

1 GeneralsurveysofChina’sresearchanddevelopmentsectorsareavailable inGaoandJefferson (2007);OECD (2009); and Zhang et al. (2009). Wei and Liu (2006) and Jin et al. (2008) also cover research anddevelopmentmarginally.

China: The Next Twenty Years of Reform and Development

342

andtechnologyasimportantsourcesofeconomicgrowthmoregenerallyhavebeenrecognisedandpromotedonlyrecently.This isclearlyenvisaged in thecountry’s‘NationalMediumandLong-termProgramforScienceandTechnologyDevelopment(2006–2020)’releasedinearly2006(hereafter,the‘2020program’).2Theaimofthe2020programistomakeChinaaninnovation-orientedsocietyby2020andoneoftheworld’sleadinginnovatorsinthelongerterm.

ThekeygoalsandprioritiesinChina’sscienceandtechnologydevelopmentinthecomingdecadearedetailedinthe2020programdocument.Accordingtothisdocument,Chinawill:

• give priority to technological development in 11 major sectors such asenergy,waterresourcesandenvironmentalprotectioninthecoming15years

• further improvethenational intellectualpropertyrights (IPR)systemandstrengthentheenforcementofIPRprotectionlawsandregulations

• encourage enterprises to play the key role in innovation through theirinvolvementinstateprojectsandtheprovisionoftaxincentivesandotherfinancialsupport

• boost investment in science and technology; by 2020, China’s researchanddevelopment expenditureswill account for about 2.5per cent of thecountry’sgrossdomesticproduct(GDP)

• by2020,derive60percentormoreofitseconomicgrowthfromtechnologicalprogress; thenumbersofpatentsgranted to and total citationsof journalarticlesbyChinesenationalsareexpectedtoberankedamongthetopfiveintheworld.

To provide an assessment of China’s indigenous innovation capacity, severalindicatorscanbeconsidered.Themostimportantfactorunderlyingindigenousinnovation is research and development spending. Associated with theimplementationofthe2020programhasbeenadramaticincreaseinresearchanddevelopmentexpenditure in recentyears.Forexample,during2005–09,researchanddevelopmentexpendituregrewatanaveragerealrateof19.4percent—twice as fast as the growth of China’s GDP.3 As a result, research anddevelopmentexpenditureasaproportionofGDP(orresearchanddevelopmentintensity)inChinarosefrom0.71percentin1990to1.62percentin2009(Figure16.1).The same figure also illustrates that China’s research and developmentpersonnelincreasedfromabout670000(full-timeequivalent)peoplein1992to1.9millionin2008.Thisgrowthwasparticularlyrapidinrecentyears,withan

2 The2020programwas releasedby theStateCouncil,People’sRepublicofChina,on9February2006(<www.gov.cn>).3 The average growth rate of research and development expenditure is calculated using data from NBS(2009,2010)andYST(2009).

Indigenous innovation for sustainable growth

343

averagerateof13.4percentduring2005–08.Inaddition,thenumberoffreshgraduatesinscience,engineeringandmedicineincreasedfrom476110in1995to2.7millionin2008(NBS2009).

Figure 16.1 China’s research and development intensity and personnel, 1990–2009

Sources:NationalBureauofStatistics(NBS)2009,China Statistical Yearbook 2009,ChtinaStatisticsPress,Beijing;NationalBureauofStatistics(NBS)2010,2009 Statistical Communiqué of National Economic and Social Development,NationalBureauofStatistics,Beijing;YHT2009,China Statistical Yearbook of High Technology Industry 2009,NationalBureauofStatisticsandMinistryofScienceandTechnology,ChinaStatisticsPress,Beijing.

With the expansion of research and development inputs, China’s innovationcapability and outcomes have increased too. For example, the numbers ofdomestic patent applications and registrations grew from 69 535 and 41 881items in 1995 to 878 000 and 502 000 items in 2009, respectively (Figure16.2).Duringthesameperiod,thenumberofChineseapplicationsforpatentregistrationoffshorealsoincreased—from13510toabout99000items,withthenumberofregisteredpatentsrisingfrom3183to80000(NBS2009;MST2010). Inaddition, it is reportedthat thenumberofpublicationsbyChinesescientistsandengineers increased from65000 to208000between1995and2007.4

There are considerable variations among the Chinese regions. Among the 31administrative regions in China, research and development intensity variedfromlessthan0.5percentinsixregionstomorethan2percentinfourregionsin 2008 (Table 16.1). In terms of human resources, the number of research

4 Thosenumbersarebasedonthesciencecitationindex(SCI),engineeringindex(EI)andindextoscientificandtechnicalproceedings(ISTP)databases,accordingtoYST(2009).


344

anddevelopmentscientistsandengineerspermillionpeoplerangedfrom161inTibet to9833 inBeijing.Table16.1also shows that thenumberofpatentregistrationspermillionpeoplewasbetween32inTibetand1296inShanghaiin2008.Ingeneral,largedisparitiesexistbetweenthecoastalregionsandtherestof thecountry (that is, thecentralandwesternregions).Furthermore, ifthenumberofpatentregistrationsper1000scientistsandengineersisdefinedas an indicator of research and development performance then, in general,provincial-level performance and inputs are positively related—as expectedandasdepictedinFigure16.3.Thefigurealsohighlights,however,twoclearoutliers, with Beijing underperforming and Zhejiang achieving an excellentperformancein2008atleast.

Figure 16.2 Numbers of patents applied for and accepted, 1995–2009

!

0

200

400

600

800

1000

1200

1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009

Year

10

00

pie

ces

Application

Acceptance

Sources:NationalBureauofStatistics(NBS)2009,China Statistical Yearbook 2009,ChinaStatisticsPress,Beijing;NationalBureauofStatistics(NBS)2010,2009 Statistical Communiqué of National Economic and Social Development,NationalBureauofStatistics,Beijing;MinistryofScienceandTechnology(MST)2010,China Science & Technology Statistics,MinistryofScienceandTechnology,Beijing,<www.most.gov.cn>

Table 16.1 China’s regional research and development statistics, 2008

Regions Expenditure over GRP (%)

Scientists and engineers per

million population

Number of patents per million

populationCoastal mean 1.96 2830 601

Beijing 5.25 9833 1047

Shanghai 2.59 4212 1296

Tianjin 2.45 3293 577

Jiangsu 1.92 1887 579

Zhejiang 1.60 2067 1034

Liaoning 1.41 1538 247

Guangdong 1.41 2186 650

Shandong 1.40 1408 283

Fujian 0.94 1345 220


345

Hebei 0.67 535 79

Middle mean 0.85 731 84

Hubei 1.31 1103 147

Anhui 1.11 655 71

Heilongjiang 1.04 1168 120

Hunan 1.01 604 96

Jiangxi 0.97 540 52

Shanxi 0.90 974 67

Jilin 0.82 1085 109

Henan 0.66 583 97

Guangxi 0.46 426 46

Hainan 0.23 172 40

Western mean 0.81 597 80

Shaanxi 2.09 1352 117

Sichuan 1.28 768 164

Chongqing 1.18 995 170

Gansu 1.00 593 40

Ningxia 0.69 694 98

Guizhou 0.57 257 46

Yunnan 0.54 357 44

Inner Mongolia 0.44 647 55

Qinghai 0.41 377 41

Xinjiang 0.38 366 70

Tibet 0.31 161 32

Sources:NationalBureauofStatistics(NBS)2009,China Statistical Yearbook 2009,ChinaStatisticsPress,Beijing;YST(2009).

Innovation at the firm levelChinese enterprises have played an important role in promoting growth inthe country’s innovation capacity. Since the mid 1990s, Chinese enterpriseshavebecometheleadingplayersinresearchanddevelopmentinvestmentandexecution(NBS2009).By2008,theywereaccountingformorethan70percentof the country’s research and development investment and spending (Figure16.4).Thisgrowthisalsoreflectedinthechangingshareofpatentsregisteredbytheenterprisesectorcomparedwiththenationaltotal.Thisshareincreasedfrom12percentin1995to34percentin2008.5TheexpandedroleofChineseenterprises could, however, have led to more market-driven research anddevelopmentinvestment.Thisisreflectedinthemovementoftwoindicators.

5 ThosesharefiguresarecalculatedusingChina’spatentdata(NBS2009).


346

First,theshareofbasicandappliedresearchexpenditureovertotalresearchanddevelopmentspendingdeclinedfrom32percentin1995to17percentin2008(NBS2009).Second,theshareof‘invention’patentsovertotaldomesticpatentsregisteredpeakedat25.9percentin2004andhassincefallen,reaching22.7percentin2008(NBS2009).6Thechallengeaheadforpolicymakersistoensurethatmarket-orientedresearchanddevelopmentactivitiesdonotgrowattheexpenseoflong-terminnovationcapacitybuildinginthecountry.

Figure 16.3 China’s research and development intensity and performance, 2008

Note:Researchanddevelopmentefficiencyisdefinedasthenumberofpatentregistrationspermillionscientistsandengineers.

Sources:ResearchanddevelopmentintensityandefficiencyvaluesarecalculatedusingdatafromNationalBureauofStatistics(NBS)2009,China Statistical Yearbook 2009,ChinaStatisticsPress,Beijing;YST2009,China Statistical Yearbook of Science and Technology 2009,NationalBureauofStatisticsandMinistryofScienceandTechnology,ChinaStatisticsPress,Beijing.

Figure 16.4 China’s research and development expenditure shares and sources of funds, 2008

Expendituresshares Sourcesoffund

Source:NationalBureauofStatistics(NBS)2009,China Statistical Yearbook 2009,ChinaStatisticsPress,Beijing.

6 Chinesepatentsaregenerallygroupedintothreecategories:inventions,utilitymodelsanddesigns.


347

Intermsofinnovationactivitiesandefforts,heterogeneityexistsacrosssectorsaswellasamongfirms.TheNationalBureauofStatistics(NBS)conductedthefirstnationalsurveyoffirminnovationactivitiesin2007.Detailedinformationabout firm-level innovation activities was collected over a three-year period(2004–06).Thesurveycoveredalllargeandmediumfirmsandasampleofsmallfirms.7Amongthe299995firmssurveyed,therewere2674largefirms(0.9percent), 29 622 medium firms (9.9 per cent) and 267 699 small firms (89.2 percent).Only86342firms—lessthanone-thirdofthetotal—wereinfactengagedininnovationactivities(NBS2008).Accordingtothesurvey,largefirmswerefoundtohavethehighestrate(83.5percent)ofparticipationinresearchanddevelopmentactivities,followedbysmall(25.2percent)andmedium(55.9percent)firms.Attheindustrylevel,thetopfivesectorsintermsofparticipationrates were pharmaceuticals (63.7 per cent), instruments and office machines(60.7 per cent), tobacco (55.2 per cent), communication and other electronicequipment(46.8percent)andspecialmeasuringinstruments(46.5percent).8Allofthesesectorsotherthantobaccobelongtotheso-calledhigh-technologysector.9

Onaverage,innovativefirmsin2006spentabout1.9percentoftheirbusinessincomeon innovation.Thoughthisfigure is larger thanChina’sresearchanddevelopment intensity, there is huge disparity among firms. Large firms onaverage investedabout2.7percentoftheirbusiness incomeonresearchanddevelopment,whichiswellaheadofthemedium(1.8percent)andsmallfirms(1percent)(NBS2008).Thelargeandmediumenterprises(LMEs)asagroupaccountedfor81.1percentoftotalexpenditureoninnovationin2006.Theyalsohadavalueshareof78.7percentintheoutputofnewproducts,whichcouldbeusedasanalternativeindicatorofinnovationoutcome.Forthisreason,moststudiesof innovationatthefirmlevel inChinafocusonLMEs(suchasJeffersonetal.2003;Girmaetal.2009).

Theanalysisbelowprovidesastudyofresearchanddevelopmentdeterminants,strategiesandintensityinChinesefirmsusingpaneldatafor19880LMEsfortheperiod2005–07.10Threedifferentyetrelatedmodelsareestimated.Thesemodelsinturndealwiththreeissues—namely,thedeterminantsofinnovation,

7 AccordingtotheofficialEnterpriseClassificationStandardsadoptedin2003,Chinesefirmsaregroupedusingthreecriteria:thenumberofemployees,salerevenueandvalueofassets.Forexample,thenumberofemployeesisabove2000forthelargefirms,between300and2000forthemediumfirmsandbelow300forthesmallfirmsinthemanufacturingsector(<www.stats.gov.cn>).8 ThepercentagefiguresinparenthesesaretheratesofparticipationinresearchanddevelopmentactivitiesaccordingtoNBS(2008).9 TheNationalBureauofStatistics(NBS)releasedacirculartointroducetheCatalogue for High-technology Industrial Statistics ClassificationinJuly2002(<www.stats.org.cn>).10 SeeWu(2010)foramoredetailedpresentationanddiscussion.


348

thechoiceofresearchanddevelopmentstrategiesandtheintensityofresearchand development spending in Chinese firms. The baseline model can bepresentedasEquation16.1.

Equation 16.1

!

InEquation16.1, *Y isa latentvariable thathasavalueofunityorzero forthe first two models (that is, the determinants of innovation and choice ofresearch and development strategy models) and measures firm-level researchanddevelopmentintensityinthethirdmodel(thatis,theintensityofresearchanddevelopmentspendingmodel).Thelaggedvariables(X)capturetheeffectsof theageandsizeofeachfirm, its levelof liabilityordebtburden, levelofproduction technology, intangible assets and long-term investment. The X-variablesarelaggedoneperiodtoavoidpotentialsimultaneityproblemsinthemodels.Specifically,thesevariables(X)aredefinedasfollows

• AGEissimplytheageofthefirm(yearsinexistence)

• SIZEreflectsthesizeofthefirm,measuredusingthenumberofemployees11

• DEBTmeasuresthedegreeofliability,definedastheratiooftotalliabilityoverthetotalvalueofassets

• TECHcapturestheleveloftechnologyinproduction,measuredbytheratioofthenetvalueofassetsoveremployment(thatis,thecapital–labourratio)

• INTANGreflectswhetherafirmhasintangibleassets(suchaspatents);itisdefinedas1ifthefirmhasintangibleassetsandzerootherwise

• INVEST takes thevalueof1 if afirmhas long-term investmentandzerootherwise.

Other independent variables—namely, the Z variables in Equation 16.1—areintroduced to reflect firms’ productivity performance and exporting status,industryconcentrationandvariationsacrossfirmswithregardtoownership,location,industryandtime.Thesevariables(Z)include

• EFF,whichisanindicatoroffirmefficiency,measuredsimplybythefirms’labourproductivity—thatis,theratioofoutputvalueovertotalemployment

• EXP,whichisabinaryvariableandhasavalueof1ifafirmisengagedinexportingandzerootherwise

• theHerfindahlindex(HERFINDAHL),whichiscomputedtomeasurethelevelofcompetitionorconcentrationofbusinessactivitiesinasector;the

11 Thereare,ofcourse,othermeasuresoffirmsizesuchastotaloutputvalue,thevalueoftotalsalesandsoon.Thenumberofemployeesischosensothatfewobservationsaredroppedduetomissingdata.


349

calculation is based on the four-digit classification of Chinese industrialsectors

• four ownership dummy variables, which are included to represent firmsownedorcontrolledthroughmajorityshareholdingbytheState,investorsfrom Hong Kong, Macau and Taiwan (HMT), foreign investors andshareholders(versusallotherfirms)

• fivedummyvariables,whichcapturevariationsamongfirmslocatedintheareasofBeijing,Shanghai,PearlRiverDelta,thesix‘middle’provinces,thethree north-eastern provinces and western China, with the north-easternprovincesbeingchosenasthereferenceregion12

• 11 sector dummy variables to reflect potential sectoral differences amongthefirms,whicharecategorisedinto12industrysectorsonthebasisoftheofficialStandardIndustryClassification(SIC)grouping.

TheestimationresultsofthethreemodelsarepresentedinTable16.2.Accordingto the results of Model 1, large or old firms are more likely than others toinvest in innovation. Exporters and capital-intensive firms also have higherprobabilitiesofspendingonresearchanddevelopment—asarefirmswithlong-terminvestments,intangibleassetsandbetterperformance(intermsoflabourproductivity).Theprobability of innovation tends to increase over time andthatmore competition increases initially and then reduces theprobabilityofinnovation.Thisisconsistentwithevidencefromothereconomies(Aghionetal.2005;TingvallandPoldahl2006).Firmsthatarelesslikelytoinvestinresearchanddevelopmentareoftenburdenedwithheavydebtorownedbyoffshoreinvestors—inparticular, investors fromHongKong,Macau andTaiwan.ThelatterhavealargepresenceinthePearlRiverDeltaregion.Thisfindingimpliesthatthecountry’sforeigninvestmentpoliciesmightnotbeprovidingincentivesforforeignfirmstoinvestinresearchanddevelopmentinChina.

Table 16.2 Econometric estimation results

Model 1 Model 2 Model 3Variables Coefficients p-values Coefficients p-values Coefficients p-values

Intercept –2.2088 0.000 –1.3537 0.000 0.1013 0.025

AGE 0.0098 0.000 0.0087 0.000 0.0024 0.000

AGE2 –0.00005 0.000

12 Specifically,China’s 31 administrative regions arepartitioned into sixgroups and representedby sixdummy variables: REG1 (Beijing,Tianjin, Hebei and Shandong), REG2 (Shanghai, Jiangsu and Zhejiang),REG3(Guangdong,Fujian,GuangxiandHainan),REG4(Shanxi,Anhui,Jiangxi,Hubei,HunanandHenan),REG5 (Liaoning, Jilin and Heilongjiang) and REG6 (Inner Mongolia, Ningxia, Tibet, Xinjiang, Gansu,Guizhou,Qinghai,Shaanxi,Sichuan,YunnanandChongqing).


350

SIZE 0.0000 0.051 0.0000 0.000 0.0000 0.543

EXP 0.3319 0.000 0.2535 0.000 0.0321 0.000

DEBT –0.1763 0.000 –0.2149 0.001 –0.6026 0.000

TECH 0.0001 0.000 0.0000 0.009 0.0000 0.832

EFF 0.0000 0.011 0.0000 0.771 –0.0001 0.000

INTANG 0.2223 0.000 0.1789 0.000 –0.0358 0.000

INVEST 0.4580 0.000 0.3566 0.000 0.2252 0.000

HERFINDAHL 3.0130 0.000 2.6715 0.000 3.8713 0.000

HERFINDAHL2 –9.6779 0.000 –8.3931 0.004

Region dummies yes yes yes

Ownership dummies yes yes yes

Year dummies yes no yes

Seudo-R 0.1480 0.0842 0.6091

Sample size 59640 27102 13446

Notes: Models 1 and 2 are estimated using quadratic hill-climbing optimisation algorithm and quasi-maximumlikelihood(Huber-White)robuststandarderrorsandco-variance.Model3isestimatedusingpanelEGLSwithcross-sectionweightsandWhitecross-sectionstandarderrorsandco-varianceprovidedinEview 6.

Source:Authors’ownestimations.

Theestimatedcoefficients (notreported in the table)of thedummyvariablesalso show that firms engaged in manufacturing pharmaceuticals, machinery,transport equipment, communication and other electronic equipment aremorelikelythanotherstoinvestininnovation.Theseproductsaremainlyinthehigh-techsectors.Infact,attheaggregatelevel,onaverage,researchanddevelopmentintensityinthehigh-techsectorsismuchhigherthanthenationalaverageof1.4percentin2007(seeFigure16.1).Forexample,thepercentageshare of research anddevelopment expenditure over sectoralvalue added in2007was4.7inpharmaceuticals,15.4inaircraftandspacecraft,6.8inelectronicandtelecommunicationequipment,3.9incomputersandofficeequipmentand6.3inmedicalequipmentandmetermanufacturing(YHT2008).Theestimationresultsalsoimplythatstate-ownedandshareholdingfirmsaremorelikelythanotherstobeinnovators.13Itisalsointerestingtonotethatfirmslocatedinthecentralandwesternregions—inparticular,thewesternregion—aremorelikelythanotherstospendonresearchanddevelopment.Thiscouldreflectthefactthatstate-ownedenterprises(SOEs)playamoreimportantroleintheeconomiesof thecentralandwesternregions—accountingfor25.1percentand30percentoffirmsinthetworegionsrespectively,comparedwithashareof15.6percent in thecoastal area.Thesefindingsabout the roleofSOEs in innovationsuggestthatprivatisationisnotalwaysconducivetoinnovation(atleastbeforeChina’sprivatefirmscanplayamoreprominentroleininnovation).

13 ThedetailedresultsarereportedinWu(2010).


351

According to the estimation results of Model 2 in Table 16.2, persistentinnovators—definedasfirmsthatinvestedinresearchanddevelopmenteveryyearduringtheperiodsurveyed—aremorelikelytobeassociatedwithlarge-scaleproduction,oldageintermsofcommencementdate,exportingstatusandhighcapital–labourratios(orcapital-intensivetechnology).Persistentinnovatorsarealsolikelytohavelong-terminvestmentsorintangibleassets.Wefoundhighindebtednesstobenegativelylinkedwiththeprobabilityofbeingapersistentinnovator(Table16.2).Relativelyefficientfirmsarenotnecessarilymorelikelythanotherstobepersistentinnovators.FirmscontrolledbyinvestorsfromHongKong,MacauandTaiwanorlocatedinthePearlRiverDeltaregionarelesslikelythanotherstobepersistentinnovators.Ingeneral,SOEsandshareholdingfirmsaremorelikelytobepersistentinnovators,asarefirmslocatedinwesternChinaorinvolvedinmanufacturingpharmaceuticals,machinery,transportequipment,communicationandotherelectronicequipment.Theseproductsonceagainareproducedmainlyinthehigh-techsectors,asexpected.

Wefoundthatfirms’researchanddevelopmentintensitywasnegativelyrelatedtotheageofthefirm(Wu2010).Furtheranalysis,however,demonstratesthatfirms’researchanddevelopmentintensityincreases initiallyandthenfallsastheir capital ages (Model3,Table16.2); theestimated turningpoint is about25years.Sincetheaverageageofthefirmsinthesampleis17,mostChineseenterprisesare stillon theupward (left) sideof the inverted-Ushape.As fortherelationshipbetweencompetitionandresearchanddevelopmentintensity,thereisnoevidenceofaninvertedU-shapedrelation,asarguedbyAghionetal. (2005) andTingvall andPoldahl (2006). Instead it is shown inTable 16.2thatresearchanddevelopmentintensityandcompetitionarenegativelyrelated.This supports the argument thatdominantfirms tend tobemore innovativethannon-dominantones(Blundelletal.1995).

ItisfurthershowninTable16.2(Model3)thatfirms’researchanddevelopmentintensityisassociatedpositivelywiththeexistenceoflong-terminvestments,exportingstatus,largesizeandhighcapitalintensity,althoughthelasttwohaveinsignificantcoefficients.Itisalsofoundthatfirms’researchanddevelopmentintensityislinkednegativelywithfirmliability,efficiencyandtheexistenceofintangibleassets.Ifafirmpossessesintangibleassets,itcouldimplythatthefirmiswellestablishedinthefield(withnewproductsorpatents,forinstance)andneedsresearchanddevelopmentinvestmentonlytomaintaintheleadingedge.The negative relationship between efficiency and research and developmentintensityisapuzzle.ItcouldreflecttheroleofSOEsininnovation.ChineseSOEsaregenerallylessefficientthanothers,buttheyarethekeyplayersinresearchanddevelopmentactivitiesinChina.ThisisconsistentwiththepositivesignofthecoefficientofthedummyvariablerepresentingSOEs.Thesefindingsimplythat China is facing a dilemma.While the need for further economic reform


352

callsforthewithdrawalorprivatisationofSOEs,China’sprivatefirmsarenotreadytotakeovertheriskybusinessofresearchanddevelopmentinvestment.Thus,asfarasinnovationisconcerned,specificpoliciesarerequiredtohelpthetransitionfromSOEstoprivatefirms.

International perspective

Among theworld’smajor spenderson researchanddevelopment,Chinawasrankedthird in2007—behindonlytheUnitedStatesandJapan(Table16.3).Itshouldbepointedout,however,thatthereisstillalargegapbetweenChinaand the world’s top-two research and development investors. For example,in 2007, China’s total research and development spending was about 28 percentand70percentof that in theUnitedStates andJapan, respectively. Intermsofresearchanddevelopmentintensity,althoughChinaisstillbehindtheworld’stopspenders,itiswellaheadofmajoreconomiesatasimilarstageofdevelopment—asisclearlydemonstratedinFigure16.5.Ifcurrentgrowthinresearchanddevelopmentspendingismaintained,wespeculatethatChinawillfollowtheinnovationpathsofSouthKoreaandJapanandbecomeoneofthemostinnovativecountriesintheworld.

Table 16.3 World’s top five research and development spenders, 2007

Shares (%)

Country

Research and development expenditure (PPP$ billion in 2000 prices) Business Government Others

Research and development personnel (million people)

US 311.4 66.2 28.3 5.5 1.426

Japan 124.6 77.7 15.7 6.7 0.938

China 87.1 70.4 24.6 5.1 1.736

Germany 58.7 68.0 27.8 4.3 0.506

France 35.6 52.0 38.2 9.8 0.372

United Kingdom

33.3 46.5 30.0 23.4 0.349

PPP=purchasingpowerparity

Note:ResearchanddevelopmentpersonnelfiguresfortheUnitedStatesare2006data.

Sources: Research and development expenditure and personnel data are drawn from the OECD onlinedatabase(<www.oecd.org>).


353

Figure 16.5 Research and development intensity and per capita GDP of major economies, 2007

Note:ResearchanddevelopmentintensityisthepercentageshareofresearchanddevelopmentexpenditureoverGDPineachcountry.GDPpercapitaisexpressedin2005constantinternationalprices.

Source:WorldBank2010,World Development Indicators,TheWorldBank,Washington,DC.

Furthermore,thecombinedshareofbasicandappliedresearchexpenditureovertotalresearchanddevelopmentspendingmaintainedadecliningtrendinrecentyears.Itwasonly17.2percentin2008,whilethisratioismuchhigherandrisingorrelativelystableovertimeinthemajorindustrialisedeconomies(Figure16.6).Thus,thepatternofChina’sresearchanddevelopmentspendingdeviatesfromtheglobal trendandisbiasedtowards investment in ‘development’research.Asdiscussedabove, this couldhave long-term implications for thecountry’sinnovationcapacitybuilding.For instance,anemergingtrendisthat,amongthepatentsgranted,only22.7percentbelongedtothe‘inventions’categoryin 2008.The structure and quality of China’s research and development arethereforechangingasinvestmentincreasesovertime.

In2006,ChinaforthefirsttimeovertooktheUnitedStatestohavetheworld’slargestnumberofresearchers.14By2008,China’sresearchanddevelopmentsectorhadmorethan1.9millionemployees,ofwhichmorethan84percent(about1.6million)werescientistsandengineers.15Meanwhile,inthesameyear,therewereabout6.1millionstudents,including759000postgraduatestudents,whowereenrolledintheschoolsofscience,engineeringandmedicalsciencesinChineseuniversities (NBS2009).Thus,China’spotential in researchanddevelopmenthumanresourceswillundoubtedlybethelargest intheworldinthecoming

14 ThisisbasedondatafromtheOECDonlinedatabase(<www.oecd.org>).15 These numbers are drawn from the Annual Statistics of Science and Technology, National Bureau ofStatisticsofChina(www.stats.gov.cn).


354

decades. The country’s comparative advantage in human resources is alsoreflectedintheresearchanddevelopmentcoststructure.Labourcompensationaccounted for about 25 per cent of total research and development costs in2007, which was much lower than that in many Organisation for EconomicCooperation andDevelopment (OECD) countries suchas Japan (39per cent),SouthKorea(44percent),theUnitedKingdom(48percent),theUnitedStates(57percent),France(57percent)andGermany(60percent)inthesameyear.16ThissuggeststhatChinastillenjoysaconsiderablecomparativeadvantage inlabourcosts.Thereare,however,risksassociatedwithlowcompensationpaidtoscientistsandengineers.Skilledlabourisverymobileintoday’sworldandlowwagescouldmakeChinalesscompetitiveintheinternationaltalentmarket.

Figure 16.6 Research and development spending shares (basic and applied research) in selected economies

Sources:NationalBureauofStatistics(NBS)2009,China Statistical Yearbook 2009,ChinaStatisticsPress,Beijing;OECDonlinedatabase(<www.oecd.org>).

Anotherimportantfactorcloselyrelatedtoinnovationisthedevelopmentofthehigh-techsector.During1996–2007,theaveragerealgrowthrateofvalueaddedinthissectorwas18.7percent,whichwastwiceasfastasthegrowthrateoftheChineseeconomy.Computerandofficeequipmentmanufacturingledthisgrowth,witharealrateofgrowthof28.8percent,followedbytheelectronicand telecommunications equipment sector, with 18.8 per cent, and medicalequipmentandmetermanufacturing,with18percent(Figure16.7).In2008,thehigh-techindustryasagroupamountedto12.9percentoftotalmanufacturingoutputinChina(DPD2009).Thevalueofexportsinthissectorhasachievedanaveragerateofgrowthof44.9percentduring2002–08.17Incomparisonwithmajor high-tech exporters in the world, China has the largest global marketshare(Table16.4).In2007,China’shigh-techsectoralsoaccountedfor29.7per

16 LabourcompensationsharesovertotalresearchanddevelopmentcostsareestimatedusingdatafromtheOECDonlinedatabase(<www.oecd.org>).17 ThisisanominalrateofgrowthcalculatedusingdatafromYHT(2008).


355

centoftotalmanufacturingexportsfromthecountry.Thisfigureiscompatiblewith that in the United States, which is the world’s second-largest exporter,butiswellbehindotherEastAsianeconomiessuchasthePhilippines(68.9percent),Malaysia(51.7percent),Singapore(46.4percent)andSouthKorea(33.4percent)(Table16.4).ItwillbeinterestingtoseewhetherChinaisabletofollowitsAsianneighboursintermsofthedevelopmentofthehigh-technologysector.

Figure 16.7 High-technology sector value added (in 1995 constant prices)

0

50

100

150

200

250

300

350

400

450

500

1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008

Bill

ion

yuan

Year

Pharmaceuticals

Aircraft and spacecraft

Electronic and telecommunication equipments

Computers and office equipments

Medical equipments and meters

Sources:YHT2002,China Statistical Yearbook of High Technology Industry 2002,NationalBureauofStatisticsandMinistryofScienceandTechnology,ChinaStatisticsPress,Beijing;YHT2008,China Statistical Yearbook of High Technology Industry 2008,NationalBureauofStatisticsandMinistryofScienceandTechnology,ChinaStatisticsPress,Beijing.

Table 16.4 High-technology sector exports in selected economies, 2007

Countries Percentage of manufactured exports

World shares (%)

China 29.7 18.6

United States 28.5 12.7

Germany 14.2 8.6

Japan 19.0 6.7

South Korea 33.4 6.1

Singapore 46.4 5.8

France 18.9 4.5

Netherlands 25.7 4.1

Malaysia 51.7 3.6

United Kingdom 19.5 3.5

Mexico 17.1 1.8

Philippines 68.9 1.6

Brazil 11.9 0.5

India 5.3 0.3

Russian Federation 6.9 0.2

Source:WorldBank2010,World Development Indicators,TheWorldBank,Washington,DC.


356

Implications for future economic growth

A precise assessment of the contribution of innovation to China’s growthis technically and conceptually challenging. The empirical literature isdominated by growth-accounting exercises and has focused on the analysisof the traditional Solow-type innovation or total factor productivity (TFP)growth.Wu(forthcoming)reviewsmore than70studieswithmore than150estimatesofTFPgrowthratesandfindsthatTFPgrowthonaverageaccountedforaboutone-thirdofChina’seconomicgrowthduringthe1990sandthefirstdecadeof thiscentury.Thesamefigureformore industrialisedeconomies is,however, much higher. For example, according to Dougherty and Jorgenson(1996), productivity growth accounted for 49.8 and 57.6 per cent of outputgrowthduring1960–89inJapanandGermany,respectively.Therefore,thereisconsiderablescopeforimprovementinChina’sproductivitygrowthingeneralandininnovationinparticular.

Themodestcontributionof innovationtoeconomicgrowth in thepast threedecadesisconsistentwiththecountry’sinnovationconditions.Asshownintheprecedingsections,evidenceatthemacroandmicrolevelsillustratesthatthereisstillaconsiderablegapbetweenChinaandtheadvancedeconomiesintermsofinnovationresourcesandcapacity.Chinais,however,catchinguprapidlywithindustrialised economies in terms of its innovation capacity measured usingvariouscriteriasuchasthenumberofpatentsregistered,scientificpublicationsandcitationsandhi-techcommodityexports.Thedrivingforcesforthecatch-upare the increasingresearchanddevelopment inputs incapitalandhumanresources. The catch-up will make it possible for the country to realise itsinnovationpotential,whichwillbevitalforChina’ssustainablegrowthinthecomingdecades.Toreachthisgoal,severalemergingissuesmustberesolvedbyChinesepolicymakers.

First,intermsofaggregateinvestmentinresearchanddevelopment,ChinaisaheadofothercountriesatasimilarstageofdevelopmentandthecountryisalsorapidlycatchingupwithOECDeconomies.Thereare,however,areaswhereChinacoulddomuchbetter.Forexample,China’s researchanddevelopmentintensity in the high-tech sector is lagging behind the major players in theworld(Table16.5).Infourofthefivehigh-techsectors(theexceptionbeingtheaircraftandspacecraftsector),thereisahugegapinresearchanddevelopmentintensity. Even in the aircraft and spacecraft industry, China’s research anddevelopmentintensitywasabouthalfthatinGermany,FranceandtheUnitedKingdomin2007(Table16.5).


357

Table 16.5 High-technology sector research and development intensity in selected economies, 2006 (per cent)

Industries China US Japan Germany France United Kingdom

Italy Korea

Manufacturing total

3.4 10.2 11.0 7.6 9.9 7.0 2.4 9.3

High-tech sector otal

5.7 39.8 28.9 21.5 31.9 26.6 11.1 21.3

Pharmaceuticals 4.7 46.2 37.1 23.9 33.4 42.3 5.0 6.3

Aircraft and spacecraft

14.9 24.1 11.5 32.9 31.1 31.1 45.2 26.1

Electronic and telecommunication equipment

6.4 43.3 13.4 28.8 50.9 23.9 11.6 25.1

Computers and office equipment

3.8 34.7 n.a. 14.9 27.7 1.4 8.4 14.2

Medical equipment and meters

5.2 48.3 31.9 13.6 19.0 7.8 6.7 10.3

n.a.notavailable

Sources: YHT 2008, China Statistical Yearbook of High Technology Industry 2008, National Bureauof Statistics and Ministry of Science andTechnology, China Statistics Press, Beijing;YHT 2009, China Statistical Yearbook of High Technology Industry 2009,NationalBureauofStatisticsandMinistryofScienceandTechnology,ChinaStatisticsPress,Beijing.

Second, with the expansion of research and development activities it isimportant not to neglect the quality of research and development in China.Theprecedingsectionspresentedevidenceof relativedeclines in investmentinbasicandappliedresearchinrecentyears.Thisisalsoreflectedinthesmallshare of ‘invention’ patents among total domestic patents in China. If thistrend continues, China’s long-term capacity in innovation—and hence thesustainabilityofeconomicgrowthinthefuture—couldbecompromised.

Third, the role of privately owned enterprises, including foreign firms, ininnovation should be strengthened through more stringent enforcement ofintellectualpropertyrightsprotectionlawsandregulationsandtheprovisionof incentivesviaappropriate innovationpolicies.Asshown in theprecedingsections,theenterprisesectorplaystheleadingroleininnovationintheworld’smajor economies. Although China’s privately owned firms are expandingrapidly, in terms of innovation, they are lagging behind their state-ownedcounterparts, not to mention privately owned firms in other countries.Thiscould be due to institutional constraints in China such as limited access tofinanceandgovernmentgrantsfornon-SOEs.


358

Finally, there is considerable regional disparity in innovation.This not onlycontributes to China’s overall regional disparity, but it is also detrimental tothe diffusion of knowledge and technology within China. There should beinstitutional mechanisms to ensure a more even distribution of innovationresourcesacrosstheChineseregions.

Conclusions

Insummary,Chinahasmadeconsiderableprogress indeveloping indigenousinnovationskillsandcapacityinthepastthreedecades.Thistrendhasbeenstrengthenedthroughtheimplementationofthe2020programofscienceandtechnologydevelopmentsince2006.ItalsolaysthefoundationforthepossibletransformationofChina’seconomicgrowthmodelfromaresource-intensiveonetoaninnovation-orientedmodel.China’sinvestmentininnovationhasalreadygrownrapidly,withinnovationoutcomesexpandingasaconsequence.

An important development in recent years is the expanded role of Chineseenterprises in innovation. Chinese firms are now the dominant investors inresearchanddevelopmentinthecountry.Intermsofinnovation-relatedfirmcharacteristics, however, there is heterogeneity across firms with differentownershipandscaleaswellasindifferentindustriesandlocations.ItisshowninthisstudythatSOEsperformmuchbetter thanforeign-investedfirmsandprivatelyownedChinesefirmsasfarasresearchanddevelopmentpropensityandeffortsareconcerned.ThisisadilemmaforChina.Aseconomicreformdeepens,SOEs are under pressure to privatise. In the meantime, non-SOEs (includingforeignandindigenousprivatefirms)arenotreadytotakerisksassociatedwithresearch and development activities. This situation calls for specific policiesencouragingtheparticipationofnon-statefirmsininnovationandimprovementofthelegalsystemtoprovideeffectiveprotectionofintellectualpropertyrightsinChina.

ThisstudyhasalsodemonstratedthegapbetweenChinaandtheworld’sleadinginnovators.Toclosethegap,Chinesepolicymakerscouldpaymoreattentiontoseveral issues inthecomingdecades.First,whileChina is theworld’s largestexporterofhigh-techproducts,China’sresearchanddevelopmentintensityinthehigh-techsectorislaggingbehindtheworld’smajorplayers.Second,astheroleofenterprisesininnovationisstrengthened,thereisthedangerofneglectingbasicandappliedresearch,whichisvitalforthecountry’sinnovationcapacitybuildinginthelongerterm.Finally,boththequantityandqualityofinnovationinvestmentsandproductsshouldbemonitoredduringtheprocessofeconomictransformation.Growthshouldnotcompromisethequalityofinnovation.


359

References

Aghion,P.,Bloom,N.,Blundell,R.,Griffith,R.andHowitt,P.2005,‘Competitionandinnovation:aninverted-Urelationship’,Quarterly Journal of Economics,vol.120,no.2,pp.701–28.

Blundell,R.,Griffith,R.andvanReenen,J.1995,‘Dynamiccountdatamodelsoftechnologicalinnovation’,Economic Journal,vol.105,pp.333–44.

Dougherty, C. and Jorgenson, D.W. 1996, ‘International comparisons of thesources of economic growth’, American Economic Review (Papers and Proceedings),vol.86,no.2,pp.25–9.

DevelopmentandPlanningDivision(DPD)2009,An analysis of China’s hi-tech industrial development in 2008,ScienceandTechnologyStatisticsReport,no.25,DevelopmentandPlanningDivision,MinistryofScienceandTechnology,Beijing.

Gao, J. and Jefferson, G. 2007, ‘Science and technology take-off in China:sourcesofrisingR&Dintensity’,Asia Pacific Business Review,vol.13,no.3,pp.357–71.

Girma,S.,Gong,Y.andGörg,H.2009,‘WhatdeterminesinnovationactivitiesinChinesestate-ownedenterprises?Theroleofforeigndirectinvestment’,World Development,vol.37,no.4,pp.866–73.

Jefferson,G.,Hu,A.G.,Guan,X.andYu,X.2003,‘Ownership,performance,and innovation in China’s large- and medium-size industrial enterprisesector’,China Economic Review,vol.14,no.1,pp.89–113.

Jin,F.,Lee,K.andKim,Y.-K.2008,‘ChangingenginesofgrowthinChina:fromexports,FDIandmarketizationtoinnovationandexports’,China and World Economy,vol.16,no.2,pp.31–49.

Kuo,C.-C.andYang,C.-H.2008,‘Knowledgecapitalandspilloveronregionaleconomicgrowth:evidencefromChina’,China Economic Review,vol.19,no.4,pp.594–604.

Liu,X.,Wang,C.andWei,Y.2009,‘DolocalmanufacturingfirmsbenefitfromtransactionallinkageswithmultinationalenterprisesinChina?’,Journal of International Business Studies,vol.40,pp.1113–30.

Ministryof Science andTechnology (MST) 2010,China Science & Technology Statistics,MinistryofScienceandTechnology,Beijing,<www.most.gov.cn>


360

NationalBureauofStatistics(NBS)2008,Statistics of 2007 Innovation Survey of Industrial Firms,ChinaStatisticsPress,Beijing.

NationalBureauofStatistics(NBS)2009,China Statistical Yearbook 2009,ChinaStatisticsPress,Beijing.

NationalBureauofStatistics(NBS)2010,2009 Statistical Communiqué of National Economic and Social Development,NationalBureauofStatistics,Beijing.

Organisation for Economic Cooperation and Development (OECD) 2009,Measuring China’s innovation system: national specificities and international comparisons, STI Working Paper 2009/1, Statistical Analysis of Science,Technology and Industry, Organisation for Economic Cooperation andDevelopment,Paris.

Tian,X.2007,‘AccountingforsourcesofFDItechnologyspillovers:evidencefromChina’,Journal of International Business Studies,vol.38,no.1,pp.147–59.

Tingvall, P. G. and Poldahl, A. 2006, ‘Is there really an inverted U-shapedrelationbetweencompetitionandR&D?’,Economics of Innovation and New Technology,vol.15,no.2,pp.101–18.

Wei,Y.andLiu,X.2006,‘ProductivityspilloversfromR&D,exportsandFDIinChina’smanufacturingsector’,Journal of International Business Studies,vol.37,pp.544–57.

WorldBank2010,World Development Indicators,TheWorldBank,Washington,DC.

Wu,Y.2010,Are Chinese enterprises innovative? R&D determinants, strategies and intensity,EconomicsDiscussionPapers,UniversityofWesternAustralia,Perth.

Wu,Y.(forthcoming),‘WhatdowereallyknowaboutproductivitygrowthinChina?’,inL.Xu(ed.),China’s Economy in the Post-WTO Environment: Stock markets, FDI and challenges of sustainability,EdwardElgar,Cheltenham.

YHT2002,China Statistical Yearbook of High Technology Industry 2002,NationalBureauofStatisticsandMinistryofScienceandTechnology,ChinaStatisticsPress,Beijing.



361


YST2009,China Statistical Yearbook of Science and Technology 2009,NationalBureauofStatisticsandMinistryofScienceandTechnology,ChinaStatisticsPress,Beijing.

Zhang,C.,Zeng,D.Z.,Mako,W.P.andSeward,J.2009,Promoting Enterprise-Led Innovation in China,TheWorldBank,WashingtonDC.

indigenous innovation for sustainable...

Documents