categorizing zhou guillen - management department

CATEGORIZING THE LIABILITY OF FOREIGNNESS:OWNERSHIP, LOCATION, AND INTERNALIZATION-SPECIFIC DIMENSIONSNAN ZHOU1,2* and MAURO F. GUILLEN3

1China Minsheng Bank, Beijing, China2Nankai Business School, Nankai University, Tianjin, China3Lauder Institute, Management Department, Wharton School, University ofPennsylvania, Philadelphia, Pennsylvania, U.S.A.

Plain language summary: This study explains the different types of additional costs faced bymultinational firms when they invest abroad. We identify product adaptation cost,discrimination cost, governance cost, and appropriation hazard as important sources of thedifficulties that multinationals face. We then link each of these costs to different dimensionsof cross-national distance, including economic, cultural, demographic, political, andadministrative distance. Finally, we show that the importance of these different distancedimensions differ when multinationals invest abroad for different reasons including marketseeking, efficiency seeking, knowledge seeking, and natural resource seeking.

Technical summary: This study bridges the literature on the liability of foreignness (LoF)and cross-national distance. Following the ownership-location-internalization paradigm,we categorize the LoF into three dimensions: ownership-specific LoF, location-specificLoF, and internalization-specific LoF, each of which reduces a specific advantage of themultinational firm. We first define the three dimensions of the LoF and then argue thateach dimension is related to different types of costs. We further explain that different typesof costs are associated with different dimensions of cross-national distance, and firmsface different types of the LoF and associated costs when they conduct different types offoreign direct investment. We test these hypotheses in the context of Chinese listed firmsinvesting abroad, finding support for most of the predictions. Copyright © 2016 StrategicManagement Society

INTRODUCTION

The concept of the liability of foreignness (LoF) is thebackbone of the fields of international strategy andinternational business. Stephen Hymer was one ofthe first scholars to recognize that multinationalenterprises (MNEs) face extra costs when doingbusiness in foreign countries compared with local

competitors because of their lack of familiarity withlocal conditions (Hymer, 1960; Hymer, 1976). TheLoF is commonly defined as ‘the costs of doingbusiness abroad that result in a competitive disadvan-tage for an MNE subunit … broadly defined as alladditional costs a firm operating in a market overseasincurs that a local firm would not incur’ (Zaheer,1995: 342–343).

The LoF has great impact on multinationals.Previous studies have found that it influences theinternationalization strategy of multinationals, includ-ing the entry mode and performance implications ofinternationalization (Chen, 2006; Mezias, 2002).Given the importance of the LoF in global strategy,

Keywords: liability of foreignness; foreign entry; distance; China;OLI paradigm*Correspondence to: Nan Zhou, China Minsheng Bank, No.13Zhichun Road, Hang Nan Building, Beijing, China, 100083.E-mail: [email protected]

Global Strategy JournalGlobal Strategy Journal, 6: 309–329 (2016)

Published online in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/gsj.1140

Copyright © 2016 Strategic Management Society

scholars have tried to understand its various aspects,such as its definition, source, and how to overcome it(Eden and Miller, 2004; Mezias, 2002; Zaheer, 1995).

Because the LoF is a complex concept, one keystream within the LoF literature is to identify differenttypes of the LoF. For instance, Eden and Miller(2004) broke down the LoF into three specifichazards: unfamiliarity hazards, discriminationhazards, and relational hazards. Our article falls intothis literature by introducing a new categorization ofdifferent types of the LoF.

Previous studies in this area usually lay out thecategorization without specifying its theoreticalfoundation. This study adopts the ownership–location–internalization (OLI) paradigm to categorizethe LoF into ownership-specific (O), location-specific(L), and internalization-specific (I) LoF. Although theOLI paradigm has been applied to explain variousaspects of internationalization (Dunning and Lundan,2008a; Fiss and Hirsch, 2005), it has not been adoptedto understand the LoF. Because the OLI paradigm isone of the most well-accepted frameworks in the inter-national business literature (Eden and Dai, 2011;Portugal Ferreira et al., 2011), adopting the OLI para-digm ensures that our categorization considers thedominant factors that influence the internationaliza-tion of multinationals. Further, the OLI paradigmhighlights the advantages that encourage firms to goabroad. It corresponds well with the concept of theLoF, which describes the disadvantages multina-tionals face when they go abroad.

Empirically, many studies on classifying differenttypes of the LoF (Bell, Filatotchev, and Rasheed,2012; Moeller et al., 2013) are largely theoreticalwithout providingmeasurements or empirical support.By measuring different dimensions of the LoF withdifferent dimensions of cross-national distance, ourstudy bridges the literature of the LoF and cross-national distance, which is a key variable determininglocational advantages and disadvantages in host coun-tries relative to the home country. These two conceptsare closely related to each other because large cross-national distance between the host and home countryis usually associated with high LoF.

Similar to the LoF, cross-national distance is alsoa complex concept with a multidimensional nature(Berry, Guillen, and Zhou, 2010; Ghemawat, 2001).Given the close relationship between cross-nationaldistance and the LoF and the fact that both conceptsare aggregated constructs with layers of components,it makes sense to relate these two concepts bycategorizing different types of the LoF according to

different dimensions of cross-national distance.Therefore, in this study, we match different typesof the LoF with different dimensions ofcross-national distance and examine their impact onforeign entry location choice.

Related to the complex nature of the LoF, it is alsoimportant to consider the fact that firms face differenttypes of the LoF under different circumstances. Thereis a lack of consideration of firm heterogeneity whenexamining different types of the LoF. Firms differ inmany aspects, such as their motivations to go abroad.Therefore, it is unrealistic to assume that all firms facethe same type of the LoF regardless of the motivationfor foreign direct investment (FDI). We argue that theimportance of each type of the LoF is not always thesame. We hypothesize that the impact of each typeof the LoF differs depending on whether the FDI ismarket, efficiency, strategic asset, or natural resourceseeking.

To summarize, we answer three research questionsin this study: (1) What are the different types of theLoF?; (2) How do different types of the LoF relate todifferent dimensions of cross-national distance?; and(3) How do they influence the choice of foreign entry?We adopt the OLI paradigm as the theoretical founda-tion of our categorization (Dunning, 1993; Dunningand Lundan, 2008b). Empirically, we find supportfor our hypotheses in the context of Chinese listedfirms investing abroad from 1999 to 2007.

By answering these three questions, this studymakes three contributions. First, it unifies the litera-tures on the LoF and on cross-national distance. Bothare important concepts in international business andmanagement. However, the existing literature doesnot integrate these two streams.We link different typesof the LoF to different dimensions of cross-nationaldistance. Creating such a connection helps us betterunderstand the nature of these two concepts and alsobetter explains the pattern of FDI both theoreticallyand empirically. Besides bridging the literatures onthe LoF and cross-national distance, this study alsocontributes to these two areas separately. It contributesto the literature on the LoF by introducing a new typol-ogy to categorize different types of the LoF andconsidering firm heterogeneity in examining theeffects of the LoF. Adopting the OLI paradigm tocategorize the LoF and measuring the categorizationby cross-national distance enable a systematicapproach to understand the LoF. Classifying theimpact of the LoF by foreign direct investmentmotivation helps us understand the distinctive difficul-ties faced by multinationals when they go abroad.

310 N. Zhou and M. F. Guillen

Copyright © 2016 Strategic Management Society Global Strategy Journal, 6: 309–329 (2016)DOI: 10.1002/gsj.1140

Finally, this study contributes to the literature on cross-national distance by highlighting the contingencynature of the relationship between distance and thelocation choice of FDI. The existing research findingson this relationship are not conclusive. Our studyshows that the relationship is complex and contingenton the types of the LoF and also the FDI motivations.

CATEGORIZING THE LIABILITY OFFOREIGNNESS

The liability of foreignness

The concept of the LoF describes the additional costsmultinationals face relative to host country competi-tors when they operate in foreign countries. Hymer(1960) identified several disadvantages of foreignfirms, such as the lack of information about the hostcountry. Building on Hymer’s insights, Zaheer(1995) developed the concept of the LoF, whichbecomes an important concept in the area of interna-tional business and management.

Since the introduction of the concept, scholars havetried to understand different aspects of the LoF (Edenand Miller, 2001; Hennart, Roehl, and Zeng, 2002;Zaheer and Mosakowski, 1997), including its catego-rization. Scholars have proposed different categoriza-tions of the LoF. In a theoretical piece, Calhoun(2002) identified cultural-driven external and internalsources of the LoF. Eden and Miller (2004) groupedthe LoF into three specific hazards: unfamiliarityhazards, discrimination hazards, and relationalhazards. Qian, Li, and Rugman (2013) distinguishedthe liability of country foreignness and the liabilityof regional foreignness. Similarly, Asmussen andGoerzen (2013) unpacked the LoF into regional,cultural, and institutional dimensions.

Cross-national distance

Existing attempts to categorize the LoF did notsystematically incorporate the concept of cross-national distance—which is also an important conceptin international business—despite the fact that manystudies captured the foreignness of the host countryby calculating the distance with the home country(Berry et al., 2010; Kostova, 1996; Xu and Shenkar,2002; Zhou and Guillén, 2015).

Previous studies on cross-national distance haverecognized that it is a multidimensional concept. Forexample, Xu, Pan, and Beamish (2004) measured

normative and regulative distance. Berry et al.(2010) disaggregated the construct of distance by pro-posing a set of multidimensional measures, includingeconomic, financial, political, administrative, cultural,demographic, knowledge, and global connectednessas well as geographic distance.

There is a large body of literature on the impact ofdifferent distance dimensions on the various globalstrategies, such as entry mode choice (Kogut andSingh, 1988) and subsidiary performance (Barkema,Bell, and Pennings, 1996), among others (for a reviewof the literature, see Werner, 2002). In this study, wefocus on the location choice of FDI, which is one ofthe most important decisions in foreign expansion.Table 1 summarizes the studies in these areas.

The first half of the table includes some of the mostinfluential papers on distance, while the second halffocuses on the studies that investigate the relationshipbetween distance and location choice. The findings ofthe first half of the studies are not consistent. For in-stance, while some found that large cultural distanceencourages wholly owned subsidiaries, other foundthe opposite. These inconsistent findings suggest thatthe relationship between distance and global strategyis complicated, and we need to adopt a contingencyapproach to understand it (Drogendijk and Slangen,2006; Gaur and Lu, 2007).

Most of the recent studies on distance investigatedifferent dimensions of cross-national distance andalso recognize the contingency nature of the complexrelationship. For instance, Slangen and Beugelsdijk(2010) found that institutional hazards are more nega-tively related to vertical foreign activity than tohorizontal foreign activity; and the impact ofgovernance hazards on each type of foreign activityis more negative than the impact of cultural hazardson that type of activity.

Building on these previous studies, we adopt acontingency perspective to understand the relationshipbetween different distance dimensions and foreignlocation choice. Different from other studies, weexplore the contingency by linking different distancedimensions to different types of the LoF, and we alsodistinguish different FDI motivations.

In this study, we adopt Berry et al. (2010)’sdistance dimensions as our pool of selection for differ-ent distance dimensions and pick up the most relevantdistance dimensions for each type of the LoF. Wechoose this distance classification for both theoreticaland empirical reasons. Theoretically, this classifica-tion is comprehensive, and Berry and her colleaguesgrounded their analyses and choice of dimensions on

Categorizing the Liability of Foreignness 311


Table1.

Different

dimensionsof

distance

andforeigndirectinvestment

Authors

Journal

Dependent

variable

Distance

dimensions

Theory

Hom

e/host

country

Findings

Kogut

andSingh

(1988)

Journalo

fInternational

Business

Studies

Jointv

enture

(JV)

vs.greenfield

vs.acquisitio

n

Culturald

istance

The

Uppsalamodel

Worldwide/

UnitedStates

Largercultu

rald

istance

encourages

JVor

greenfield

over

acquisition

Padm

anabhan

andCho

(1996)

Managem

ent

International

Review

Jointv

enture

vs.

wholly

owned

subsidiaries

(WOS)

Culturald

istance

Transactio

ncost

theory

Japan/worldwide

Largercultu

rald

istance

encourages

WOS

Brouthersand

Brouthers

(2001)

Journalo

fInternational

Business

Studies

Jointv

enture

vs.

wholly

owned

subsidiaries

Culturald

istance

Transactio

ncost

theory

Britain,G

ermany,

The

Netherlands,

UnitedStates/CEE

Largercultu

rald

istance

encourages

JV

Barkema,Bell,

andPennings

(1996)

Strategic

Managem

ent

Journal

Longevity

offoreignventure

Culturald

istance

The

Uppsalamodel

The

Netherlands/

worldwide

Largercultu

rald

istance

produces

shorter

longevity

ofJV

sPark

andUngson

(1997)

Academyof

Managem

ent

Journal

Dissolutio

nrate

Culturald

istance

Transactio

ncost

theory

JVswith

U.S.

partners

Largercultu

rald

istance

produces

alower

dissolutionrateof

JVs

Dow

and

Ferencikova

(2010)

International

BusinessReview

Entry

ornot(0/1)

Psychicdistance

The

Uppsalamodel

Various/Slovakia

Psychicdistance

isa

significantly

stronger

predictorof

market

selectionandforeign

directinvestment

performance

than

Kogut

andSingh

index

Nachum,Z

aheer,

andGross

(2008)

Managem

ent

Science

Entry

ornot(0/1)

Knowledge,market,

andresource

proxim

ity

Networktheory,

econom

icgeography

UnitedStates/

worldwide

Proxim

ityto

theworld’s

know

ledgeandmarkets

isastronger

driver

oflocatio

nchoice

than

isproxim

ityto

theworld’s

resources

Slangenand

Beugelsdijk

(2010)

Journalo

fInternational

Foreignaffiliates’

salesto

affiliated

andlocal

Institu

tionalh

azards;

cultu

rald

istance

Institu

tionaltheory,

foreigndirect

investmenttheory

UnitedStates/

worldwide

The

impactof

distance

onforeignsalesdiffers

bythetype

offoreign

Contin

ues



institutional theories of national business, governance,and innovation systems (Henisz and Williamson,1999; La Porta et al., 1998; Nelson and Rosenberg,1993; Whitley, 1992), resulting in nine dimensions,including cultural, economic, demographic, political,administrative, financial, knowledge, globalconnectiveness, and geographic distance.1 Theyapproached cross-national distance from an institu-tional perspective so as to capture the rich diversityof ways in which countries differ, thus followingrecent institutional theorizing in the field of interna-tional business (Jackson and Deeg, 2008; Pajunen,2008). Empirically, in order to overcome the limita-tions of the Euclidean approach, they calculateddyadic distances using the Mahalanobis method,which is scale invariant and takes into considerationthe variance–covariance matrix.

Categorizing different types of the LoF by the OLIparadigm

The OLI paradigm suggests that the decision to investinternationally and the impact of the investment reflectthe joint effects of O-specific, L-specific, andI-specific advantages (Dunning, 1993; Dunning andLundan, 2008b). O-specific advantage arises frommultinationals’ ownership of, or access to, a set ofincome-generating assets, or capabilities to coordinatethese assets, in a way that benefits them relative totheir competitors. L-specific advantages refer to spe-cific resources and market conditions of a host countrythat are potentially available to all firms. I-specificadvantage reflects the greater organizational effi-ciency or superior incentive structure of hierarchies,or the ability to exercise monopoly power over theassets under common governance. The OLI paradigmhas been used widely by scholars to explain why firmsinvest abroad and the patterns of FDI across countries(Hill, Hwang, and Kim, 1990; Li et al., 2013; Singhand Kundu, 2002). The paradigm is designed toexplain the advantages of multinationals and howthese advantages motivate FDI. Few previous studiesadopted this paradigm to examine the disadvantagesthat multinationals face when they invest abroad(Denk, Kaufmann, and Roesch, 2012; Dunning,2000).

1 In our analyses, we excluded financial, knowledge, and globalconnectiveness dimensions because of high correlations withother dimensions.T

able1.

(Contin

ued)

Authors

Journal

Dependent

variable

Distance

dimensions

Theory

Hom

e/host

country

Findings

Business

Studies

unaffiliated

custom

ers

activ

ities

(horizontalo

rvertical)andthetype

ofdistance

(institutional

hazardsandcultu

ral

distance)

Procher(2011)

Annalsof

Regional

Science

Entry

ornot(0/1)

Culturald

istance;

geographic

distance

Agglomeration

theory

France/worldwide

Culturalp

roximity

increase

thelik

elihoodof

aparticular

locatio

nto

bechosen,w

hereas

alarger

distance

betweena

foreignlocatio

nandthe

headquartersdeterforeign

directinvestment

investments



In this study, we adopt the OLI paradigm to catego-rize different types of the LoF for two reasons. First,the OLI paradigm is a comprehensive framework thatsynthesizes many theories to explain the globalizationof multinationals. It is one of the most well-acceptedframeworks in the international business literature(Eden and Dai, 2011; Portugal Ferreira et al., 2011).As we can see from Table 1, many of the studies ondistance and foreign direct investment use transactioncost theory, the Uppsala model, and institutionaltheory. The OLI paradigm is a synthesis of thesedifferent theories. For instance, O-specific andI-specific advantages are related to transaction costtheory because the internalization of O-specific advan-tages reduces transaction costs. L-specific advantageis related to the Uppsala model, because locationalfactors such as cross-national distance determine thepath of globalization. Adopting the OLI paradigmensures that our categorization considers the dominantfactors that influence the internationalization of multi-nationals. Second, the OLI paradigm highlights theadvantages that encourage firms to go abroad. It isexactly the opposite of the LoF, which describes thedisadvantages multinationals face when they goabroad. Such a corresponding relationship betweenthe two concepts enables us to consider the pros andcons of globalization using the same framework.

Ownership-specific LoF refers to the costs thatreduce O-specific advantage. An important part ofO-specific LoF is product adaptation cost. O-specificadvantages are usually exemplified by the superiorityof products in terms of quality, price, technology, orspecial features (Agarwal and Ramaswami, 1992;Anand and Delios, 2002). Differences between hostand home countries in terms of consumer demandmay reduce O-specific advantage because multina-tionals need to pay additional costs—first to identifythe differences and then to adjust their products to suitthe demands of local customers. We label such costsas product adaptation costs. For example, Nokiadeveloped handset models specific to India withfeatures such as a flashlight, a dust cover, and aslip-free grip.

Location-specific LoF refers to the costs thatreduce L-specific advantage. Discrimination cost isan important part of L-specific LoF. Discriminationcosts refer to the costs derived from unfavorable treat-ment by the host country government when multina-tionals try to acquire locational-bound resources inforeign countries. Governments increase discrimina-tion costs by imposing rules and regulations againstforeign firms (Kobrin, 1987; Meschi, 2009),

especially when foreign firms try to acquire local nat-ural resources (e.g., minerals, oil, or gas) or technol-ogy (Garcia-Canal and Guillen, 2008; Henisz,Dorobantu, and Nartey, 2014). Biases held by the hostcountry government against foreign firms reducemultinationals’ legitimacy in the foreign country(Brouthers, O’Donnell, and Hadjimarcou, 2005;Kostova and Zaheer, 1999), resulting in unfavorableconsequences, such as disapprovals of foreign invest-ments (Ambos and Ambos, 2009), and more stringentrequirements on the operation of multinationals(Wang, 2007).

Internalization-specific LoF refers to the costs thatreduce MNEs’ internalization advantage. Governancecosts and appropriation costs are two important partsof I-specific LoF. Governance costs refer to multina-tionals’ additional costs of managing foreign subsidi-aries because of increased communication andcoordination costs (Hennart, 2001). For instance, aforeign manager will spend more time and effort com-municating with his/her local subordinates when theydo not share the same language, thus increasing gover-nance costs, which become part of the I-specific LoF.

Appropriation costs are the costs associated withthe possibility of expropriation of the foreign invest-ment by the government. Foreign direct investmentis more risky than other modes of foreign entry (suchas exports or franchising) because it involves directinvestment in foreign countries (Tallman, 1991).When firms choose to internalize foreign operationsby conducting foreign direct investment, appropria-tion costs become a primary concern. The risk origi-nates from the possibility that once an MNE hassunk the capital necessary to conduct business opera-tions in a country, the government may at some futurepoint face incentives to renegotiate the terms of invest-ment in order to redistribute the MNE’s returns(Henisz and Zelner, 2004). An extreme case wouldbe the seizure of the foreign investment by the govern-ment. The three types of the LoF and the associatedcosts are summarized in Columns 1 and 2 of Table 2.

DIFFERENT TYPES OF THE LIABILITYOF FOREIGNNESS BY TYPE OF FOR-EIGN DIRECT INVESTMENT

The LoF can influence multinationals in differentways (Miller and Parkhe, 2002). In particular, itaffects the process by which multinationals interna-tionalize (Denk et al., 2012): they first enter countriesassociated with lower degrees of the LoF (Johanson



and Wiedersheim-Paul, 1975). Although the threedifferent types of the LoF pose additional costs forall foreign direct investments, their relative impor-tance differs by type.

In his work, Dunning identified four types offoreign direct investment in terms of motivations:market-seeking, efficiency-seeking, strategic asset-seeking, and natural resource-seeking foreign directinvestment (Dunning and Lundan, 2008b). Market-seeking FDI is undertaken to sell products in foreigncountries. The objective of efficiency-seeking FDI isto gain from the common governance of geographi-cally dispersed activities to reduce costs. Strategicasset-seeking FDI is conducted to promote long-termstrategic objectives, such as sustaining or advancingglobal competitiveness through the acquisition ofnew capabilities. Natural resource-seeking FDI occurswhen multinationals invest abroad to acquire particu-lar and specific resources that do not exist in the homecountry or are available at a higher quality and/orlower cost in the host country. Although the threetypes of the LoF are relevant to all four types offoreign direct investment, their relative importancediffers, as shown in Columns 3–6 of Table 2. Next,we will explain the rationales in detail.

Market-seeking foreign direct investment

Because the primary objective of market-seeking for-eign direct investment is to sell products in the hostcountry by exploiting the multinationals’ O-specificadvantages, differences in product demands betweenthe home and host markets become the major concernof multinationals. Such differences jeopardize theO-specific advantages that multinationals enjoy.Therefore, multinationals need to adjust their productsto address the differences. Product adaptation costs areespecially important in market-seeking foreign directinvestment.

We measure product adaptation costs by differentdimensions of cross-national distance. Product adapta-tion costs increase with cross-national distancebecause it reflects differences in consumer demandand, thus, the need to adapt. As mentioned earlier,we apply multiple distance dimensions developed byBerry et al. (2010) to measure product adaptationcosts.

Product adaptation costs are high when consumerdemands in host and home countries are different.The most relevant distance dimensions are thoseclosely related to consumer preference: cultural,T

able2.

Different

componentsof

theliabilityof

foreignnessanddifferenttypes

offoreigndirectinvestment

Types

oftheLoF

Corresponding

cost

Market-seeking

foreigndirect

investment

Efficiency-seeking

foreigndirect

investment

Strategicasset-

seekingforeign

directinvestment

Naturalresource-

seekingforeign

directinvestment

Corresponding

distance

Ownership-specific

LoF

Productadaptation

cost

Moreim

portant

Lessim

portant

Lessim

portant

Lessim

portant

Culturald

istance,

econom

icdistance,

demographicdistance

Location-specific

LoF

Discrim

inationcost

Lessim

portant

Moreim

portant

Moreim

portant

Moreim

portant

Politicaldistance

Internalization-

specificLoF

Governancecost,

appropriationcost

Lessim

portant

Moreim

portant

Moreim

portant

Lessim

portant

Adm

inistrativedistance,

politicalhazard

LoF

,liabilityof

foreignness



economic, and demographic distance. Culturaldistance refers to the difference in cultural valuesand norms, and it creates product adaptation costs byinfluencing the choices that consumers make becauseof their preference (Ghemawat, 2001). For example,consumer durable industries are particularly sensitiveto differences in consumer tastes and preferences.The Japanese prefer small automobiles and householdappliances because of the social norm common incountries where space is highly valued.

Product adaptation costs are also associated witheconomic distance. Economic distance reflects thedifference in consumer purchasing power betweentwo countries. Purchasing power is the ability ofconsumers to buy. In general, the more economicallydeveloped a country, the more purchasing powerconsumers in that country have. Because beforeexpanding abroad, the firm typically sells goods andservices tailored to the characteristics of the homecountry (Vernon, 1979), differences in purchasingpower and other economic characteristics have animpact on demand conditions and, therefore, the needto adapt.

Product adaptation costs are also affected by demo-graphic distance, which captures national difference interms of the size, growth, and age structure of popula-tion (Berry et al., 2010). These differences have directimplications for market attractiveness and growthpotential. Differences in factors such as life expec-tancy, birth rates, and the resulting age structure ofthe population, among others, attest to fundamentalcharacteristics of the population of countries andmay affect consumer behavior and, thus, the need toadapt.

Summarizing these arguments on product adapta-tion costs, we predict that:

Hypothesis 1: Cultural distance, economic dis-tance, and demographic distance reducemarket-seeking foreign direct investment.

Efficiency-seeking foreign direct investment

Efficiency-seeking foreign direct investment criticallydepends on the balance between the advantagesgained from spreading value-added activities acrossvarious locations and the costs of communicationand coordination over distance (Aarland et al.,2007). Because of the relatively low internationalmobility of labor, wage differentials between home

and host countries can become a major determinantof efficiency-seeking foreign direct investment(Kimino, Saal, and Driffield, 2007). Various empiricalstudies have supported the prediction that lower wagesattract foreign direct investment (Taylor, 2000).Therefore, efficiency-seeking foreign direct invest-ment is conducted to acquire L-specific advantage(low labor cost) in foreign markets. Accordingly, theL-specific LoF is important. Discrimination cost fromthe actions by the foreign government becomes amajor concern for efficiency-seeking foreign directinvestment. For example, the government couldimpose more stringent regulations on the operationof multinationals (Wang, 2007).

Discrimination cost is closely related to politicaldistance, which captures the difference in institutionalchecks and balances (Demirbag, Glaister, andTatoglu, 2007; Dow and Karunaratna, 2006; Henisz,2000), democratic character, the size of the state rela-tive to the economy, and external trade associations(Brewer, 2007; Hirschberg, Sheldon, and Dayton,1994). For efficiency-seeking foreign direct invest-ments, the primary source of discrimination by hostcountry government is political distance, becausediscrimination treatment is more likely to happen tomultinationals from countries with large politicaldistance because of the host country government’sunfamiliarity with the home country’s political system(Brandt and Li, 2003).

Host country governments can increase discrimina-tion costs by imposing rules and regulations againstforeign firms (Kobrin, 1987; Meschi, 2009). Hostcountry governments tend to associate foreign firmswith negative images because of unfamiliarity(Ewing, Windisch, and Newton, 2010). Such imagesreduce foreign multinationals’ legitimacy in foreigncountries (Brouthers et al., 2005; Kostova and Zaheer,1999), resulting in unfavorable consequences such asthe reluctance of the host country government to allowforeign investment from emerging markets (Ambosand Ambos, 2009) and more stringent requirementson the operation of foreign multinationals (Wang,2007). Therefore, discrimination costs faced byefficiency-seeking FDI increase with the politicaldistance between host and home countries.

Efficiency-seeking foreign direct investment is alsoinfluenced by I-specific LoF. Because hiring foreignemployees with lower wages is an important driverfor efficiency-seeking FDI, governance costs associ-ated with managing foreign employees become theprimary concern for managers. Governance costsgrow with administrative distance, which refers to



differences in bureaucratic patterns owing to colonialties, language, religion, and the legal system(Ghemawat, 2001; La Porta et al., 1998; Whitley,1992). While one could argue that administrativedistance is related to cultural distance, we believe itis distinct because it goes beyond it to include bothformal and informal institutional arrangements in asociety. Administrative distance increases governancecosts by raising communication, coordination, andadministrative costs. For example, a Chinese managermay find it difficult to communicate with its Frenchsubordinates because of language barriers.

Appropriation costs are also important inefficiency-seeking FDI because establishingmanufacturing facilities in foreign countries usuallyinvolves large investments in plants, equipments,raw materials, and other necessary inputs (Heniszand Delios, 2001). Therefore, the concern aboutpotential appropriation is also a primary considerationin efficiency-seeking foreign direct investment.

Different from other costs that are increased bycross-national distance, appropriation costs areenhanced by political hazards (Delios and Henisz,2000). Political hazards in a country are high whenpolicy makers can act unilaterally or have highcertainty that a subservient or allied legislature andjudicial branch will support their actions, becausefuture policies are likely to be particularly volatile inresponse to exogenous shocks, to changes in the iden-tity of policy makers, or to changes in the preferencesof existing policy makers (Delios and Henisz, 2003).When political hazards are high, multinationals facehigh appropriation costs because the government caneasily change policy or, at the extreme, seize the firm’sforeign assets.

Summarizing these arguments on discriminationcosts, governance costs, and appropriation costs, wepredict:

Hypothesis 2: Political distance, administrativedistance, and political hazards reduceefficiency-seeking foreign direct investment.

Strategic asset-seeking foreign direct investment

Strategic asset-seeking FDI is driven by the multina-tionals’ need to access new resources and capabilitiesin foreign countries. In strategic asset-seeking FDI,there are two primary objectives: to acquire foreignstrategic assets such as knowledge and to transfer it

within the multinational (Li et al., 2013; Lu, Liu,and Wang, 2011). While the former is related to dis-crimination cost, the latter is related to governancecost and appropriation cost.

Discrimination costs are important for strategicasset-seeking FDI because multinationals need toobtain approval from the host country governmentto acquire strategic assets. Strategic assets such astechnology and brand are best acquired through ac-quisition (Elango and Pattnaik, 2011; Luo et al.,2011). Discrimination costs are high when acquisi-tions by foreign firms are conceived by local stake-holders as sensitive and undesirable (Graham andMarchick, 2006). The host country governmentmay create barriers by banning acquisitions, becauseof the fear that national security would be threat-ened. For example, Huawei, a Chinese telecommu-nication device provider, failed in its attempt toacquire 3leaf, an American company, because theU.S. government did not approve the acquisitionout of national security concerns. Therefore,discrimination costs are important in strategicasset-seeking foreign direct investment. We haveargued earlier that discrimination costs are increasedby political distance.

In strategic asset-seeking foreign direct investment,another important consideration is how to transfer theacquired assets or knowledge within the firm effec-tively (Makino and Delios, 1996; Steensma et al.,2000). Therefore, I-specific costs are important. Manyscholars have argued that some knowledge is partiallytacit and transfer requires frequent interactions withinthe organization (Almeida, 1996; Kogut and Zander,1992; Nelson and Winter, 1982). The efficiency ofsuch transfer is closely related to governance costsbecause external or ‘imported’ knowledge needs tobe adapted to different conditions and shared withinthe MNE (Teece, 1986), and such adaptation andtransfer increases the costs of monitoring and manag-ing. We have argued earlier that governance costs areenhanced by administrative distance.

Another component of I-specific LoF is appropria-tion costs. Strategic asset-seeking FDI usually takesthe form of mergers and acquisitions (Batson, 2007),and conducting mergers and acquisitions in foreigncountries requires a large amount of investment.Therefore, concern of appropriation is also a primaryconsideration in strategic asset-seeking foreign directinvestment.

Summarizing these arguments on discriminationcosts, governance costs, and appropriation costs, wepredict:



Hypothesis 3: Political distance, administrativedistance, and political hazards reduce asset-seeking foreign direct investment.

Natural resource-seeking foreign direct investment

In natural resource-seeking FDI, the primary goal is toacquire foreign natural resources. Accordingly, dis-crimination costs are crucial. Similar to strategicasset-seeking FDI, it is critical for natural resource-seeking FDI to obtain approval from the host countrygovernment. The difficulties frequently faced bymultinationals when it comes to obtaining licensesand approvals from the host country government innatural resource industries have been documentedextensively in the literature (Globerman and Shapiro,2009; Henisz et al., 2014). We have argued earlier thatdiscrimination costs increase with political distance.Thus, we expect that:

Hypothesis 4: Political distance reduces naturalresource-seeking foreign direct investment.

METHODOLOGY

Empirical setting

To test the hypotheses, we used data on the foreign en-tries made by Chinese listed firms from 1999 to 2007.The data begin in the year of 1999 because the ac-counting rules in China required listed firms to reportforeign subsidiaries systematically only after 1999.The data on foreign subsidiaries before 1999 is likelyto be unreliable or uncomprehensive. Including dataprior to 1999, however, does not change the resultswe report. This Chinese sample provides an excellentresearch setting for three reasons. First, China hasbecome one of the largest outward FDI investors inthe world. Chinese outward FDI is increasing rapidly,from $0.9 billion in 1991 to $101 billion in 2013(UNCTAD, 1996, 2014). Second, Chinese firms haveinvested in a wide array of countries, including 23developed countries and 52 developing countries inour sample. Third, Chinese firms started to globalizeduring the 1990s. Examining Chinese firms from theearly stage of globalization avoids left-censoringproblems.

For each Chinese firm in the sample, we reviewedtheir annual reports to identify any foreign subsidiary.

In cases in which not enough information was pro-vided, we searched other internet sources or contactedthe company. Chinese listed firms are required to dis-close information on major subsidiaries (e.g., countryof entry, amount of capital invested, percentage ofownership, and primary business activities or majorproducts) in the appendix of their annual reports.And then, we determined the establishment year ofeach subsidiary as given in the annual report, on thecompany’s website, or through further internetsearches. We obtained ownership information fromGuo Tai An (GTA), a research service center. Finally,we collected financial data from China Stock MarketAccounting Research (CSMAR), a database that ispart of Wharton Research Data Services. Overall,there are 256 Chinese firms owning 649 foreignsubsidiaries in 38 countries from 1999 to 2007.

Variable definitions

Dependent variables

We identified different types of foreign direct invest-ment through the description of subsidiaries’ businessactivities in the company’s annual report. Market-seeking foreign direct investment includes foreignsubsidiaries that sell products in a given host country.Efficiency-seeking foreign direct investment includesforeign subsidiaries that produce but do not sell prod-ucts in a host country. Strategic asset-seeking foreigndirect investment includes foreign subsidiaries thatconduct research and development activities in a hostcountry. Natural resource-seeking foreign direct in-vestment includes foreign subsidiaries that exploitnatural resources, such as mining or fishing, in a hostcountry. It is possible that an MNE enters a foreigncountry in a year for multiple reasons. In this case, thisparticular firm-country-year appears in our datasetmultiple times, but each type of foreign directinvestment appears only once. When the informationon foreign subsidiaries is not comprehensive enoughto determine the purpose of a foreign subsidiary, wesearched the company’s website or did a web searchto find out detailed information on the nature of theforeign subsidiary. In some cases, we contacted thecompany to ask about the particular foreign subsidiary.We hired two research assistants to code the differenttypes of FDI, and the inter-rater reliability correlationcoefficient was +0.92. When there was a disagree-ment, one of the authors decided which coding toadopt based on further research on the subsidiary.



For each Chinese firm, the foreign direct invest-ment dependent variables are dummies that equalto ‘1’ if a firm makes a certain type (market seek-ing, efficiency seeking, strategic asset seeking, ornatural resource seeking) of FDI in a certain hostcountry in a certain year, and it is ‘0’ otherwise.The level of analysis is firm-country-year. Becausenot all parent firms are listed from the beginningof our study period, we have an unbalanced paneldataset.

Independent variables

Cross-national distance measures come from the data-base developed by Berry et al. (2010). Details about

cultural, economic, demographic, political, andadministrative distance are summarized in Table 3.

Political hazard is measured by the POLCONVindex. This annual time-varying measure obtainedfrom Henisz (2000) quantifies the extent to whichany one institutional actor—e.g., the executive or alegislative chamber—in a given country is uncon-strained in its choice of policies. It has been usedwidely by scholars to measure political hazard(Delios and Henisz, 2003; Holburn and Zelner,2010; Slangen, 2013).

Control variables

We used the percentage held by the top 10 share-holders to measure ownership concentration.Managers may choose to invest abroad to pursuetheir own interests rather than shareholders’. Highownership concentration suggests large owners

2 The information in this table is from Berry et al. (2010).

Table 3. Details about distance dimensions2

Dimension Component variables Source

Cultural distancePower distance WVS questions on obedience and respect for authority World Value SurveyUncertainty avoidance WVS questions on trusting people and job security World Value SurveyIndividualism WVS questions on independence and the

role of government in providing for its citizensWorld Value Survey

Masculinity WVS questions on the importance of family and work World Value SurveyEconomic distanceIncome GDP per capita (in 2000 U.S. dollars) WDIInflation GDP deflator (% GDP) WDIExports Exports of goods and services (% GDP) WDIImports Imports of goods and services (% GDP) WDIDemographic distanceLife expectancy Life expectancy at birth, total (years) WDIBirth rate Birth rate, crude (per 1,000 people) WDIPopulation under 14 Population ages 0–14 (% of total) WDIPopulation under 65 Population ages 65 and above (% of total) WDIPolitical distancePolicy-making uncertainty Political stability measured by considering independent

institutional actors with veto powerPOLCON

Democratic character Democracy score Freedom HouseSize of the state Government consumption (% GDP) WDIWTO member Membership in WTO (GATT before 1993) WTORegional trade agreement Dyadic membership in the same trade bloc WTOAdministrative distanceColonizer-colonized link Whether dyad shares a colonial tie CIA FactbookCommon language % population that speaks the same language in the dyad CIA FactbookCommon religion % population that shares the same religion in the dyad CIA FactbookLegal system Whether dyad shares the same legal system La Porta et al., 1998

CIA, Central Intelligence Agency; GATT, General Agreement on Tariffs and Trade; GDP, gross domestic product; POLCON, politicalconstraint; WDI, World Development Index; WTO, World Trade Organization; WVS, World Value Survey.



have more incentive and power to monitormanagers’ behavior, thus reducing agencyproblems (Fama and Jensen, 1983; Jensen andMeckling, 1976).

Product diversification also influences the proba-bility of foreign direct investment in that it isresource consuming and, thus, competes withglobalization for scarce resources such as capital.However, the experience generated from productdiversification also might be applied to globaliza-tion. Research on the relationship between productdiversification and globalization remains inconclu-sive (Doukas and Lang, 2003; Tallman and Li,1996). We controlled for product diversificationusing the concentric index developed by Montgom-ery and Wernerfelt (1988).

A firm’s size may influence the pattern of itsmultinational activity (Swaminathan and Delacroix,1991) and provide an indicator of its ability tosurvive in foreign markets. Larger firms should bemore likely to go abroad, whereas smaller firmslikely lack the knowledge and experience to expandoverseas. We measured firm size by the logarithm oftotal sales.

A firm’s agemay also influence the probability of itinvesting abroad because older firms may be subjectto inertia, which will prevent strategic changes(Hannan and Freeman, 1977). We measured age bythe number of years since founding.

Better-performing firms are more likely to investabroad because they have the necessary resourcesand capabilities. Therefore, we controlled for firm per-formance by return on assets, or ratio of net profit tototal assets.

Because firms can learn from their internationalexperiences (Barkema and Vermeulen, 1998; Deliosand Henisz, 2003; Hitt, Li, and Worthington, 2005),we also controlled for a firm’s prior internationalexperience, which is measured by the total numberof foreign investments a firm made before the yearof a potential entry.

We controlled for geographic distance because itinfluences transportation costs. It is calculated byusing the great circle method. For each type offoreign direct investment, we also control for hostcountry locational advantage. For market-seekingforeign direct investment, we include the size ofthe population, which is measured by a country’spopulation. For efficiency-seeking foreign directinvestment, we include gross domestic product percapita as a control for the level of labor cost. Forstrategic asset-seeking foreign direct investment,

we include the number of patents per million peopleas a control for the stock of knowledge and technol-ogy in a country. For natural resource-seekingforeign direct investment, we include total naturalresource rents as a percentage of gross domesticproduct to control for the stock of natural resourcein a country. Total natural resource rents are thesum of oil rents, natural gas rents, coal rents (hardand soft), mineral rents, and forest rents. The datasource for these locational factors is the WorldDevelopment Index.

Finally, to control for possible industry and timeeffects, we included industry and year dummies inall regressions.

Estimation method

Because the dependent variables are dummy vari-ables, we used logit regression. We constructed thedataset by including all of the possible combinationsof firm-country-year observations. Given that only asmall proportion of the dependent variable has thevalue of one, we estimated rare-event logistic regres-sion models. The relogit command in STATA generatesapproximately unbiased and lower-variance estimatesof logit coefficients and their variance–covariancematrix by correcting for small samples and rare events(King and Zeng, 2001). We used the cluster option inSTATA to account for intragroup variance. We did notuse the conditional logit model because it has severallimitations, such as the inability to accommodate orcontrol for the effects of firm-level heterogeneity(Martin, Swaminathan, and Tihanyi, 2007). We alsoprovide additional results using seemingly unrelatedregression.

RESULTS

Table 4 provides the means, standard deviations, andcorrelations for the sample. The mean values of thedependent variables are close to zero, which confirmthat our choice of rare-event logit regression is appro-priate. In order to reduce the correlations amongdifferent distance dimensions and also to comparecoefficients, we normalized them. The correlationsof most normalized distance dimensions are not high.We also checked the variance inflation factor scores,and they are lower than 10. Therefore, there is noserious multicollinearity.



Table4.

Means,standarddeviations,and

correlations

#Variables

Mean

Standard

deviation

12

34

56

78

910

1112

1314

1516

1Market-seekingforeigndirectinvestment

0.0004

0.02

1.00

2Efficiency-seekingforeigndirectinvestment

0.0000

0.00

0.00

1.00

3Strategicasset-seekingforeigndirectinvestment

0.0001

0.01

0.00

0.00

1.00

4Naturalresource-seeking

foreigndirectinvestment

0.0000

0.00

0.00

0.00

0.00

1.00

5Culturald

istance

0.00

1.00

0.00

0.00

0.00

0.00

1.00

6Economicdistance

0.00

1.00

0.01

0.00

0.02

0.01

�0.02

1.00

7Dem

ographicdistance

0.00

1.00

0.00

0.00

0.01

0.00

�0.28

0.40

1.00

8Po

liticaldistance

0.00

1.00

0.01

0.00

0.01

0.00

�0.36

0.24

0.03

1.00

9Po

liticalhazard

0.00

1.00

0.01

0.00

0.01

0.00

�0.01

0.41

0.34

0.02

1.00

10Adm

inistrativedistance

0.00

1.00

0.00

0.00

0.00

0.00

�0.04

0.39

0.39

0.11

0.22

1.00

11Ownershipconcentration

61.23

13.06

0.00

0.00

0.00

0.00

0.05

�0.01

0.00

�0.10

0.01

0.00

1.00

12Productd

iversificatio

n0.17

0.35

0.00

0.00

�0.01

0.00

�0.01

0.00

0.00

0.03

0.00

0.00

�0.07

1.00

13Log

ofsales

20.46

1.31

0.04

0.00

0.01

0.00

�0.15

0.03

0.02

0.21

�0.02

0.01

0.12

�0.10

1.00

14Age

7.92

3.79

�0.01

0.00

�0.01

0.00

�0.24

0.05

0.04

0.35

�0.02

0.02

�0.41

0.02

0.08

1.00

15Returnon

assets

0.03

0.09

0.00

0.00

0.00

0.00

0.03

�0.01

�0.01

�0.04

�0.01

0.00

0.11

�0.01

0.12

�0.14

1.00

16Priorinternationalexperience

0.18

2.00

0.01

0.00

0.00

0.04

�0.01

0.00

0.00

0.02

0.00

0.00

0.04

0.03

0.15

�0.03

0.01

1.00

17Geographicdistance

0.00

1.00

0.00

0.00

0.00

0.00

�0.08

�0.10

�0.12

�0.03

0.13

�0.10

0.00

0.00

0.00

0.00

0.00

0.00

n=136,513.Correlatio

nsgreaterthan

0.009aresignificantatthe

0.01

level(two-tailedtest).



Table 5 shows the results of the rare-event logit re-gressions for Chinese firms to test Hypotheses 1–4.There are eight model specifications in the table. Thedependent variable of Models 1 and 2 in Table 6 ismarket-seeking FDI; of Models 3 and 4 is efficiency-seeking FDI; of Models 5 and 6 is strategic asset-seeking FDI; and of Models 7 and 8 is naturalresource-seeking FDI. Models 1, 3, 5, and 7 are thebaseline models that include only control variables.In Models 2, 4, 6, and 8, we entered the distance di-mensions that measure the three different types ofthe LoF.

First, we focused onmarket-seeking FDI. InModel2, all three measures of O-specific LoF (product adap-tation costs)—cultural, economic, and demographicdistance—are negative and significant, while neitherL-specific LoF (discrimination costs: political dis-tance) nor I-specific LoF (appropriation costs: politi-cal hazard; and governance costs: administrativedistance) are significant. The results suggest that thethree measures of product adaptation costs havegreater impact on the location choice of market-seeking foreign direct investment. Therefore,Hypothesis 1 is supported.

In addition to being statistically significant, theseeffects are also large in magnitude. In Model 2, hold-ing other variables at their mean values, when culturaldistance increases from the 25th percentile to the 75thpercentile, the probability of investing decreases by97.7 percent. The same number is 60.3 percent for de-mographic distance and 23.5 for economic distance.The results show that although absolute risk is low,product adaptation costs do have a large impact onthe relative risk of foreign direct investment.

In Model 4, political distance, political hazard, andadministrative distance are negative and significant,while cultural, economic, and demographic distanceare not significant. The results suggest that L-specificLoF and I-specific LoF, not O-specific LoF, are themajor obstacles of efficiency-seeking foreign directinvestment. Therefore, Hypothesis 2 is supported. InModel 4, holding other variables at their mean values,when political distance, political hazard, and adminis-trative distance increase from the 25th percentile to the75th percentile, the probability of investing decreasesby 11.9 percent, 15.5 percent, and 47.1 percent,respectively.

The results inModel 6 are similar to those inModel4. Political distance, political hazard, and administra-tive distance are all negative and significant. Therefore,Hypothesis 3 is also supported. In Model 6, holdingother variables at their mean values, when political

distance, political hazard, and administrative distanceincrease from the 25th percentile to the 75th percentile,the probability of investing decreases by 20.3 percent,8.9 percent, and 37.8 percent, respectively.

In Model 8, only political distance is negative andsignificant. The results show that discrimination costsare the primary concern for natural resource-seekingforeign direct investment. So, Hypothesis 4 is alsosupported. In Model 8, holding other variables at theirmean values, when political distance increases fromthe 25th percentile to the 75th percentile, the probabil-ity of investing decreases by 23.6 percent.

Among the control variables, product diversifica-tion is negative and significant in most models. Firmsthat engage in greater product diversification are lesslikely to invest abroad, because of resource con-straints. Ownership concentration is negative in somemodels, suggesting that firms with concentratedowners are less prone to encounter agency problemand, thus, are less likely to invest abroad. Log of salesis positive and significant: larger firms are more likelyto invest abroad because they are more likely to pos-sess the resources to support foreign expansion. Firmage is negative and significant, indicating that olderfirms are subject to greater inertia and less likely to un-dertake strategic changes (Freeman, Carroll, andHannan, 1983). Return on assets is positive and signif-icant in some models: better-performing firms aremore likely to invest abroad. Prior international expe-rience is positive and significant in efficiency-seekingand natural resource-seeking foreign direct invest-ment, showing that prior international experience en-courages firms to conduct these two types of foreigndirect investment. Host country locational advantageis positive and significant in most models, indicatingthat host country locational advantage is indeed an im-portant driver for different types of foreign direct in-vestment. Finally, geographic distance is negativeand significant only in efficiency-seeking foreign di-rect investment. In other types of foreign direct invest-ment, geographic distance is insignificant or evenpositive and significant, meaning that multinationalsdo overcome physical distance and its associated costssuch as transportation costs in market-seeking, strate-gic asset-seeking, and natural resource-seekingforeign direct investment.

Robustness checks

Because it is possible that a multinational enters a for-eign country in a year for multiple motivations, we



Table5.

Rare-eventlogitregressionsof

distance

dimensionson

differenttypes

offoreigndirectinvestment

Variables

Model1

Model2

Model3

Model4

Model5

Model6

Model7

Model8

Dependent

variable

Market-seekingFD

I(H

1)Efficiency-seekingFD

I(H2)

Strategicasset-seekingFD

I(H

3)Naturalresource-seeking

FDI(H

4)Independentv

ariables

Culturald

istance

�0.65***

�1.23

�0.12

�0.04

(0.19)

(3.40)

(0.39)

(0.18)

Economicdistance

�0.26+

�0.12

�0.08

�0.05

(0.15)

(0.19)

(1.29)

(0.33)

Dem

ographicdistance

�0.41**

�0.09

�0.07

�0.04

(0.15)

(8.13)

(0.08)

(0.03)

Politicaldistance

�0.07

�0.05***

�0.09*

�0.18***

(0.27)

(0.00)

(0.04)

(0.03)

Politicalhazard

0.22

�0.07***

�0.04**

�0.04

(0.18)

(0.00)

(0.01)

(0.03)

Adm

inistrativedistance

�0.23

�0.44***

�0.26*

�0.74

(0.16)

(0.08)

(0.12)

(0.92)

Productd

iversificatio

n�0

.30+

�0.29+

�1.22

�1.19

�0.56*

�0.49+

�1.35

�0.82***

(0.17)

(0.17)

(2.58)

(2.59)

(0.26)

(0.26)

(0.14)

(0.18)

Ownershipconcentration

�0.02

�0.02

�0.03

�0.03

�0.01

�0.01

�0.11***

�0.08***

(0.01)

(0.01)

(0.03)

(0.03)

(0.03)

(0.02)

(0.01)

(0.02)

Log

ofsales

1.30***

1.33***

0.64

0.63

0.78***

0.80***

0.67***

0.63***

(0.24)

(0.24)

(0.59)

(0.60)

(0.21)

(0.22)

(0.13)

(0.15)

Age

�0.28***

�0.28***

�0.06

�0.06

�0.29*

�0.39**

�0.04

�0.07

(0.07)

(0.08)

(0.12)

(0.12)

(0.13)

(0.13)

(0.08)

(0.10)

Returnon

assets

4.04

3.67

4.02

5.60

+7.48***

6.39***

29.53

43.93***

(3.37)

(3.64)

(3.30)

(3.33)

(0.99)

(1.15)

(59.28)

(0.57)

Geographicdistance

0.21*

0.13

�0.60

�0.82***

�0.04

�0.63

�0.48

�0.31

(0.10)

(0.11)

(0.81)

(0.09)

(0.25)

(0.75)

(1.44)

(0.39)

Priorinternationalexperience

�0.08

�0.09

0.18***

0.15***

0.28

0.05

0.20***

0.14*

(0.07)

(0.07)

(0.03)

(0.02)

(0.25)

(0.30)

(0.04)

(0.06)

Locationaladvantage

0.10**

0.11

0.07*

0.11***

0.01***

0.01

0.17

0.28***

(0.03)

(0.07)

(0.03

(0.01)

(0.00)

(0.01)

(0.21)

(0.01)

Constant

�34.87***

�34.43***

5.15

8.59***

8.89***

6.51***

17.31***

9.18***

(7.86)

(8.15)

(5.07)

(0.12)

(0.46)

(0.39)

(0.36)

(0.36)

Industry

andyear

dummies

Included

Included

Included

Included

Included

Included

Included

Included

Num

berof

observations

136,513

136,513

136,513

136,513

136,513

136,513

136,513

136,513

Log

pseudo-likelihood

�325.89

�305.09

�27.99

�20.67

�120.81

�112.47

�14.22

�11.36

Pseudo

r-square

0.31

0.35

0.20

0.41

0.25

0.30

0.41

0.53

Robuststandarderrorsarein

parentheses.FD

I,foreigndirectinvestment.

+p<

0.1*

p<

0.05

**p<

0.01

***

p<

0.001(two-tailedtests).



checked for the robustness of our results by simulta-neously estimating the four models of market-seeking,efficiency-seeking, strategic asset-seeking and naturalresource-seeking FDI using seemingly unrelated re-gression model. STATA does not support seemingly un-related regression for rare-event logit models.Therefore, we used a logit model instead. The resultsare summarized in Table 6. The results are largelyconsistent with those in Table 5. Every distance di-mension that was significant in Table 6 remained neg-ative and significant in Table 6.

DISCUSSION AND CONCLUSION

In this study, we adopted the OLI paradigm and cate-gorized the LoF into three different types: O-specific,L-specific, and I-specific LoF. We linked the threedifferent types of LoF with different dimensions ofcross-national distance, and we then argued that mul-tinationals face different types of the LoF (as mea-sured by different dimensions of cross-nationaldistance) when they operate abroad for different rea-sons. The empirical test on a sample of Chinese firms

Table 6. Robustness checks

Variables Model 1 Model 2 Model 3 Model 4

Dependent variables Market- seekingFDI

Efficiency- seekingFDI

Strategic asset- seekingFDI

Natural resource-seeking FDI

Independent variablesCultural distance �0.68*** �0.50 �0.59 �0.19

(0.19) (0.87) (0.40) (0.36)Economic distance �0.27+ �0.10 �0.02 �0.09

(0.15) (0.18) (1.28) (1.41)Demographic distance �0.43** �0.54 �0.86 0.09

(0.15) (2.52) (0.83) (0.21)Political distance �0.10 �0.15*** �0.07* �0.12***

(0.27) (0.02) (0.03) (0.03)Political hazard �0.26 �0.05*** �0.02+ �0.11

(0.18) (0.00) (0.01) (0.19)Administrative distance �0.24 �0.43*** �0.48** �0.56

(0.16) (0.09) (0.20) (0.78)Product diversification �0.22 �0.19 �0.39 �1.67***

(0.17) (0.26) (0.28) (0.18)Ownership concentration �0.02 �0.04 �0.01 �0.11***

(0.02) (0.03) (0.03) (0.02)Log of sales 1.37*** 0.68 0.95*** 0.75***

(0.24) (0.60) (0.22) (0.15)Age �0.28*** �0.10 �0.41*** 0.02

(0.08) (0.12) (0.12) (0.10)Return on assets �1.02 5.60+ 4.10*** 5.67***

(1.31) (3.33) (1.26) (0.57)Locational advantage 0.10 0.11*** 0.01 0.03

(0.07) (0.01) (0.01) (0.04)Geographic distance 0.14 �0.08*** �0.09 �1.56***

(0.11) (0.00) (0.08) (0.39)Prior international experience �0.10 �0.15 �0.32 0.30***

(0.08) (0.17) (0.32) (0.06)Industry and year dummies Included Included Included IncludedConstant �35.52*** �264.36*** �45.17*** �11.43***

(8.15) (20.45) (7.74) (3.15)Number of observations 136,513 136,513 136,513 136,513

Robust standard errors are in parentheses.+ p < 0.1 * p < 0.05 ** p < 0.01 *** p < 0.001 (two-tailed tests).



supported our hypotheses. Our analyses have boththeoretical and practical implications.

Theoretically, we contribute to the literature on theLoF by identifying and measuring different types ofthe LoF. Although scholars have tried to categorizedifferent types of the LoF, we advance our under-standing of the LoF in three ways. First, we adoptedthe OLI paradigm as the theoretical foundation ofour categorization. Unlike other studies that lack acomprehensive framework to support their categoriza-tion, we adopted the OLI paradigm—one of the mostcomprehensive and well-accepted theories to explainFDI—as the theoretical foundation for our categoriza-tion. By adopting the OLI paradigm, we also catego-rize the LoF by its impact and not by its source, asmost other studies did. Second, we highlighted therole of firm heterogeneity in assessing the impact ofthe LoF. We showed that firms face different typesof the LoF according to different motivations forengaging in foreign direct investment. Our study isone of the first to examine how the various types ofthe LoF influence foreign entry decisions when multi-nationals invest abroad for different motivations.Third, we also empirically measure different types ofthe LoF by different dimensions of cross-nationaldistance. Previous attempts to categorize the LoF arelargely theoretical. Our study advances our under-standing of the LoF by empiricallymeasuring and test-ing different types of the LoF.

Our study also contributes to the literature on dis-tance by linking distance to the LoF. Although priorstudies have recognized the multidimensional natureof cross-national distance, they have not linked it withthe concept of the LoF. By creating such a linkage, weadvance our understanding of the impact of distanceon foreign entry decision. Moreover, the context ofthis study is firms from emerging markets, while themajority of existing studies on distance focus on firmsfrom developed countries. Thus, this study adds newinsights to distance research by exploring newcontexts.

The findings of this article provide insights into therecent debate on whether the world is flat (Friedman,2005), spiky (Florida, 2005), or semi-globalized(Ghemawat, 2003). While proponents of the view thatthe world is flat believed that the advance in technol-ogy has made the world borderless, believers in theview that the world is spiky or semiglobalized observethat there are still substantial globalization barriers.Our findings support the latter view by showing thatthe world is still spiky and that different dimensionsof cross-national distance still play significant roles

in deterring foreign entries, although the impact ofgeographic distance itself becomes negligible.

Finally, the findings regarding the role of priorforeign experience on foreign direct investment speakto the validity of staged theories of internationalization(Johanson and Vahlne, 1977, 1990). Although priorexperience helps overcome the LoF in efficiency-seeking and natural resource-seeking foreign directinvestment, its impact is limited in the other types offoreign direct investment.

Our findings also have practical implications.MNE managers need to carefully assess the extent towhich their own firms are affected by each type ofthe LoF. It is imperative not to simply emulate otherfirms going abroad. By the same token, policy makersfocused on attracting foreign direct investment to thehost country would need to carefully analyze what toemphasize as a magnet for foreign multinationals.Our analysis suggests that they need to adopt policiesthat reduce different types of the LoF for each type offoreign direct investment.

This research suffers from several limitations. First,the sample is from only one home country. Therefore,one should be cautious in generalizing the implica-tions of our findings to firms from other countrieswithout examining the peculiar characteristics ofChina. Future research could analyze multiple homecountries to test the external validity of the resultsreported in this article. A related point is that, whenexamining multiple host and home countries, scholarsneed to recognize the asymmetric nature of cross-national distance, which we did not address in thisarticle because we have only one home country. Third,we focused on only three types of the LoF. It is possi-ble that there are other types of the LoF that hinder for-eign investments. Finally, we did not consider theeffects of prior international experience based on priorinvestments within specific types, which would enabletesting if prior experience in efficiency-seekingforeign direct investment, for instance, is relevant formarket-seeking or strategic asset-seeking FDI. Theselimitations offer additional avenues for future researchwithin the systematic framework of analysis proposedin this article.

ACKNOWLEDGEMENTS

The authors are grateful to GSJ Editor Jay Anand andtwo anonymous reviewers for their invaluableguidance and suggestions during the review process.



REFERENCES

Aarland K, Davis JC, Henderson JV, Ono Y. 2007. Spatialorganization of firms: the decision to split production andadministration. RAND Journal of Economics 38(2):480–494.

Agarwal S, Ramaswami SN. 1992. Choice of foreign marketentry mode: impact of ownership, location, and internaliza-tion factors. Journal of International Business Studies23(1): 1–27.

Almeida P. 1996. Knowledge sourcing by foreign multina-tionals: patent citation analysis in the U.S. semiconductorindustry. Strategic Management Journal, Winter SpecialIssue 17: 155–165.

Ambos TC, Ambos B. 2009. The impact of distance onknowledge transfer effectiveness in multinational corpora-tions. Journal of International Management 15: 1–14.

Anand J, Delios A. 2002. Absolute and relative resources asdeterminants of international acquisitions. StrategicManagement Journal 23(2): 119–134.

Asmussen CG, Goerzen A. 2013. Unpacking dimensions offoreignness: firm-specific capabilities and internationaldispersion in regional, cultural, and institutional space.Global Strategy Journal 3(2): 127–149.

Barkema HG, Vermeulen F. 1998. International expansionthrough start-up or acquisition: a learning perspective.Academy of Management Journal 41: 7–26.

Barkema HG, Bell JHJ, Pennings JM. 1996. Foreign entry,cultural barriers, and learning. Strategic ManagementJournal 17(2): 151–166.

Batson A. 2007. China’s merger boom continues. Wall StreetJournal – Eastern Edition 249(10): C5.

Bell GR, Filatotchev I, Rasheed A. 2012. The liability offoreignness in capital markets: sources and remedies.Journal of International Business Studies 43(2):107–122.

Berry H, Guillen MF, Zhou N. 2010. An institutionalapproach to cross-national distance. Journal of Interna-tional Business Studies 41(9): 1460–1480.

Brandt L, Li H. 2003. Bank discrimination in transition econ-omies: ideology, information, or incentives? Journal ofComparative Economics 31: 387–413.

Brewer PA. 2007. Operationalizing psychic distance: arevised approach. Journal of International Marketing 15:44–66.

Brouthers KD, Brouthers L. 2001. Explaining the nationalcultural distance paradox. Journal of InternationalBusiness Studies 32(1): 177–189.

Brouthers LE, O’Donnell E, Hadjimarcou J. 2005. Genericproduct strategies for emerging market exports into triadnation markets: a mimetic isomorphism approach. Journalof Management Studies 42: 225–245.

Calhoun MA. 2002. Unpacking liability of foreignness:identifying culturally driven external and internal sourcesof liability for the foreign subsidiary. Journal of Interna-tional Management 8(3): 301–321.

Chen TJ. 2006. Liability of foreignness and entry modechoice: Taiwanese firms in Europe. Journal of BusinessResearch 59: 288–294.

Delios A, Henisz WJ. 2000. Japanese firms’ investment strat-egies in emerging economies. Academy of ManagementJournal 43: 305–323.

Delios A, Henisz WJ. 2003. Political hazards, experience, andsequential entry strategies: the international expansion ofJapanese firms, 1980-1998. Strategic Management Journal24(11): 1153–1164.

Demirbag M, Glaister KW, Tatoglu E. 2007. Institutional andtransaction cost influences on MNEs’ ownership strategiesof their affiliates: evidence from an emerging market.Journal of World Business 42: 418–434.

Denk N, Kaufmann L, Roesch J-F. 2012. Liabilities offoreignness revisited: a review of contemporary studiesand recommendations for future research. Journal ofInternational Management 18: 322–334.

Doukas JA, Lang LHP. 2003. Foreign direct investment,diversification, and firm performance. Journal ofInternational Business Studies 34(2): 153–172.

Dow D, Ferencikova S. 2010. More than just national culturaldistance: testing new distance scales on FDI in Slovakia.International Business Review 19: 46–58.

DowD,Karunaratna A. 2006. Developing amultidimensionalinstrument to measure psychic distance stimuli. Journal ofInternational Business Studies 37: 578–602.

Drogendijk R, Slangen A. 2006. Hofstede, Schwartz, ormanagerial perceptions? The effects of different culturaldistance measures on establishment mode choices bymultinational enterprises. International Business Review15: 361–380.

Dunning JH. 1993.Multinational Enterprises and the GlobalEconomy. Addison-Wesley: Wokingham, U.K.

Dunning JH. 2000. The eclectic paradigm as an envelope foreconomic and business theories of mne activity.International Business Review 9: 163–190.

Dunning JH, Lundan SM. 2008a. Institutions and the OLIparadigm of the multinational enterprise. Asia PacificJournal of Management 25: 573–593.

Dunning JH, Lundan SM. 2008b. Multinational Enterprisesand the Global Economy (2nd edn). Edward Elgar:Cheltenham, U.K.

Eden L, Dai L. 2011. Rethinking the O in Dunning’sOLI/eclectic paradigm. Multinational Business Review 18:13–34.

Eden L, Miller SR. 2001. Opening the black box: multina-tionals and the costs of doing business abroad. Paperpresented at the Academy ofManagement AnnualMeeting,Washington, D.C.

Eden L, Miller SR. 2004. Distance matters: liability of for-eignness, institutional distance, and ownership strategy. InThe Evolving Theory of the Multinational Firm (Advancesin International Management Vol. 16), Hitt MA, ChengJLC (eds). Elsevier: Amsterdam, The Netherlands;187–221.



Elango B, Pattnaik C. 2011. Learning before making the bigleap. Management International Review 51(4): 461–481.

Ewing MT, Windisch L, Newton FJ. 2010. Corporate reputa-tion in the People’s Republic of China: a B2B perspective.Industrial Marketing Management 39: 728–736.

Fama EF, Jensen MC. 1983. Separation of ownership andcontrol. Journal of Law and Economics 26(2): 301–326.

Fiss PC, Hirsch PM. 2005. The discourse of globalization:framing and sensemaking of an emerging concept.American Sociological Review 70(1): 29–52.

Florida R. 2005. The world is spiky. Atlantic MonthlyOctober: 48–50.

Freeman J, Carroll GR, Hannan MT. 1983. The liability ofnewness: age dependence in organizational death rates.American Sociological Review 48: 692–710.

Friedman TL. 2005. The World Is Flat: A Brief History of theTwenty-First Century (1st edn). Farrar, Straus, and Giroux:New York.

Garcia-Canal E, Guillen MF. 2008. Risk and the strategy offoreign location choice in regulated industries. StrategicManagement Journal 29(10): 1097–1115.

Gaur AS, Lu JW. 2007. Ownership strategies and survival offoreign subsidiaries: impacts of institutional distance andexperience. Journal of Management 33(1): 84–110.

Ghemawat P. 2001. Distance still matters: the hard reality ofglobal expansion. Harvard Business Review 79: 137–147.

Ghemawat P. 2003. Semiglobalization and internationalbusiness strategy. Journal of International Business Studies34(2): 138–152.

Globerman S, Shapiro D. 2009. Economic and strategic con-siderations surrounding Chinese FDI in the United States.Asia Pacific Journal of Management 26: 163–183.

GrahamEM,MarchickDM. 2006.U.S. National Security andForeign Direct Investment. Institute for InternationalEconomics: Washington, D.C.

HannanM, Freeman J. 1977. The population ecology of orga-nizations. American Journal of Sociology 92: 929–964.

Henisz WJ. 2000. The institutional environment for multina-tional investment. Journal of Law, Economics, and Organi-zation 16(2): 334–364.

Henisz WJ, Delios A. 2001. Uncertainty, imitation, and plantlocation: Japanese multinational corporations, 1990-1996.Administrative Science Quarterly 46: 443–475.

Henisz WJ, Williamson OE. 1999. Comparative economicorganization: within and between countries. Business &Politics 1(3): 261–277.

Henisz WJ, Zelner BA. 2004. Explicating political hazardsand safeguards: a transaction cost politics approach.Industrial and Corporate Change 13: 901–915.

Henisz WJ, Dorobantu S, Nartey L. 2014. Spinning gold: thefinancial returns to stakeholder engagement. StrategicManagement Journal 35(12): 1727–1748.

Hennart JF. 2001. Theories of the multinational enterprise. InOxford Handbook of International Business, Rugman AM,Brewer T (eds). Oxford University Press: Oxford, U.K.;141–170.

Hennart JF, Roehl T, Zeng M. 2002. Do exits proxy a lia-bility of foreignness? The case of Japanese exits fromthe U.S. Journal of International Management 8:241–264.

Hill CWL, Hwang P, Kim WC. 1990. An eclectic theory ofthe choice of international entry mode. Strategic Manage-ment Journal 11(2): 117–128.

Hirschberg JG, Sheldon IM, Dayton JR. 1994. An analysis ofbilateral intra-industry trade in the food processing sector.Applied Economics 26: 159–167.

Hitt MA, Li H, Worthington WJ. 2005. Emerging markets aslearning laboratories: learning behaviors of local firms andforeign entrants in different institutional contexts.Manage-ment and Organization Review 1: 353–380.

Holburn GLF, Zelner BA. 2010. Political capabilities, policyrisk, and international investment strategy: evidence fromthe global electric power generation industry. StrategicManagement Journal 31(12): 1290–1315.

Hymer SH. 1960. The International Operations of NationalFirms: A Study of Direct Foreign Investment. MIT Press:Cambrige, MA.

Hymer SH. 1976. The International Operations of NationalFirms: A Study of Direct Foreign Investment. MIT Press:Cambridge, MA.

Jackson G, Deeg R. 2008. Comparing capitalisms: under-standing institutional diversity and its implications for inter-national business. Journal of International Business Studies39: 540–561.

Jensen MC, Meckling WH. 1976. Theory of the firm: mana-gerial behavior, agency costs and ownership structure.Journal of Financial Economics 3: 305–360.

Johanson J, Vahlne JE. 1977. The internationalization processof the firm: a model of knowledge development andincreasing foreign market commitments. Journal ofInternational Business Studies 8: 25–34.

Johanson J, Vahlne JE. 1990. The mechanism of internation-alism. International Marketing Review 7(4): 11–24.

Johanson J, Wiedersheim-Paul F. 1975. The internationaliza-tion of the firm: four Swedish cases. Journal of Manage-ment Studies 12: 305–322.

Kimino S, Saal DS, Driffield N. 2007. Macro determinants ofFDI inflows to Japan: an analysis of source countrycharacteristics. The World Economy 30(3): 446–469.

King G, Zeng L. 2001. Logistic regression in rare events data.Political Analysis 9: 137–163.

Kobrin SJ. 1987. Testing the bargaining hypothesis in themanufacturing sector in developing countries. InternationalOrganization 41: 609–638.

Kogut B, Singh H. 1988. The effect of national culture on thechoice of entry mode. Journal of International BusinessStudies 19: 411–432.

Kogut B, Zander U. 1992. Knowledge of the firm, combina-tive capabilities, and the replication of technology.Organi-zation Science 3: 383–397.

Kostova T. 1996. Success of the transnational transfer oforganizational practices withinmultinational companies.



Unpublished doctoral dissertation, University of Minne-sota, Minneapolis, MN.

Kostova T, Zaheer S. 1999. Organizational legitimacy underconditions of complexity: the case of the multinationalenterprise. Academy of Management Review 24: 64–81.

La Porta R, Lopez-de-Silanes F, Shleifer A, Vishny RW.1998. Law and finance. Journal of Political Economy106: 1113–1154.

Li JY, Stucchi T, Visholm A, Jansen JS. 2013. Chinesemultinationals in denmarktesting the eclectic frameworkand internalization theory. Multinational Business Review21: 65–86.

Lu J, Liu X, Wang H. 2011. Motives for outward FDI ofChinese private firms: firm resources, industry dynamics,and government policies. Management and OrganizationReview 7: 223–248.

Luo Y, Zhao H, Wang Y, Xi Y. 2011. Venturing abroad byemerging market enterprises. Management InternationalReview 51(4): 433–459.

Makino S, Delios A. 1996. Local knowledge transfer andperformance implications for alliance formation in Asia.Journal of International Business Studies 27: 905–927.

Martin X, Swaminathan A, Tihanyi L. 2007. Modeling inter-national expansion. Research Methodology in Strategy andManagement 4: 103–119.

Meschi PX. 2009. Government corruption and foreign stakesin international joint ventures in emerging economies. AsiaPacific Journal of Management 26: 241–261.

Mezias JM. 2002. Identifying liabilities of foreignness andstrategies to minimize their effects: the case of labor lawsuitjudgments in the United States. Strategic ManagementJournal 23(3): 229–244.

Miller SR, ParkheA. 2002. Is there a liability of foreignness inglobal banking? An empirical test of banks’ X-efficiency.Strategic Management Journal 23(1): 55–75.

Moeller M, Harvey M, Griffith D, Richey G. 2013. Theimpact of country-of-origin on the acceptance of foreignsubsidiaries in host countries: an examination of the‘liability-of-foreignness’. International Business Review22(1): 89–99.

Montgomery CA, Wernerfelt B. 1988. Diversification,Richardian rents, and Tobin’s Q. RAND Journal ofEconomics 19: 623–632.

Nachum L, Zaheer S, Gross S. 2008. Does it matter wherecountries are? Proximity to knowledge, markets andresources, and MNE location choices. ManagementScience 54: 1252–1265.

Nelson RR, Rosenberg N. 1993. Technical innovation andnational systems. In National Innovation Systems, NelsonRR (eds). Oxford University Press: New York; 3–21.

Nelson RR, Winter SG. 1982. An Evolutionary Theory ofEconomic Change. Belknap Press of Harvard UniversityPress: Cambridge, MA.

Padmanabhan P, Cho KR. 1996. Ownership strategy for aforeign affiliate: an empirical investigation of japanesefirms. Management International Review 36: 45–65.

Pajunen K. 2008. Institutions and inflows of foreign directinvestment: a fuzzy-set analysis. Journal of InternationalBusiness Studies 39: 652–669.

Park SH, Ungson GR. 1997. The effect of national culture,organizational complementarity, and economic motivationon joint venture dissolution. Academy of ManagementJournal 40: 279–307.

Portugal Ferreira M, Frias Pinto C, Ribeiro Serra FA, GasparLF. 2011. John Dunning’s influence in internationalbusiness/strategy research: a bibliometric study in thestrategic management journal. Journal of StrategicManagement Education 7: 77–99.

Procher V. 2011. Agglomeration effects and the location ofFDI: evidence from French first-time movers. Annals ofRegional Science 46(2): 295–312.

Qian G, Li L, Rugman AM. 2013. Liability of countryforeignness and liability of regional foreignness: theireffects on geographic diversification and firm performance.Journal of International Business Studies 44(6): 635–647.

Singh N, Kundu S. 2002. Explaining the growth ofe-commerce corporations (ECCs): an extension and appli-cation of the eclectic paradigm. Journal of InternationalBusiness Studies 33: 679–697.

Slangen AHL. 2013. Greenfield or acquisition entry? Theroles of policy uncertainty and MNE legitimacy in hostcountries. Global Strategy Journal 3(3): 262–280.

Slangen AHL, Beugelsdijk S. 2010. The impact of institu-tional hazards on foreign multinational activity: a contin-gency perspective. Journal of International BusinessStudies 41(6): 980–995.

Steensma HK, Marino L, Weaver KM, Dickson PH. 2000.The influence of national culture on the formation oftechnology alliances by entrepreneurial firms. Academy ofManagement Journal 43: 951–973.

Swaminathan A, Delacroix J. 1991. Differentiation within anorganizational population: additional evidence from thewine industry. Academy of Management Journal 34:679–692.

Tallman SB. 1991. Strategic management models andresource–based strategies among MNEs in a host market.Strategic Management Journal 12(1): 69–82.

Tallman SB, Li J. 1996. Effects of international diversity andproduct diversity on the performance of multinationalfirms, Academy of Management Journal 39: 179–196.

Taylor CT. 2000. The impact of host country governmentpolicy on U.S. multinational investment decisions. TheWorld Economy 23(5): 635–647.

Teece DJ. 1986. Profiting from technological innovation:implications for integration, collaboration, licensing, andpublic policy. Research Policy 15(6): 285–305.

UNCTAD. 1996.World Investment Report 1996: Investment,Trade, and International Policy Arrangements. UnitedNations: Geneva, Switzerland.

UNCTAD. 2014.World Investment Report 2014: Investing inthe SDGs: An Action Plan. United Nations: Geneva,Switzerland.



Vernon R. 1979. Foreign Production of Technology-IntensiveProducts by U.S.-Based Multinational Enterprises.National Technical Information Service: Springfield, VA.

Wang N. 2007. Why Chinese automobile manufacturers arediscriminated in foreign countries, Commercial VehicleNews, 4 Aug: 21–23.

Werner S. 2002. Recent development in internationalmanagement research: a review of 20 top managementjournals. Journal of Management 28: 277–305.

Whitley R. 1992. Business Systems in East Asia: Firms,Markets, and Societies. SAGE Publications: London, U.K.

XuD, Shenkar O. 2002. Institutional distance and themultina-tional enterprise. Academy of Management Review 27:608–618.

Xu D, Pan Y, Beamish PW. 2004. The effect of regulativeand normative distances onMNE ownership and expatriatestrategies. Management International Review 44:285–307.

Zaheer S. 1995. Overcoming the liability of foreignness.Academy of Management Journal 38: 341–363.

Zaheer S, Mosakowski E. 1997. The dynamics of theliability of foreignness: a global study of survival in fi-nancial services. Strategic Management Journal 18(6):439–463.

ZhouN, GuillénMF. 2015. From home country to home base:a dynamic approach to the liability of foreignness. StrategicManagement Journal 36(4): 907–917.



Copyright of Global Strategy Journal is the property of Wiley-Blackwell and its content maynot be copied or emailed to multiple sites or posted to a listserv without the copyright holder'sexpress written permission. However, users may print, download, or email articles forindividual use.

categorizing zhou guillen - management department

Documents