05_On the link between Japanese ODA and FDI in China

8/7/2019 05_On the link between Japanese ODA and FDI in China

1/6

Applied Economics, 2005, 37, 51-55 RoutledgeTaylor 6. Francii Croup

On the link between Japanese ODAand FDI in China: a microeconomicevaluation using conditionallogit analysisSeverine BlaiseCEF/-CN RS, Facitlte des Sciences Econotniques, U niversite de laMediterranee, C EREF I, Facttlte d'Econom ie App liquee, Universitt;cl'Aix-Matseille III, 15-19 Allee Claude Forhin 13627Aix-en-Provence, FranceE-tnail: [email protected]

This study investigates the effectiveness of Japan's official developmentassistance in promoting foreign direct investments inflows in the case ofthe People's Republic of China. Conditional logit analysis using provincelevel statistics from 1980 to 1999, show s that Ja pa nes e aid Oows did h ave asignificant po sitive impact on private investors location choice even th oug hother profit-maximizing factors such as the level of economic activity hada leading spillover effect. In a context of growing scarcity of aid, the studyconcludes by asserting the importance of a complementary process inwhich foreign aid is aimed at enhancing the development of infrastruc-tures , acting as a pre-requisite for future direct investments. Finally, Japanproviding an interesting case study, we stress the need for a bettercooperation between public and private sectors in development assistanceprogrammes.

I. IntroductionEver since Japan emerged as the largest donor ofofficial development assistance (ODA) in volume, itsmotives and practice for giving aid got more andmore attention. Heavily criticized from both insideand outside Japan, Japanese aid policy has longbeen depicted as 'mercantilist ' and routinely accusedof 'unfair aid practices'. The concentration of aidfiows on Asian countries, the emphasis on capitalprojects for economic infrastructures, the stronginvolvement of Japan's private sector in developmentco-operation are the key features of the Japanese aidprogramme that have been under harsh scrut iny(Ensign, 1992).

The purpose of this study is to verify whetherJapanese aid leans on a better economic rationaliza-

tion of aid programmes by evaluating its effectivenessin promoting foreign direct investments (FDI). Inthe case of the People's Republic of China (Chinahereafter) the link between the Japanese public andprivate sector in development cooperation is investi-gated. Indeed, the underlying idea is that ODAprojects act as a prerequisite for future privateinvestment, notably through the development ofinfrastructures in recipient countries.China is one of the Asian countries that had themost intense relationship with Japan for cultural,historical and strategic reasons. Even though Japanhas always been distrustful of China, after theSecond World War the 1972 Sino-Japanese rappro-chement and the 1978 Sino-Japanese Peace andFriendship Treaty laid foundations for therapid development of bilateral relations. China

Applied Economics ISSN 0003-6846 print/ISSN 1466-4283 online 2005 Taylor & Francis Group Ltdhtlp://www.tandf.co.uk/journalsDOI: 10.1080/000368404200028153451


2/6

52 S. Blaisebegan its economic reform and open-door policy andfor the first titne showed a willingness to accept for-eign aid. As a consequence, the Japanese aid pro-gramme to China began much later than in otherAsian countries (see Story, 1987 for a survey). Eventhoug h, from 1979 Japan ese aid grew su bstantially:from 1982 to 1986 Ch ina w as the largest recipient ofJapanese aid and ever since it has been the secondlargest recipient next to Indonesia. ' In the meantime,China has been economically very important toJapan: the enormous potential of its market has ledtnany Japanese companies to invest in China, and thetwo countries rapidly became major trade partners.

The next section presents an econometric analysisaitned at verifying the role of ODA projects in thelocation decision of Japanese private investors usingconditional logit analysis. The literature on the eco-nomic itnpact of FDI is vast and this issue will not bediscussed here. However, following Ichikawa (1991),one could assert that FDI, unlike portfolio invest-ment, promotes the dissemination of valuable knowl-edge and entrepreneurship in the form of researchand development, production technology, marketingskills, managerial expertise, and so on. That is, directinvestment is a vehicle for the dissemination of theseinputs and services, not simply a provider of finance.Thus, direct investtiient - particularly in manufactur-ing - is more likely than portfolio investment to pro-mote econotnic growth. Therefore, one could expectODA to contribute substantial ly to the economicgrowth of a recipient country not only through itsdirect impact, but also indirectly by promoting theinflows of FDI.

tion lies in their restriction to tnacroeconomicframework. As pointed out by Mosley (1986), whileat the microeconomic level evaluation of aid effec-tiveness brings generally satisfactory results, macro-economic analysis remain quite atnbiguous (theso-called 'Macro-Micro paradox'). Moreover, in thecase of large countries such as China, because there isa great disparity in the geographical distribution ofthose financial fiows, it does not seem very relevant toconsider the country as a whole. Consequently, onewould rather perform a tnicroeconomic evaluation ofthe relat ionship between Japanese ODA and FDI,using province level statistics of the Chinese economyas well as detailed data on Japanese affiliates'activity. This evaluation is carried out on the period1980-1999. Statistics for ODA are issued frotn theJapan Bank for Internat ional Cooperat ion (JBIC)online database for loan aid, and ODA AnnualReport of the Ministry of Foreign Affairs for grantaid. ODA is referred to as the sutn of loan and grantaid, setting aside technical cooperation. Thosestatistics have been classified by provitice (calculationfrom the author). As far as FDI are concerned, ToyoKeizai Overseas Japanese Companies Data are usedfrom the 1990 and 1999 editions which also allowone to have a classification by Chinese provinceof Japanese affiliates activities. Statistics for eachChinese province are cotnpiled from ChinaStatistical Yearbook published by China StatisticalPublishing House, various volumes. This processwill allow one to have a tnore precise perception ofthe 'spillover effect' of foreign aid.

I I . MethodologyA nutnber of works was carried out on the relation-ship between ODA and FDI in the case of Japan.Nevertheless, the causal relation that has been eval-uated is rather the opposite. That is, to assert theimpact of FDI as a detenninant of ODA fiows.^ Asfar as we know, few studies (with contrasted results)have considered the possibility that ODA act as aprerequisite for future Japanese FDI.^ These studiessuffer a nutnber of shortcotnings ranging fromtnisspecification for some, to invalid estimationtnethodology for others. Yet, their cotnrnon limita-

ill. ModelThe location decision of a Japanese investor in Chinais analysed by means of a conditional logistic regres-sion in which ODA is introduced along with otherprofit-maximizing factors.

Conditional logit tnodel has been widely used inprevious empirical studies of location choice. Forexample Head et al. (1995) examined the agglomera-tion benefit and location choice of Japanese tnanufac-turing in the USA. Fukao and Yue (1997) analysedthe determinants of FDI by Japanese electronic firms.Belberdos and Carree (2002) also examined the loca-tion of Japanese Investment in China, focusing on

' ODA Annual Report, Ministry of Foreign Affairs, various years.^See for exatnple Teranishi (1983); Okamota and Yokota (1992)."^The International Development Center of Japan (1997); Yosioka et at. (1998); Inui (2000) and tnore recently Nakatnura et al.(2001).


3/6

Link between Japanese ODA and FDI in Chinaagglomeration effects, Keiretsu and firm heterogene-ity. Urata and Kawai (1999) studied the determinantsof the location of FDI by Japanese small andmedium-sized enterprises. ' ' The conditional logitmodel was first developed in economic analysis byMcFadden (1973).

The purpose of this study is to verify if, in a givenprovince, the amount of Japanese aid has an iiTipacton the location choice of Japanese private investors,for instance through the development of infrastruc-ture. To model the location of Japanese FDI inChina, it is assumed that Japanese firms undertakeFDI in a ptovince where they can iTiaximize theirprofit after evaluating relevant characteristics ofalternative locations. By assuming that the firm hasa production function of Cobb-Douglas form, let usdescribe the profit (JT) of firm / obtain ed from under-taking F DI in province y as Equa tion 1.TTjj ^QA ]J . . . afiiAifij .C (1)

where a^,..., , are unknow n p arame ters, X^(s=\,... ,m ) are variables describing the characteris-tics of the province j (f=\,...,ti), an d /x,y is arandom disturbance term capturing province- andinvestment-specific heterogeneity in total factorproductivity.Giv en profit E qu atio n 1, if and o nly if jXij is dis-tr ibuted as Type I extreme value (independent ran-dom variable) according to the Weibull distribution,then the probability that provincey will yield investor/the highest profit among all the provinces is given by

the logit expression (Equation 2) (McFadden, 1973).ex p

7=1

(2)

One expresses the number of FDI selections made byJapa nese firm /' in the p rovince j as W jj ( / = 1 , . . . , //).This dependent variable takes the value 1 if the prov-ince is selected by the investor and 0 otherwise.Finally, one obtains the probability of observingsuch FDI pattern as Equation 3.(3)

The para me ter (c/ o,. . . , c/,,,), which indicate thecharacteristics of potential host provinces toJapanese FDI, are estimated by the maximum

53likelihood method, which maximizes the likelihoodfunction (Equation 3).

In other respects, as pointed out in previous studiesthe location choice criteria may vary across differentsectors. Therefore, the estimation is carried out inboth manufacturing and noti-tnanufacturing sectors.The following variables enter into consideration asindependent varia-bles:

- Agglomeration effect: economic activities ofexisting Japanese firms in one province thatgenerate positive externalities for nearby firmengaged in similar activity. This is tneasured byth e numb er of e.xisting Ja panese affiliates beforethe venture began operation {JAtnc and ./Antnc,respectively for lTtanufacturing and non-manufacturing sector). Note that this numberis used plus one in order to be able to use thelog of the variable.

- Level of economic activity: economic size ofthe provinces is controlled for by including theper capita GDP (GDPc). Indeed, the larger theeconomic size of a region, the tnote likely thatit will receive foreign investments.

- Production cost: labour cost is recognized as oneof the most important factors. It is given by theaverage wage level of workers (Wage), which isexpected to discourage investtnent especially inthe case of manufacturing location.

- Infrastructure: a well developed transportationinfrastructure reduces the costs of itnportinginputs and exporting or distributing output aswell as a good commu nication infrastructurefacilitates and reduces the cost of cotnmunica-tion of affiliates. Consequently, the infrastruc-ture indicator is expected to have a positiveimpact on the location decision of privateinvestors.

The dtitntny Coast (C) is also introduced,which ta kes the value 1 if the region is locatedon the coast side and 0 otherwise, as well asth e distance from Japan (DIST).A measure of the quality of telecotnmunica-tion infrastructure is also included: the nutnberof long-distance telephone lines (Telecotn).Finally, the eutnulative atnount of Japanese aid(ODAc) is also introduced, which is also sup-posed to reinforce the developtnent of infrastruc-ture, especially in the transportation sector.

- Human capital: the level of education is expectedto have a positive impact on the location deci-

"* Other examples can be found in Fukao (1996), Fukao and Tei (1996).


4/6

54 S. Blaisesion especially in non-manufacturing sectorsas it enhatices the quality of human capital.The education level is measured by the numberof .stuiletits enrolled in senior middle schools as apercentage of population (EDU), The log of allvariables is used, except for the dummy C as wellas the Distance variable.

IV. ResultsResults of the conditional logistic regression are givenin Tab le 1, Th e agg lom eratio n effect an d the level ofeconornic activity appear to have the largest impacton location choice by Japanese investors in manufac-turing as well as non-manufacturing sectors. For bothsectors the two variables are strongly significant,holding the highest positive coefficient (especiallyfor non-tnanufacturing activities).

As expected, high wage level strongly discouragesJapanese investors in the manufacturing sector. Thisshows that Japanese affiliate activities in labour inten-sive sector took advantage of China's cheap labour.This finding is corroborated by the significant nega-tive itnpact of education level in the same sector. Innon-tnanufacturing activities, wage level also holds anegative coefficient, but the variab le is less sign ificant.Education level is positive but not significant. It canbe suggested that in the later sector, investors givemore importance to qualified labour force.The amount of Japanese ODA attributed to eachprovince is strongly significant in the two sectors and

Table I. Conditional (fixed-etfeets) logistie regressionVariableCLJAcLODAcLGDPcLWAGELEDUDISTLTELECOMLog likelihood

Manufacture0,297***(3,245)0,623***(9,552)0,129***(3,880)0,638***(4,514)-0 ,67 9 * * *(-3,617)-0 ,63 5 * *(2.218)-0 ,174***(-2,599)0,285***(3,823)-4147,6929

Non-manufacture- 0 , 0 4 5(-0.275)0.605***(6.872)0.189***(2.715)

1.220***(4.829)- 0 , 6 5 1 * *(-1,923)0,39(0,745)-0 ,2 9 9 * *(-2,160)0,399***(3,737)-1290,6864*, **, *** tnark the results which are respectively 10%, 5%,1% significant.

confirms that aid has a spillover effect on the locationdecision of Japanese investors. This positive effectis shown to be slightly more important for non-manufacturing activities. Indeed, ODA projectshave been concentrated mainly on economic infra-structures, considered as high priority by theChinese governm ent itself, in order to resolve seriousbottlenecks in the econotny (Soderberg, 1996), Thisis also proved by the impact of infrastructure indi-cator such as telecommunication facilities. This hadan important 'spillover effect ' on promoting Japaneseinvestors' activities. As far as the loan component ofODA is concerned (which constitutes the bulk ofJapanese aid to China), this finding confirms theidea that the incentive for future private investmentand technology transfer not only allowed a high rateof reimbursement of loan aid, but that the duty toreirnburse constitutes an indicator of solvabilitywell perceived by private investors (Teboul andBassino, 1997),

Furthermore, i t is found that manufacturingactivities tend to locate in coastal areas, which offerconsiderable transport facilities for itnportinginputs and export ing production output . However,this factor is not significant in the location decisionof non-manufacturing activities. Finally, the distancefrotn Japan is proved to have a negative itnpacton FDI location in both sectors, still favouringnorth and coastal provinces.

V. ConclusionThis econometric analysis allowed one to supportthe view that Japan's ODA has been quite effectivein promoting Japanese FDI in China, Even thoughother profit-maximizing factors such as the level ofeconomic activity or the agglotneration effect ofJapanese firtns had a leading role in location decisionof Japanese investors, the allocation of aid projectsdid have a significant positive impact, Japanese ODAprogrammes in China have been mainly allocated toinfrastructure projects (especially in transportations),paving the way for future private investment. InJapan, this complementary process is generated byboth institutional and infonnal links between thepublic and private sectors and appears to be substan-tially beneficial to the recipient country. In a contextof growing scarcity of aid, a detailed analysis of thiseffective framework of develop men t co oper ationcould be an enriching and useful exercise.


5/6

Link between Japanese ODA and FDI in ChinaAcknowledgementsI would like to express all my gratitude to professorKyoji Fukao for his precious advice and commentsregarding this study and to professor Ding Jianpingfor generous provision of Chinese statistics.

ReferencesBeiberdos, R. and Carrcc, M. (2002) The location ofJapanese investments in China: agglotTieration effects,Keiretsu, and firm heterogeneity, Jourtial of theJapanese and International Econotnies, 16(2), 194-211.Ensign, M. M. (1992) Doitig Good or Doitig Well? Japan'sForeign Aid Progratn, Colombia University Press,New York.Fukao, K. (1996) Kokunaika Kaigaika-WagakuniSeizogyo no Richi Sentaku ni kansuru Jisho Bunseki[At home or abroad? Empirical analysis regardinglocation selection by Japanese companies], KeizaiKenkyu, 41(\), N I.Fukao, K. and Teil , L. (1996) Chokusctsu Toushi Sakikokuno Kcttei Youin ni tsuite [Concerning the decisionfactors of direct investor countries]. Financial Review,Finance Research Center of MOF, Tokyo.Fukao, K. and Yue, X. (1997) Denki maker no richi sen-taku [The location selection of Japanese electronicfirms], Mita Gakkai Za.sshi, 90(2).Head, K., Ries, J. and Swcnson, D. (1995) Agglomerationbenefits and location choices: evidence from Japanesemanufacturing investments in the US, Journal ofInternational Econotnies, 38, 223-47.Ichikawa, H. (1991) Japanese direct investment and devel-opment finance, in Yen for Development (Ed.) S. Islam,Council on Foreign Relations Press, New York.International Development Center of Japan (1997) Th eRole of Japan's ODA for Asian Countries.Inui, T. (2000) Characteristics of Overseas Activities ofJapanese Companies in Developing Countries,Symposium for Aid Effectiveness, November 20th,Economic Planning Agency, Tokyo.JICA (1999) The Second Country Study for Japan's OfficialDevelopment Assistance to the People's Republic ofChina.JBIC (2001) JBIC ODA Loatis to the People's Repuhlic ofChina.

55McFadden, D. (1973) Conditional logit analysis of qualita-tive choice behavior, in Frontiers in Econom etrics (Ed.)P. Zaremka, Acadetnic Ptess, New York.Mosley, P. (1986) Aid effectiveness: the Micro-MacroParodox, Institute of Development Studies Bulletin,17(2), 22-7 .Nakamura, O. , Shishido, S. , Minato, N. , Kutokawa, M.and Kawakami, A. (2001) Chugoku to Indonesia ni

taisuru wagakuni no ODA no hikaku kenkyu: ODA-FD I wo fukwnu tahumongata tnodelu hunseki [A cotn-parative analysis of Japan's ODA to China andIndonesia: a multi-sector tnodel analysis of ODA andFDI] , The Japan Society for InternationalDevelopment, 2001 National General Meeting.Okamoto, Y. and Yokota, K. (1992) Nihon no EtijoSeisaku no Suuryou Bunseki-Nikoku Et i jo noKunibetsu Haibun Hiritsu no Kettei , [Etnpirical ana-lysis of Japan's Aid Policy-Dctcrtninants of theCountry Allocation of Japan's bilateral ODA],Kokusai Kaihatsu Kenkyuu, 1(2).Soderberg, M. (1996) 77)e Busittess of Japanese Foreign Aid:Five Case Studies frotn Asia, Rout ledge, London.Story, G. (1987), Japan's official developtnent assitance toChina: a survey. Pacific Econotnic Paper, N 150,Australia-Japan Research Center.Tcboul, R., Bassino, J. P. (1996) La contribution dc l'aidepublique et des investissements directs Japonais a lacroissance des pays en deveioppetiicnt dc l'Asie deI'Est, Seminar Strategie de croissance et marchesetnergents organized by the 'Caisse des Depots ctConsignations' , the Development Center OECTD, theCEFI, the CEPII, the 'Direction des RelationsEconomiques Exterieures' (DREE), the 'Ministerc del 'Economie et des Finances' and the RcvucEconomique, Ho Chi Minh Ville, Novetiibcr 1996.Teranishi, O. (1983) Waga Kuni no Seifu Kaihatsu EnjouSeisaku ni tsuite, [On the Japanese Foreign Aid

Policy], Discussion Paper Series, N 8 3 , The Instituteof Economic Research, Hitotsubashi Univctsity,Tokyo .Urata, S. and Kawai, H. (1998) The Detertnitiants of theLocation of Foreign Direct Investtnent by JapaneseSmall and Mediwn-Sized Enterprises, The EPAEconomic Research Industry, Tokyo.Yoshioka, S., Moro, K. atid Sawada, M. (1998) ODA atidAsian FDI Itiflow: An Etnpirical Analysis in SeleetedSoutheast Asian Countries, The EPA EconotniicResearch Industry, Tokyo.


6/6

05_On the link between Japanese ODA and FDI in China

Documents