A study on the spillover effect of FDI on literature review

是 否 +2 论坛币

k人 参与回答

经管之家送您一份

应届毕业生专属福利!

求职就业群

赵安豆老师微信：zhaoandou666

经管之家联合CDA

送您一个全额奖学金名额~ !

立即领取

感谢您参与论坛问题回答

经管之家送您两个论坛币！

+2 论坛币

A study on thespillover effect of FDI on literature review

      Abstract:

ever since 1970s, FDI has become hotissue in the research field of international investment, based on thedefinition of spillover effect and some literature reviews, this paper summarizedfive channels of spillover effect and makes some comments relevant on the existenceand determinant factors of FDI spillover.

      Key words:

FDI, spillover effects, channels,determinant factors.

     

扫码加我 拉你入群

请注明：姓名-公司-职位

以便审核进群资格，未注明则拒绝

分享0 收藏1 回帖

关键词：literature Spillover Effect Review study literature investment research channels relevant

     I. Introduction

Spillover, also called as technologyspillover or knowledge spillover in foreign literature, FDI (Foreign DirectInvestment) spillover effect means the effect takes via multinationalinvestment. MacDougall (1960) firstly put forward the entrance of multinationalcompany will cause external economics for host country, that is , spillovereffect.

Blomstrom and Kokko (1997, 1998) callthe external economics as productivity spillover. They claim that the spillovereffect come into being when the entrance of multinational company leads tonon-voluntary diffusion through various channels, it could be divided intopositive spillover(multinational companies’ entrance promote the productivityof the host country ) and negative spillover(multinational companies’ entrancedecline the productivity of the host country).

      II. The channels of spillover effect.

Caves (1974) says there are 3channels: in the first place, the entrance will promote the monopoly, strengthenthe free competition, and optimize the allocation of resources; in the second,the competition effect and demonstration effect of Multi-National Corporationwill increase the production efficiency of the host; in the third, Multi-NationalCorporation's entering speeds up the transfer of technology and innovation. Kokko(1994) alleged there are four main method: 1.demonstration/imitation effects, localenterprises improve their technology by imitation; 2. Linkage effect, multinationalcorporations set up forward linkage or forward linkage, and assist thetechnical innovation; 3. Training effect, multinational  ones will train the worker and the managersfor a better adjustment to their own company, and those human resources mayflow into the local enterprises. 4. Competition effect, the foreigners applypressures on the host company, which will promote their productivity via hightechnology. Gorg and Greenway (2004) as well as GreaPo and Fontours (2007) putforward the fifth channel: export.

   

  III. Empirical test of spillover effect            1. Empirical test of spillover effect for Intra-industry

Caves (1974) first tests the existenceof spillover effect with measurement method, Using regression method, he makes fulluse of cross section data from the Australian industry to test the existence ofspillover effect, and finds that the local enterprises industrial added valueper capital is into proportional to the Multi-National Corporation employeesaccounted for industry employees, confirming the existence of spillover effect.

Globeman (1979) use the same way andthe same variables as Caves (1974), that is, the local enterprises industrialadded value per capital for the dependent variables, while capital, humanresource, and control effects from foreign enterprise for variables. He useregression method to analyze the cross section data from the Canada industryduring the year from 1971 to 1972, and finds the proofs of the existence ofpositive spillover.

Blomstrom and Persson (1983) analyzethe cross section data from the Mexico industry, find positive spillover, too.

Table1. The channel and origin of spillover effect:

Channels of spillover	The origin of the spillover
Imitation effect	Imitation of the high technology    Imitation if the advanced management
Competition effect	Reduction  of X- non efficiency    Reduction  of monopoly & promotion of efficiency
Human resource flow	“win-win ” communication about technology    Flow of the covered knowledge
Linkage effect	Assistant of the local enterprises, backward linkage    Supplication of intermediate goods, forward linkage
Export effect	Pooling export experience    Learning advanced technology;

Resources: basedon reference [6] and [7]

Subsequently, many scholars doresearch in line with the measurement method put forward by Gorg and Greenway(2004). After reviewing all the literature testing intra-industry spillovereffect during the year of 1974 to 2002, it could be concluded that all thescholar has the same measurement thoughts---using the production efficiency orthe total factor productivity as the dependent variable, making regression withexplanatory variables of FDI participation degree and some of its variables. Ifthe coefficient of the participation degree is positive, positive spillovereffect exist, otherwise, negative spillover effect. Of the whole literature, 16are based on cross data, 2 based on industry panel data, 24 based on companypanel data. And they think the panel data is more efficient than cross data. Sothere are only 7 literatures focusing on developed countries and signalingpositive spillover. For the developing countries, all are negative spillover.

However, Kolasa (2008) gets positivespillover with panel data of companies in Poland, with panel data of Chinesefrom 1995 to 1999, Liu (2008) finds that the spillover is negative in shortterm while positive in long term. Chang and Xu (2008) holds the view that it’sis unbelievable with the local productivity as the independent variable, forthe entrance of foreign  enterprises willlock out the local companies with low productivity but save the ones with highproductivity.  Thus, although the resultsare always positive, they can’ t be for the existence of spillover effect. Theyuse Chinese industrial census data during 1998 ---2005 to test the FDI overflowdata level of countries and regions, They used local enterprises exists as thedependent variables, (two variables, survival for 1, death for 0), With the FDIindustry employees proportion as a variable and some control variables asdiscrete time logistic regression, Results FDI positive spillover effects isfinally found in the analysis of national level.

     

        2. Empirical test of spillover effect for Inter-industry

On the contrast to spillover effect forInter-industry, the test is relatively later; the argument is relatively less,and FDI inter-industry spillover effect appears positive in national level.

Javorck (2004) test inter-industryspillover effect with the Statistical panel data of LiWanTao during 1996 to2000, the result notices that the backward linkage has positive correlationwith productivity, while forward linkage have no positive correlation, However,instead of directly using the related FDI intermediate value to measure theresult, he takes approximation method in the model to measure backward andforward FDI contact index, comprehensively analyze the weighted average of FDIindustry participation, the results might be some errors. In addition, Blockand Gertler (2003 ) Harrison and Robinson (2004), Liu (2008) test the firmlevel panel data respectively by China ,British and Indonesia, and the interindustry spillover effect is found positive. for the choice of FDI forward andbackward linkage variables, the method is basically the same.

Kugler 2006 noticed the contactdegree of FDI forward and backward linkage is approximately standby theweighted average of industry FDI participation. Besides, the research methodfor the local enterprise productivity regression is not convincing. To studythe cross-industry and inter-industry spillover effect, he establishes a multi-sectorDynamic stochastic equilibrium model. The model is a competitive equilibriummodel which endogens the technological change, and allows “dry middle school”effect to occur in both inter-industry and intra-industries. With the help ofthe model, the panel data of Columbia companies during years 1974-1988 test thedart of spillover effect .Results show that the spillover effects occur ininter-industry rather than the intra-industry, and backward linkage is the mainchannel of spillover effect.

According to the test, the intra-industryspillover effect test results are controversial, while has more consistentresults. On one hand, the test for inter-industry spillover effect is concentratedin the early time, because of the data and measurement limitations, Manystudies have not found positive spillover effects, even they find some negativespillover effect in the end. In addition, Kokko (1996) considered that thesestudies ignored the nature of different industries, In the industry of bigtechnology gap and high foreign share, the economy scale of foreign capitalenterprises will have a crowding out effect on the host country enterprise, onthe other hand, Some scholars believe that there is no spillover effects in intra-industry.Javorcik (2004) thinks that these scholars may be looking for the spillovereffect in the wrong way, for the local competitive enterprises, Multi-NationalCorporation will maximize to prevent the technology spillover to their competitors;spillover effect brought by the flow of human resources is not obvious inpractice. For the Multi-National Corporation will attract excellent talents withhigh salaries. As to the local supplier or the attachments, considering thepurchase cost-saving and production efficiency improvement, Multi-NationalCorporation tends to enhance their technical capacity. competitive supplysystem adopted by the Multi-National Corporation will also urge the localsuppliers to improve production efficiency. Therefore, the spillover effect ismore likely to occur in inter-industry.

      3. Empirical test ofspillover effect for export
The initial research on spillovereffect of export is very late, related literature is scarce, Aitken (1997) set foodin the field earliest, he used the Bangladesh garment industry case toillustrate the existence of spillovers. In the absence of Multi-NationalCorporation, the country's apparel exports are very small, since a South Koreanclothing Multi-National Corporation enters the country, a large number of localexport-oriented clothing enterprises shoot as bamboo after a spring rain, Fromthen on, the clothing industry plays an important role in the country'sexports. They believe that the Multi-National Corporation entry will spread informationto the host country enterprise, The Multi-National Corporation know overseasmarket, customer demand as well as the technology frontier,. Local businessesget the information through the Multi-National Corporation, at the same time,local enterprises can get the distribution channels by the way of cooperating withMulti-National Corporation, thereby reducing the risk of export, increasing thelikelihood of export. They use two stage Probit model to test the industrialpanel data from Mexico, the results found a significant positive correlation betweenthe geographical distance and the possibility of export.
Kokko ( 2001) researches exportspillover effect caused by Multi-National Corporation on two different tradeoriented stage on the Uruguay （import substitution  trade form before 1973 and export orientedtrade model after 1973.  The empiricalstudy found that in the background of export-oriented trade, the Multi-NationalCorporation will have export spillover effect on local enterprises; TheMulti-National Corporation will produce export spillover effect of localenterprises, exports incline to increase. Besides, Greenway’s (2004) empiricalresearch on the British also finds export spillover effect caused by theMulti-National Corporation.
      

     IV. The factors ofspillover effect
Early the studies represented byCaves (1974) and Globerman (1979) are only concerned about whether there isspillover effect, while ignoring what factors will affect the spillover effect,and what kind of environment is conducive to the spillover effect. In recentyears, some scholars shift their research focus to the factors of spillovereffects, these studies suggest that the Multi-National Corporation investmentmotivation, the absorptive capacity of the host country, The FDI property, thehost country enterprise property and industry agglomeration effect will affectthe spillover effect of FDI.
      1. Investment motivation of Multi-National Corporation
The traditional FDI theory believesthat the Multi-National Corporation have a comparative advantage over domesticenterprises in technology, so there will be competition effect and demonstrationeffect, Its theoretical basis is the theory of monopoly advantage by hymer andKindleber, The theory claims that the transnational operation of enterpriseshave some special property, when the host country market is incomplete, theMulti-National Corporation may use its monopoly competition, maintain themonopoly price to obtain excess profits, which could be called astechnology-exploiting FDI. The monopolistic advantage theory can explain theFDI in developing countries from developed countries , but for the FDI in developedcountries from the developing countries the theory can’t fully explain it. Sothere may a technology-sourcing FDI, aiming to get advanced technology from thehost countries..
Driffield and Love (2002) believe thatFDI investment motives will influence the size of FDI spillover effects, spillovereffects brought by technology-sourcing FDI should be small, they inspect theEngland panel data (1984 -1995), find a significant positive correlation betweentechnology-exploiting FDI and local business productivity, a significantnegative correlation between technology-sourcing FDI and local business productivity.To distinct the two different types of FDI, they compare R & D intensity ofeach national Multi-National Corporation with R & D intensity of localenterprises. if R & D intensity of each national Multi-National Corporationis larger than the R & D intensity of local enterprises, the Multi-NationalCorporation should be called technology-sourcing FDI, otherwise, technology-exploitingFDI.Then the Girma (2005) based on Driffield and Love (2002), using threshold regressionanalysis (threshold regression analysis) test method for the British panel dataof 1989---1999, once again confirmed the above results.
      2. The allure of host country
The allure of the host country willhave effect on the spillover effect. BlomstrǒM, Globerman and Kokko (1999) holdthe view that the more attractive the host country , the larger the spillover effect.So the Smaller technology gap with the Multi-National Corporation, the greater isspillover effect. Wang and Blomstrom (1992) think that the bigger the technologygap, the greater the spillover effect, because this will provide host countryenterprise with opportunity of learning advanced technology from Multi-NationalCorporation.
Girma (2001), through the empiricalresearch on the British, finds the spillover effect of FDI does not exist withoutconsidering the absorption ability. But in consideration of differentabsorption capacity, local enterprises of strong absorption ability obtainpositive spillover effects, and local enterprises of the weak absorptionability have no spillover effects and spillover effects is negative. In theirmodel, absorptive capacity is measured by the initial total factor productivityand the corresponding industry average TFP gap.
When Haleg and Longc （2006） made the empiricalresearch on Chinese, they use the ratio of initial single enterprise's totalfactor productivity and the industry's most complete element productivity tomeasure the initial technology absorption of local enterprises , they found ifthe initial absorption capacity exceeds a critical value, the spillover effectis positive, otherwise, the spillover effect is negative.
Girma (2005) uses more advancedthreshold regression analysis to analyze the UK data, which ranges the absorptionability to cause spillover effect naturally, results show that when the absorptionability are within a certain range, the spillover effect is positive, butbeyond this range, the spillover effect does not exist, or below the lowerlimit, spillover effect is negative.
      3. The FDI property and host country property.
For the FDI property, scholarsgenerally consider two main aspects, one is the source of FDI, the other FDIentry modes, namely joint ventures, acquisitions or sole way to enter the hostcountry,; and for the host country enterprise nature, different research emphasesdifferent things, so it could be divided in various way.
Belderbos Capannelli and Fukao (2001)agree that, in many countries, especially developing countries, encourageMulti-National Corporation to purchase in the local procurement, and the policydefining the minimum local sourcing proportion will promote the backwardconnection between the Multi-National Corporation and local enterprises, thiswill promote the technologic spillover effect of Multi-National Corporation,they made empirical test on local procurement done by Japanese electronicscompanies in different countries, although the regulations that the theMulti-National Corporation must purchase in the local company can promoteJapanese purchase proportion, however these local sourcing mostly flow to localsuppliers in Japan, rather than the local suppliers in host country.
Buckley (2007), via China industrialstatistics data in 2001, verifies the spillover effect that different kinds of FDImade on different nature of host country enterprises, in accordance with theFDI technology, FDI is divided into two types, one is the FDI in Hong Kong andMacao area, where the Multi-National Corporations have advantage of standardtechniques and labor-intensive production, the other is the FDI in westerncountries such as, USA, Europe, Japan, where the Multi-National Corporation hasthe advantage of new technology, and latest R & D, so their ownershipadvantage lies in technology intensive industry, at the same time, t the hostenterprises are divided into labor-intensive enterprises and technology-intensiveenterprises or state-owned enterprises and nonstate-owned enterprises. Theempirical results show that, FDI from Hong Kong and Macao have larger spillovereffect on labor-intensive industry of local enterprises, and FDI from thewestern mainly have spillover effect on technology intensive industries in thelocal enterprises.
Javorcik (2008) argue that, when Multi-NationalCorporation and local enterprises the format a joint venture, in order toprevent the advanced technology spillovers to local competitive enterprises,generally technology transfer from the parent company to the joint venture, isrelatively non advanced technology, joint ventures in host can easily access tothe technology and absorb it, but the parent company is more likely  to transfer advanced technology to ownedenterprises, because this technology is not easy to overflow and not easily absorbedby local enterprises, so the spillover effect of joint venture is larger thanthe wholly owned enterprises. In addition, the joint venture use relativelysimple technique, it’s more likely for them to find proper local suppliers,besides, the host country party is more familiar with local suppliers,  costs for searching for appropriate localsuppliers reduces. Therefore, considering the cost, the joint ventureenterprises are inclined to corporate with local suppliers, thus, compared withthe wholly owned enterprises, joint ventures have larger spillover effect.
      4. Industrial agglomeration
In the trend of globalization, FDI playsa decisive role in the economy of the host country, Multi-National Corporationhave gradually infiltrated into the local industrial clusters, and even somelocal industry cluster is formed in the direct promotion of Multi-NationalCorporation, geographical proximity promote the interaction between theMulti-National Corporation and local enterprises, spillover effect of clustereconomy makes the FDI spillover effect more likely, Jordaan (2005) think the spillovereffects generated within both intra-industry and inter-industry will becomemore and more obvious through demonstration, imitation, human capital flow. Econometrictest on Mexico industry statistical data showed that FDI spillover effect ofthe geographical concentration of industries is larger, while Chang and Xu (2008)hold the view that the crowding out effect is greater than the spillover effectin the region. in the inspection of Chinese panel data, they found regional FDIhave significantly negative correlation with the survival of local enterprises,so, in this region, the crowding out effect on local enterprise is more obvious.
   

V. Brief review

At present, the focus of domestic andforeign scholars have been from the research of the existence of spillovereffects to the influence factors of technology spillover effects, with thedifferent influence factors, whether there is spillover effect on host countryenterprises made by Multi-National Corporation, selection of the panel data andthe measurement model also makes the analysis accuracy increase continuously. Inaddition, research perspective also continues to expand, such as Chang and Xu (2008),the first inspect the interactional effects of foreign and domestic enterprisesfrom the perspective of strategic management, Yuan Cheng and Lu Ting (2005)research FDI "training effect on private entrepreneurs in China throughthe survey data ". The study breaks through the traditional framework by Caves(1974) and Globerman (1979), convince constantly enhanced. It will haveimportant implications for FDI spillover effect theory for further research tostep into the traditionally economics dominated the field, combined with themethod of case study.

Reference

[1] MacDougall G. The Benefits andCosts of Private Investment From Abroad: A Theoretical Approach [J], EconomicRecord, 1960, 36:13-35

[2] Blomstrǒm M and kokko A. The impactof foreign Direct Investment On Host Countries: A review of the EmpiricalEvidence [R], World Bank Polly Research Working Paper, 1997

[3] Blomstrǒm M and Kokko A.Multinational corporations and spillover [J], Journal of Economic Survey, 1998,12 (2) : 247-277.

[4] Caves R E. Multinational Firms. Competitionand Productivity in Host-Country Markets [J]. Economic Record, 1974, 41 (162) :176-193.

[5] Kokko A. Technology, MarketCharacteristics, and spillover. Journal of Development Economics. 1994, 43 (2)292-293.

[6] Gorg H & Greenway D. Much Adoabout Nothing? Do Domestic Firms Really Benefit From Foreign Direct Investment?[J] World Bank Research Observer, 2004, 19 (2) : 171-197.

[7] Crepo N & Fontourn M P.Determinant Factors of FDI Spillover-What Do We Really Know? [J]. WorldDevelopment. 2007, 35 (3) 410-425.

[8] Globerman S. Foreign DirectInvestment and Spilover Efficiency Benefits in Canadian ManufacturingIndustries. Canadian Journal of Ecomomics. 1979, 12 (1) : 42-56

[9] Blomstrǒm & Persson. ForeignInvestment & spillover Efficiency in an Underdeveloped Economy, Evidencefrom the Mexican manufacturing [J]. World Development. 1983, 11 (1) :493-501.

[10] Kolasa M. How does FDI inflowaffect productivity of domestic firms? [J]. Journal of International Trade& Economic Development. 2008, 7 (1): 155-173

[11] Liu Z. Foreign Direct Investmentand Technology Spillover: Theory and Evidence [J]. Journal of Development Economics:2008, 85: 176-183

[12] Chang S & Xu D. Spilloverand Competition among Foreign and Local forms in China [J]. Journal ofStrategic Management. 2008, 29 (5): 495-625

[13] Javorcik B S. does ForeignDirect Investment Increase the Productivity of Domestic Firms? [J]. AmericanEconomic Review. 2004. 94 (3): 605-627.

[14] Kugler M. Spillovers fromForeign Direct Investment: Within or Between Industries? [J]. Journal ofDevelopment Economics. 1997, 43: 103-132.

[15] Kokko A. Productivity Spilloverfrom Competition between Local Firms and Foreign Affiliates [J]. Journal of DevelopmentEconomics. 1996, 8 (4) 517-530

[16] Aitken B. Hanson G H &Harrison A E. Spillovers, Foreign Investment, and Export behavior [J]. Journalof International Economics. 1997, 43: 103-132.

[17] Kokko A. Zejan M & TansiniZ. The Regimes and Spillover Effect of FDI: Evidence from Uruguay [J].Weltwirtschaftliches  Archiv. 2001, 137(1): 125-149.

[18] Greenway D, Sousn N &Wakelin K. Do Domestic Firms learn to Export from Multinationals? [J]. EuropeanJournal of Political Economy. 2004, 20: 1027-1043.

[19] Girma S Absorptive Capacity andProductivity Spilovers from FDI: A Threshold Regression Analysis [J]. OxfordBulletin of Economic and Statistics.2005, 67 (3): 282-306.

[20] Hale G & Long C. WhatDetermines Technological Spillover of Foreign Direct Investment: Evidence FromChina [R]. Economic Growth Center, Yale University Working Paper Series No.934,2006.

[21] Girma S, Greenaway D &Wakelin K. Who Benefits Foreign Direct Investment in the UK? [J]. ScottishJournal of Political Economiy. 2001, 48 (2): 119-133.

[22] Belderbos R, Capnnellin G &Fukao K. Backard Vertical linkages of Foreign Manufacturing Affiliates:Evidence From Japanese Multinationals [J]. Development. 2001, 29 (1): 189-208

[23] Buckley P J, Wang C, Clegg J.the impact of Foreign Ownership, Local Ownership and Industry characteristic onSpillover Benefits from Foreign Direct Investment in China[J]. InternationalBusiness Review. 2007, 16: 142-158.

[24] Javorcik B S& Spatareanu. Toshare or Not to Share: Does Local Participation Matter for Spillover fromForeign Direct Investment? [J] Journal  of Development Economics. 2008, 85: 194-217.

[25] Thompson E R. Clustering ofForeign Direct Investment and Enhanced Technology Transfer [J]. WorldDevelopment. 2002, 30 (5): 873-889.

谢谢分享