《国际政治经济学》文献资料（Use of History and Sociology in International Political Economy）Familiarity Breeds Investment. Diaspora Networks and International Investment

American Political Science Review Vol.104.No.3 August 2010 doi:10.1017/S0003055410000201 Familiarity Breeds Investment: Diaspora Networks and International Investment DAVID LEBLANG University of Virginia That explains cross-national patterns of international portfolio and foreign direct investment (FDI)?While existing explanations focus on the credibility of a policy maker's commitment,we emphasize the role of diaspora networks.We hypothesize that diaspora networks-connections between migrants residing in investing countries and their home country-influence global investment by reducing transaction and information costs.This hypothesis is tested using dyadic cross-sectional data for both portfolio and FDI.The findings indicate that even after controlling for a multitude of factors, disapora networks have both a substantively significant effect and a statistically significant effect on cross-border investment. igration is an increasingly important facet and Prskawetz 1998).There is growing evidence that of the global political economy.Like flows emigres do benefit their homeland:return migration of commodities and capital,flows of labor constitutes a flow of knowledge of skills.and remit- have increased dramatically in recent years.The United tances provide enormous flows of capital. Nations'(UN's)Population Division put the share of In this article,we add another layer to the liter- the world's population residing outside their nation of ature on global migration by arguing that diaspora birth at almost 200 million people-or approximately networks-cultural and/or familial linkages between 3%of the world's population-in 2007.That number migrant communities in the investing country and the continues to grow:the Organisation for Economic Co- migrant's country of origin-play a pivotal role in operation and Development (OECD;2006)estimates the global allocation of capital.We provide some of that the number of legal immigrants entering OECD the first quantitative evidence that diaspora networks countries stands at about 3 million annually.Just as act as a conduit for both portfolio and foreign di- with flows of commodities and capital,scholars have rect investment (FDI)across country pairs.A focus devoted enormous energies to the exploration of the on diasporas-an explicitly cross-border phenomena- consequences of immigration.We know that migration also provides a fresh perspective on the factors influ- is associated with a rise in support for extreme right encing international investment.The existing literature wing parties and that it has mixed effects on social on investment is primarily monadic,emphasizing the spending and on wages in the destination country (e.g., importance of credible commitments and institutional Borjas 1999;Borjas,Freeman,and Katz 1996;Knigge quality in the countries that seek global capital(e.g., 1998). Alfaro,Kelemli-Ozcan,and Volosovych 2006;Buthe Less is known,however,about the consequences of and Milner 2008:Jensen 2003). migration for the migrant's country of origin.Some We do not question the importance of credible com- scholars argue that migration results in a"brain drain," mitments.However,focusing on institutional design whereby educated and skilled members of a country only gets us part of the way toward understanding pat- leave in search of better wages.Others argue that this terns of international investment.Because diasporas brain drain is offset because the prospect of leaving provide connections between home and host countries. provides an incentive for those left behind to invest these networks facilitate cross-border investment in in their own human capital(e.g.,Stark,Helmenstein, numerous ways.Broadly speaking,migrant networks foster a greater degree of familiarity between home and host countries than may occur in their absence.Just as David Leblang is J.Wilson Newman Professor of Governance. Department of Politics,University of Virginia,P.O.Box 400787, migrants may have a taste for commodities produced Charlottesville,VA 22904 (leblang@virginia.edu). in their home country,they may also have a preference This article originally circulated under the title "Diaspora Bonds for home country investments-leading them to invest and Cross-Border Capital."I am grateful to Zane Kelly and Jessica money in their country of origin.Diaspora networks Teets for outstanding research assistance.Lee Alston,Ben Ansell, Andy Baker.Bernd Beber.William Bernhard.David Brown,Tim can also help decrease asymmetries of information Buithe,Steve Chan,Rafaela Dancygier,Jennifer Fitzgerald,John that,from the perspective of the theory of portfolio Freeman,Daniel Gingerich,Jude Hays,Nathan Jensen,Joseph investment.can result in a less than optimal portfo- Jupille,Moonhawk Kim.Robert McKnown.Sonal Pandya.Tom lio.The reduction in information asymmetries works Pepinsky,Andy Rose,Idean Salehyan,Kathryn Sikkink.David Singer,Andy Sobel,Enrico Spolaore,Michael Tomz,Romain through two channels.First,migrant communities in Wacziarg,and Jennifer Wolak provided helpful comments and/or destination countries can provide investors with infor- generously shared their data.I am also grateful to Roger,Keith,Pete, mation about their homeland-information regarding and John for helping me get through extensive revisions.Replication the tastes of consumers in their country of origin- materials are available at http://journals.cambridge.org/psr2010008 that can influence the decision of investors to invest The research was funded in part by a developmental grant from the National Institute of Child Health and Human Development there.Second,diaspora networks can have an indirect (NICHD)-funded University of Colorado Population Center(grant effect on investment because they may have knowl- R1HD51146). edge about investment opportunities,information 584

American Political Science Review Vol. 104, No. 3 August 2010 doi:10.1017/S0003055410000201 Familiarity Breeds Investment: Diaspora Networks and International Investment DAVID LEBLANG University of Virginia What explains cross-national patterns of international portfolio and foreign direct investment (FDI)? While existing explanations focus on the credibility of a policy maker’s commitment, we emphasize the role of diaspora networks. We hypothesize that diaspora networks—connections between migrants residing in investing countries and their home country—influence global investment by reducing transaction and information costs. This hypothesis is tested using dyadic cross-sectional data for both portfolio and FDI. The findings indicate that even after controlling for a multitude of factors, disapora networks have both a substantively significant effect and a statistically significant effect on cross-border investment. Migration is an increasingly important facet of the global political economy. Like flows of commodities and capital, flows of labor have increased dramatically in recent years. The United Nations’ (UN’s) Population Division put the share of the world’s population residing outside their nation of birth at almost 200 million people—or approximately 3% of the world’s population—in 2007. That number continues to grow: the Organisation for Economic Co￾operation and Development (OECD; 2006) estimates that the number of legal immigrants entering OECD countries stands at about 3 million annually. Just as with flows of commodities and capital, scholars have devoted enormous energies to the exploration of the consequences of immigration. We know that migration is associated with a rise in support for extreme right wing parties and that it has mixed effects on social spending and on wages in the destination country (e.g., Borjas 1999; Borjas, Freeman, and Katz 1996; Knigge 1998). Less is known, however, about the consequences of migration for the migrant’s country of origin. Some scholars argue that migration results in a “brain drain,” whereby educated and skilled members of a country leave in search of better wages. Others argue that this brain drain is offset because the prospect of leaving provides an incentive for those left behind to invest in their own human capital (e.g., Stark, Helmenstein, David Leblang is J. Wilson Newman Professor of Governance, Department of Politics, University of Virginia, P.O. Box 400787, Charlottesville, VA 22904 (leblang@virginia.edu). This article originally circulated under the title “Diaspora Bonds and Cross-Border Capital.” I am grateful to Zane Kelly and Jessica Teets for outstanding research assistance. Lee Alston, Ben Ansell, Andy Baker, Bernd Beber, William Bernhard, David Brown, Tim Buthe, Steve Chan, Rafaela Dancygier, Jennifer Fitzgerald, John ¨ Freeman, Daniel Gingerich, Jude Hays, Nathan Jensen, Joseph Jupille, Moonhawk Kim, Robert McKnown, Sonal Pandya, Tom Pepinsky, Andy Rose, Idean Salehyan, Kathryn Sikkink, David Singer, Andy Sobel, Enrico Spolaore, Michael Tomz, Romain Wacziarg, and Jennifer Wolak provided helpful comments and/or generously shared their data. I am also grateful to Roger, Keith, Pete, and John for helping me get through extensive revisions. Replication materials are available at http://journals.cambridge.org/psr2010008. The research was funded in part by a developmental grant from the National Institute of Child Health and Human Development (NICHD)–funded University of Colorado Population Center (grant R1 HD51146). and Prskawetz 1998). There is growing evidence that emigr ´ es do benefit their homeland: return migration ´ constitutes a flow of knowledge of skills, and remit￾tances provide enormous flows of capital. In this article, we add another layer to the liter￾ature on global migration by arguing that diaspora networks—cultural and/or familial linkages between migrant communities in the investing country and the migrant’s country of origin—play a pivotal role in the global allocation of capital. We provide some of the first quantitative evidence that diaspora networks act as a conduit for both portfolio and foreign di￾rect investment (FDI) across country pairs. A focus on diasporas—an explicitly cross-border phenomena— also provides a fresh perspective on the factors influ￾encing international investment. The existing literature on investment is primarily monadic, emphasizing the importance of credible commitments and institutional quality in the countries that seek global capital (e.g., Alfaro, Kelemli-Ozcan, and Volosovych 2006; Buthe ¨ and Milner 2008; Jensen 2003). We do not question the importance of credible com￾mitments. However, focusing on institutional design only gets us part of the way toward understanding pat￾terns of international investment. Because diasporas provide connections between home and host countries, these networks facilitate cross-border investment in numerous ways. Broadly speaking, migrant networks foster a greater degree of familiarity between home and host countries than may occur in their absence. Just as migrants may have a taste for commodities produced in their home country, they may also have a preference for home country investments—leading them to invest money in their country of origin. Diaspora networks can also help decrease asymmetries of information that, from the perspective of the theory of portfolio investment, can result in a less than optimal portfo￾lio. The reduction in information asymmetries works through two channels. First, migrant communities in destination countries can provide investors with infor￾mation about their homeland—information regarding the tastes of consumers in their country of origin— that can influence the decision of investors to invest there. Second, diaspora networks can have an indirect effect on investment because they may have knowl￾edge about investment opportunities, information 584

American Political Science Review Vol.104.No.3 about regulations and procedures,or familiarity with asset pricing model.The standard international cap- language and customs that can decrease the transaction ital asset pricing model(ICAPM)predicts that in the costs associated with cross-border investment. absence of information asymmetries and transactions We test our argument using a dyadic data set com- costs.investors should hold domestic assets in their posed of both portfolio and FDI from as many as 56 portfolio in proportion to their country's share of source countries into up to 154 destination countries global capital market capitalization.2 However con- for the year 2002.The use of portfolio and FDI al- vincing in theory,overwhelming empirical evidence lows us to be more general about the implications demonstrates that investor behavior deviates consid- of our findings as these investment types are funda- erably from this benchmark model.The empirical find- mentally different.While portfolio investors purchase ing that investors forgo substantial gains by investing stocks and bonds in open markets,foreign direct in- at home rather than abroad has spawned a huge lit- vestors purchase fixed shares in plants or in machinery. erature trying to explain this "home bias,"a systemic More important,within a country,portfolio investment preference for assets that are available in their home opportunities are constrained by the shares issued by market (e.g.,Lewis 1999). corporate or government entities,whereas FDI possi- What does the literature say about the lack of inter- bilities are unbounded in both their content and their national diversification?In early work on the "home ownership stake. bias,"French and Poterba(1991)argue that investors Giving pride of place to migrant networks in ex- purchase domestic assets as a consequence of what they plaining the cross-national allocation of capital al- call"familiarity"effects.Tesar and Werner (1995)are a lows us to speak to a number of seemingly disparate bit more specific when they attribute the taste for home literatures.Broadly speaking,the emphasis on dias- rather than foreign assets to factors such as"language pora networks as a conduit for capital flows is a nat- [and]institutional and regulator difference"(p.479).In ural extension of Keohane and Nye's (1974)work addition to taste.Coval and Moskowitz(2001)attribute on transnational-or nongovernmental-relations be- the home bias to asymmetries of information and argue tween states.Although other scholarship from inter- that investors have better information about assets sold national relations fits within the category of transna- in geographically closer markets.3 tionalism,it tends to focus on nongovernmental or In a cross-national context,it is difficult to separate intergovernmental organizations (e.g.,Keck and familiarity and cultural preferences for home products Sikkink 1998:Slaughter 2004).Emphasizing the role from informational asymmetries.To minimize prob- of migrant networks in cross-national investment pro- lems of model misspecification associated with omitted vides another mechanism-this one noninstitutional- variable bias,empirical studies of cross-border invest- by which we can understand the growing degree of ment tend to exploit gravity-type models of interna- interdependence that exists within the international tional transactions.Drawn from gravity models of in- system.From the perspective of international relations ternational trade,which in turn are (loosely)derived theory,our emphasis on migrant communities influ- from Newton's Law of Gravity,these models hold that encing their homelands allows us to speak to the grow- bilateral transactions are a positive function of the size ing interest in diaspora politics (e.g.,Shain and Barth of the two economies(their mass)and a negative func- 2003;Sheffer 2003).By privileging human networks,we tion of the distance between them.In dealing with the highlight the importance of mechanisms other than for- empirical fact that investors exhibit a home bias,schol- mal institutions for channeling economic activity(e.g., ars"augment"these gravity models with a variety of Greif1989.1993). variables.Eichengreen and Luengnaruemitchai (2006), Our arguments and evidence are developed in this Lane and Milesi-Ferretti(2004),and Rose and Spiegel article.The first section contains our theoretical dis- (2008),for example,include variables that measure cussion and embeds our argument within the context whether the country pair shares a common language,a of international asset pricing theory.The next section common border,or a common colonial heritage.These lays out the empirical model and details the statisti- cal methodology we use.In the following section,we present our central statistical results and document the We reference much of this literature in what follows.For overviews, robustness of those findings.The last section concludes see Lane and Milesi-Ferretti (2004),Lewis (1999).Portes and Rey and offers suggestions for future research. (2005).and Sarkissian and Schill (2004). 2 See Lane (2005)and Lane and Milesi-Ferretti (2004)for studies of bilateral investment that are explicitly derived from the ICAPM Elton et al.(2003)provide a textbook exposition of the capital asset pricing model. Grinblatt and Keloharju(2001),Huberman(2001),and Portes and DIASPORA NETWORKS AND INVESTMENT Rey (2005)also find that investors tend to purchase assets sold in [more proximate markets.These articles attribute this behavior to Modeling Bilateral Investment the idea that investors have better(less asymmetric)information as distance between markets decreases.Coval and Moskowitz (1999, Our argument is that migrant networks are a mech- 2001)attempt to distinguish between familiarity and information anism helping direct portfolio and FDI from the mi- asymmetries.Their 1999 study documents that investors prefer to grant's host country to their country of origin.We invest in the "familiar,"whereas their 2001 article provides some evidence in favor of the information asymmetry hypothesis.It is situate our argument within the international finance worth noting that their earlier article does not distinguish between literature and ground it in the international capital the two explanations. 585

American Political Science Review Vol. 104, No. 3 about regulations and procedures, or familiarity with language and customs that can decrease the transaction costs associated with cross-border investment. We test our argument using a dyadic data set com￾posed of both portfolio and FDI from as many as 56 source countries into up to 154 destination countries for the year 2002. The use of portfolio and FDI al￾lows us to be more general about the implications of our findings as these investment types are funda￾mentally different. While portfolio investors purchase stocks and bonds in open markets, foreign direct in￾vestors purchase fixed shares in plants or in machinery. More important, within a country, portfolio investment opportunities are constrained by the shares issued by corporate or government entities, whereas FDI possi￾bilities are unbounded in both their content and their ownership stake. Giving pride of place to migrant networks in ex￾plaining the cross-national allocation of capital al￾lows us to speak to a number of seemingly disparate literatures. Broadly speaking, the emphasis on dias￾pora networks as a conduit for capital flows is a nat￾ural extension of Keohane and Nye’s (1974) work on transnational—or nongovernmental—relations be￾tween states. Although other scholarship from inter￾national relations fits within the category of transna￾tionalism, it tends to focus on nongovernmental or intergovernmental organizations (e.g., Keck and Sikkink 1998; Slaughter 2004). Emphasizing the role of migrant networks in cross-national investment pro￾vides another mechanism—this one noninstitutional— by which we can understand the growing degree of interdependence that exists within the international system. From the perspective of international relations theory, our emphasis on migrant communities influ￾encing their homelands allows us to speak to the grow￾ing interest in diaspora politics (e.g., Shain and Barth 2003; Sheffer 2003). By privileging human networks, we highlight the importance of mechanisms other than for￾mal institutions for channeling economic activity (e.g., Greif 1989, 1993). Our arguments and evidence are developed in this article. The first section contains our theoretical dis￾cussion and embeds our argument within the context of international asset pricing theory. The next section lays out the empirical model and details the statisti￾cal methodology we use. In the following section, we present our central statistical results and document the robustness of those findings. The last section concludes and offers suggestions for future research. DIASPORA NETWORKS AND INVESTMENT Modeling Bilateral Investment Our argument is that migrant networks are a mech￾anism helping direct portfolio and FDI from the mi￾grant’s host country to their country of origin. We situate our argument within the international finance literature and ground it in the international capital asset pricing model.1 The standard international cap￾ital asset pricing model (ICAPM) predicts that in the absence of information asymmetries and transactions costs, investors should hold domestic assets in their portfolio in proportion to their country’s share of global capital market capitalization.2 However con￾vincing in theory, overwhelming empirical evidence demonstrates that investor behavior deviates consid￾erably from this benchmark model. The empirical find￾ing that investors forgo substantial gains by investing at home rather than abroad has spawned a huge lit￾erature trying to explain this “home bias,” a systemic preference for assets that are available in their home market (e.g., Lewis 1999). What does the literature say about the lack of inter￾national diversification? In early work on the “home bias,” French and Poterba (1991) argue that investors purchase domestic assets as a consequence of what they call “familiarity” effects. Tesar and Werner (1995) are a bit more specific when they attribute the taste for home rather than foreign assets to factors such as “language [and] institutional and regulator difference” (p. 479). In addition to taste, Coval and Moskowitz (2001) attribute the home bias to asymmetries of information and argue that investors have better information about assets sold in geographically closer markets.3 In a cross-national context, it is difficult to separate familiarity and cultural preferences for home products from informational asymmetries. To minimize prob￾lems of model misspecification associated with omitted variable bias, empirical studies of cross-border invest￾ment tend to exploit gravity-type models of interna￾tional transactions. Drawn from gravity models of in￾ternational trade, which in turn are (loosely) derived from Newton’s Law of Gravity, these models hold that bilateral transactions are a positive function of the size of the two economies (their mass) and a negative func￾tion of the distance between them. In dealing with the empirical fact that investors exhibit a home bias, schol￾ars “augment” these gravity models with a variety of variables. Eichengreen and Luengnaruemitchai (2006), Lane and Milesi-Ferretti (2004), and Rose and Spiegel (2008), for example, include variables that measure whether the country pair shares a common language, a common border, or a common colonial heritage. These 1 We reference much of this literature in what follows. For overviews, see Lane and Milesi-Ferretti (2004), Lewis (1999), Portes and Rey (2005), and Sarkissian and Schill (2004). 2 See Lane (2005) and Lane and Milesi-Ferretti (2004) for studies of bilateral investment that are explicitly derived from the ICAPM. Elton et al. (2003) provide a textbook exposition of the capital asset pricing model. 3 Grinblatt and Keloharju (2001), Huberman (2001), and Portes and Rey (2005) also find that investors tend to purchase assets sold in [more] proximate markets. These articles attribute this behavior to the idea that investors have better (less asymmetric) information as distance between markets decreases. Coval and Moskowitz (1999, 2001) attempt to distinguish between familiarity and information asymmetries. Their 1999 study documents that investors prefer to invest in the “familiar,” whereas their 2001 article provides some evidence in favor of the information asymmetry hypothesis. It is worth noting that their earlier article does not distinguish between the two explanations. 585

Familiarity Breeds Investment August 2010 measures of commonality are then interpreted in the constraints imposed by information asymmetries and context of greater familiarity,better information,lower transactions costs.6 barriers to entry,and smaller transactions costs. Before discussing the role of familiarity and informa- Scholars also interpret negative coefficients on the tion asymmetries,we note that migrant networks can (log of)distance in gravity models of international have a direct and observable impact on cross-border investment as a proxy for information asymmetries investment when they,themselves.are the actors.In his The greater the distance from a market,so the argu- study of the Korean diaspora,for example,Choi(2003) ment goes,the greater the information cost and the documents how FDI into Korea-a form of investment lower the level of investment.Arguing that distance that came late to that country-was originally spurred proxies for information costs,however,is theoretically by ethnic Koreans residing in Japan.Migrant-based unsatisfying,and,in a series of articles,Portes and Rey investment also comes from nonwealthy expatriates. (2005;Portes,Rey,and Oh 2001)employ a more direct Schuttler(2007)and Schulte(2008)study,respectively, measure of information asymmetries and transactions the investment behavior of Moroccan and Turkish mi- costs by including a measure of bilateral telephone traf- grants living in Germany.Using ethnographic and sur- fic.They find that telephone traffic is associated with vey research,both scholars conclude that diaspora in- higher levels of bilateral portfolio investment.+ vestors are well-situated investment in their homeland These approaches remain incomplete.Even mod- because of cultural and linguistic familiarity.There are els that include measures of communication flows and larger literatures that make similar points about the communication costs find that investors retain a home ways in which migrant entrepreneurs have used home bias (e.g.,Loungani,Mody,and Razin 2002).Taking country connections to channel human,physical,and this as their point of departure,a growing literature ar- investment capital.9 gues that investment across country may be driven not Cross-national models of home country investment by familiarity,but rather by"cultural affinity,"whereby are difficult to test in the aggregate because even if individuals have more trust in individuals and institu- we surveyed investors it would be surprising if they tions from countries that share common cultural char- would admit to making bad investments or to being acteristics (e.g.,Guiso,Sapienza,and Zingales 2005; privy to private information.Because our interest is in Siegel,Licht,and Schwartz 2008).Cultural similarity, understanding broad patterns of global investment,we viewed within the context of ICAPMs.would constitute embed variables capturing migrant networks in well- a more direct measure of(the lack of)information costs established empirical models;our inferences are then and should be correlated with lower transactions costs conditional on the set of conditioning variables,and and a greater ease of doing business across border. our conclusions are thus statements about average be- This is where we join the literature on cross-border havior across a set of observations investment.Rather than developing empirical proxies At the aggregate level,then,how do migrants chan- for cultural affinity,economic familiarity,and infor- nel investment?We identify two channels:one result- mation asymmetries,we argue that diaspora networks ing from increased familiarity,and another associated help link sources of investment to specific destinations. with a decrease in informational asymmetries.These are discussed in turn. The familiarity effect takes hold when investors in Migrant Networks and Cross-border Capital a source country become familiar with characteristics How do migrant networks fit into an explanation for of the migrant's homeland through their connections cross-national investment?A network can be under- to,and observation of,migrant communities that ex- stood as a group of actors that either know about or can ist in their country.This "familiarity effect"occurs learn about each other's characteristics (Granovetter as migrants provide a signal to investors that allows 1973).Scholars have long recognized the importance of social networks for fostering economic exchange when formal institutions are absent or incomplete(e.g., 6 We should note that there are other mechanisms by which migrant networks channel capital back to their home country.Studies of over- Greif 1989:North 2005).5 From this perspective,we seas migrant communities have documented the role that coethnic can identify numerous ways in which migrant net- networks play in transmitting technical information and investment works facilitate cross-border investment,ways that are capital back to their country of origin (e.g.,Saxenian,2002,2006). More recent contributions have demonstrated the importance of consistent with prior research documenting departures migrant networks for channeling capital in the form of remittances from the ICAPM of international investment.Within (Leuth and Ruiz-Arranz 2006:Ratha and Shaw 2007). this context,migrant networks influence investment 7 Choi (2003)tells the story of Kyuk-ho Shin and Son Masayoshi- by facilitating the familiarity effect and by decreasing both of whom amassed their wealth while living in Japan and both of whom engaged in millions of dollars worth of FDI into Korea. Migrants are aware of this advantage,as Schulte(2008)documents with a quote from one of her interviewees:"Logically,one has advan 4 Loungani,Mody,and Razin(2002)reach a similar conclusion re- tages as a Turk when approaching business partners [in Turkey.One garding the positive relationship between bilateral telephone traffic speaks the same language,one is aware that small talk is required and FDI and one is on the same level of understanding,one socializes,one The relational approach to economic sociology focuses on rela- has the same humour,one shares interest in specific topics.That is a tions between parties to a transaction rather than on the transaction major advantage"(p.8). itself.This view,that economic processes are "embedded"in social Examples include Freinkman(2002)on Armenia,Kapur(2001) relations,has been used to study labor markets(Granovetter 1973). and Saxenian (2002)on India,Kleinman(1996)on Israel,and Rauch business transactions (Uzzi 1996),and FDI(Bandelj 2002.2007). and Casella (2002)and Weidenbaum and Hughes(1996)on China 586

Familiarity Breeds Investment August 2010 measures of commonality are then interpreted in the context of greater familiarity, better information, lower barriers to entry, and smaller transactions costs. Scholars also interpret negative coefficients on the (log of) distance in gravity models of international investment as a proxy for information asymmetries. The greater the distance from a market, so the argu￾ment goes, the greater the information cost and the lower the level of investment. Arguing that distance proxies for information costs, however, is theoretically unsatisfying, and, in a series of articles, Portes and Rey (2005; Portes, Rey, and Oh 2001) employ a more direct measure of information asymmetries and transactions costs by including a measure of bilateral telephone traf- fic. They find that telephone traffic is associated with higher levels of bilateral portfolio investment.4 These approaches remain incomplete. Even mod￾els that include measures of communication flows and communication costs find that investors retain a home bias (e.g., Loungani, Mody, and Razin 2002). Taking this as their point of departure, a growing literature ar￾gues that investment across country may be driven not by familiarity, but rather by “cultural affinity,” whereby individuals have more trust in individuals and institu￾tions from countries that share common cultural char￾acteristics (e.g., Guiso, Sapienza, and Zingales 2005; Siegel, Licht, and Schwartz 2008). Cultural similarity, viewed within the context of ICAPMs, would constitute a more direct measure of (the lack of) information costs and should be correlated with lower transactions costs and a greater ease of doing business across border. This is where we join the literature on cross-border investment. Rather than developing empirical proxies for cultural affinity, economic familiarity, and infor￾mation asymmetries, we argue that diaspora networks help link sources of investment to specific destinations. Migrant Networks and Cross-border Capital How do migrant networks fit into an explanation for cross-national investment? A network can be under￾stood as a group of actors that either know about or can learn about each other’s characteristics (Granovetter 1973). Scholars have long recognized the importance of social networks for fostering economic exchange when formal institutions are absent or incomplete (e.g., Greif 1989; North 2005).5 From this perspective, we can identify numerous ways in which migrant net￾works facilitate cross-border investment, ways that are consistent with prior research documenting departures from the ICAPM of international investment. Within this context, migrant networks influence investment by facilitating the familiarity effect and by decreasing 4 Loungani, Mody, and Razin (2002) reach a similar conclusion re￾garding the positive relationship between bilateral telephone traffic and FDI. 5 The relational approach to economic sociology focuses on rela￾tions between parties to a transaction rather than on the transaction itself. This view, that economic processes are “embedded” in social relations, has been used to study labor markets (Granovetter 1973), business transactions (Uzzi 1996), and FDI (Bandelj 2002, 2007). constraints imposed by information asymmetries and transactions costs.6 Before discussing the role of familiarity and informa￾tion asymmetries, we note that migrant networks can have a direct and observable impact on cross-border investment when they, themselves, are the actors. In his study of the Korean diaspora, for example, Choi (2003) documents how FDI into Korea—a form of investment that came late to that country—was originally spurred by ethnic Koreans residing in Japan.7 Migrant-based investment also comes from nonwealthy expatriates. Schuttler (2007) and Schulte (2008) study, respectively, ¨ the investment behavior of Moroccan and Turkish mi￾grants living in Germany. Using ethnographic and sur￾vey research, both scholars conclude that diaspora in￾vestors are well-situated investment in their homeland because of cultural and linguistic familiarity.8 There are larger literatures that make similar points about the ways in which migrant entrepreneurs have used home country connections to channel human, physical, and investment capital.9 Cross-national models of home country investment are difficult to test in the aggregate because even if we surveyed investors it would be surprising if they would admit to making bad investments or to being privy to private information. Because our interest is in understanding broad patterns of global investment, we embed variables capturing migrant networks in well￾established empirical models; our inferences are then conditional on the set of conditioning variables, and our conclusions are thus statements about average be￾havior across a set of observations. At the aggregate level, then, how do migrants chan￾nel investment? We identify two channels: one result￾ing from increased familiarity, and another associated with a decrease in informational asymmetries. These are discussed in turn. The familiarity effect takes hold when investors in a source country become familiar with characteristics of the migrant’s homeland through their connections to, and observation of, migrant communities that ex￾ist in their country. This “familiarity effect” occurs as migrants provide a signal to investors that allows 6 We should note that there are other mechanisms by which migrant networks channel capital back to their home country. Studies of over￾seas migrant communities have documented the role that coethnic networks play in transmitting technical information and investment capital back to their country of origin (e.g., Saxenian, 2002, 2006). More recent contributions have demonstrated the importance of migrant networks for channeling capital in the form of remittances (Leuth and Ruiz-Arranz 2006; Ratha and Shaw 2007). 7 Choi (2003) tells the story of Kyuk-ho Shin and Son Masayoshi— both of whom amassed their wealth while living in Japan and both of whom engaged in millions of dollars worth of FDI into Korea. 8 Migrants are aware of this advantage, as Schulte (2008) documents with a quote from one of her interviewees: “Logically, one has advan￾tages as a Turk when approaching business partners [in Turkey]. One speaks the same language, one is aware that small talk is required and one is on the same level of understanding, one socializes, one has the same humour, one shares interest in specific topics. That is a major advantage” (p. 8). 9 Examples include Freinkman (2002) on Armenia, Kapur (2001) and Saxenian (2002) on India, Kleinman (1996) on Israel, and Rauch and Casella (2002) and Weidenbaum and Hughes (1996) on China. 586

American Political Science Review Vol.104.No.3 investors to make inferences about the quality of labor. ern European and North American investors after the the work ethic,and/or the business culture that exists opening of markets in Eastern Europe.Investment in in a particular destination.A migrant community from Eastern European countries,she writes,was "often India residing in the United States,for example,can based on ethnic ties between sizable and relatively provide U.S.investors with a signal of the work ethic. affluent expatriate communities and their home coun- labor quality.and business culture that exists in India. tries"(p.421).There was an informational advantage These signals enhance the quality of information that as "firms amassed information about investment op- U.S.investors have about India allowing them to make portunities through their business or personal ties" forecasts about their ability to invest in potentially (p.412).This informational advantage can translate profitable assets offered on the Indian market.Kapur into higher than average expected returns if the mi- (2001),in his study of the Indian community residing in grant him-or herself has a higher level of human the United States,explains how the mere presence of capital.In his assessment of the Armenian diaspora, that community enhances investment opportunities in Freinkman(2002)notes that "when compared to the India:"Companies like Yahoo,Hewlett Packard and average economic agent,diaspora businessmen and General Electric have opened R&D centers in India professionals face a lower risk of becoming the first largely because of the confidence engendered by the movers.They benefit from a specific informational ad- presence of many Indians working in their US opera- vantage:common cultural background and established tions.This points to the cognitive effects arising from social links between diaspora and local entrepreneurs the projection of a coherent,appealing,and progressive help them to reduce transaction costs of new entry and identity on the part of the diaspora which signals an building new partnerships"(p.4). image of prosperity and progress to potential investors Connections between migrant communities across and consumers."10 countries may enhance cross-national investment,even Along with the provision of an image of their home when these connections do not provide specific infor country,migrant networks can provide business op- mation about investment opportunities.In reference portunities by decreasing asymmetries of information to the Maghribi traders of the eleventh century,Greif and by reducing transaction costs through formal(e.g., (1989,1993)argues that this trading network was effec- business)or information(e.g.,familial)contacts in their tive because it was able to credibly threaten collective home country.Again,the case of ethnic Turks resid- punishment by all merchants if even one of them de- ing in Germany is illustrative.Turkish migrants living fected.He shows that this coethnic network was able in Germany.Schulte (2008)notes,are more likely to to share information regarding the past actions of ac- invest in their homeland rather than in,say China tors (they could communicate a reputation),something because language and cultural knowledge more than that was essential for the efficient functioning of mar- compensate for what may be marginally lower returns. kets in the absence of formal legal rules.Weidenbaum This informational advantage does not privilege only and Hughes (1996)reach a similar conclusion about individual investors,rather it can work through the the effectiveness of The Bamboo Network,remarking larger network of coethnics.In The Bamboo Network, that "If a business owner violates an agreement,he is Weidenbaum and Hughes (1996)detail the compara- blacklisted.This is far worse than being sued,because tive advantage overseas Chinese have when it comes the entire Chinese networks will refrain from doing to investing in China and argue that it goes well be- business with the guilty party"(p.51). yond commonality of language,knowledge of cultural These informational effects are important because and legal barriers,and preexisting familial connec- they help investors overcome general barriers associ- tions.Wang's study shows how ethnic Chinese residing ated with transactions costs.Migrant communities can abroad provide a"linkage between China and the rest help facilitate cross-national portfolio investment by of the world [in that they]facilitate the understanding reducing barriers to entry-through knowledge of lan- of and access to guanxi networks by other foreign in- guage,institutional rules,and/or regulatory hurdles- vestors.Without the agency of ethnic Chinese,it would that may otherwise prevent a foreign investor from pur- have been much more difficult for foreign companies chasing equities or bonds.Knowledge of on-the-ground to use informal personal networks to complement and conditions is costly (and not necessarily private)and compensate for the weak formal legal institutions in provides investors with the ability to"match"invest- China"11(Wang 2000,p.161). ments with investment opportunities that may exist. Migrant networks provide investors with an infor- This"matching"function of migrant networks has been mational advantage because they are in a position to observed in studies of international trade.where Rauch have information regarding investment opportunities and Trindade(2002)find that migrant generated infor- in their home country.Bandelj(2002)provides some mation helps match buyers with sellers,a function that evidence concerning the investment behavior of West- becomes more important as goods become increasingly heterogeneous.12 It is important to note that the homeland govern- 10 Kapur and McHale (2006)refer to this as"branding"and argue ments are aware of key role that migrants play in that the Indian diaspora has created a brand name by signaling the oprodiityduofoy Guanxi is a close personal connection between individuals.A 12 The role of migrant networks in facilitating bilateral trade has guanxi network,therefore,is one that is based on trust and reputation been studies by Gould(1994)for the United States and by Head and ather than on external monitoring or enforcement. Reis (1998)for Canada 587

American Political Science Review Vol. 104, No. 3 investors to make inferences about the quality of labor, the work ethic, and/or the business culture that exists in a particular destination. A migrant community from India residing in the United States, for example, can provide U.S. investors with a signal of the work ethic, labor quality, and business culture that exists in India. These signals enhance the quality of information that U.S. investors have about India allowing them to make forecasts about their ability to invest in potentially profitable assets offered on the Indian market. Kapur (2001), in his study of the Indian community residing in the United States, explains how the mere presence of that community enhances investment opportunities in India: “Companies like Yahoo, Hewlett Packard and General Electric have opened R&D centers in India largely because of the confidence engendered by the presence of many Indians working in their US opera￾tions. This points to the cognitive effects arising from the projection of a coherent, appealing, and progressive identity on the part of the diaspora which signals an image of prosperity and progress to potential investors and consumers.”10 Along with the provision of an image of their home country, migrant networks can provide business op￾portunities by decreasing asymmetries of information and by reducing transaction costs through formal (e.g., business) or information (e.g., familial) contacts in their home country. Again, the case of ethnic Turks resid￾ing in Germany is illustrative. Turkish migrants living in Germany, Schulte (2008) notes, are more likely to invest in their homeland rather than in, say China, because language and cultural knowledge more than compensate for what may be marginally lower returns. This informational advantage does not privilege only individual investors, rather it can work through the larger network of coethnics. In The Bamboo Network, Weidenbaum and Hughes (1996) detail the compara￾tive advantage overseas Chinese have when it comes to investing in China and argue that it goes well be￾yond commonality of language, knowledge of cultural and legal barriers, and preexisting familial connec￾tions. Wang’s study shows how ethnic Chinese residing abroad provide a “linkage between China and the rest of the world [in that they] facilitate the understanding of and access to guanxi networks by other foreign in￾vestors. Without the agency of ethnic Chinese, it would have been much more difficult for foreign companies to use informal personal networks to complement and compensate for the weak formal legal institutions in China”11 (Wang 2000, p. 161). Migrant networks provide investors with an infor￾mational advantage because they are in a position to have information regarding investment opportunities in their home country. Bandelj (2002) provides some evidence concerning the investment behavior of West- 10 Kapur and McHale (2006) refer to this as “branding” and argue that the Indian diaspora has created a brand name by signaling the potential productivity and trustworthiness of their countrymen. 11 Guanxi is a close personal connection between individuals. A guanxi network, therefore, is one that is based on trust and reputation rather than on external monitoring or enforcement. ern European and North American investors after the opening of markets in Eastern Europe. Investment in Eastern European countries, she writes, was “often based on ethnic ties between sizable and relatively affluent expatriate communities and their home coun￾tries” (p. 421). There was an informational advantage as “firms amassed information about investment op￾portunities through their business or personal ties” (p. 412). This informational advantage can translate into higher than average expected returns if the mi￾grant him- or herself has a higher level of human capital. In his assessment of the Armenian diaspora, Freinkman (2002) notes that “when compared to the average economic agent, diaspora businessmen and professionals face a lower risk of becoming the first movers. They benefit from a specific informational ad￾vantage: common cultural background and established social links between diaspora and local entrepreneurs help them to reduce transaction costs of new entry and building new partnerships” (p. 4). Connections between migrant communities across countries may enhance cross-national investment, even when these connections do not provide specific infor￾mation about investment opportunities. In reference to the Maghribi traders of the eleventh century, Greif (1989, 1993) argues that this trading network was effec￾tive because it was able to credibly threaten collective punishment by all merchants if even one of them de￾fected. He shows that this coethnic network was able to share information regarding the past actions of ac￾tors (they could communicate a reputation), something that was essential for the efficient functioning of mar￾kets in the absence of formal legal rules. Weidenbaum and Hughes (1996) reach a similar conclusion about the effectiveness of The Bamboo Network, remarking that “If a business owner violates an agreement, he is blacklisted. This is far worse than being sued, because the entire Chinese networks will refrain from doing business with the guilty party” (p. 51). These informational effects are important because they help investors overcome general barriers associ￾ated with transactions costs. Migrant communities can help facilitate cross-national portfolio investment by reducing barriers to entry—through knowledge of lan￾guage, institutional rules, and/or regulatory hurdles— that may otherwise prevent a foreign investor from pur￾chasing equities or bonds. Knowledge of on-the-ground conditions is costly (and not necessarily private) and provides investors with the ability to “match” invest￾ments with investment opportunities that may exist. This “matching” function of migrant networks has been observed in studies of international trade, where Rauch and Trindade (2002) find that migrant generated infor￾mation helps match buyers with sellers, a function that becomes more important as goods become increasingly heterogeneous.12 It is important to note that the homeland govern￾ments are aware of key role that migrants play in 12 The role of migrant networks in facilitating bilateral trade has been studies by Gould (1994) for the United States and by Head and Reis (1998) for Canada. 587

Familiarity Breeds Investment August 2010 facilitating investment and in further integrating their receive capital tend to have better institutions,more countries into the global economy.Gamlen (2008) stable economic policies,and better governance.14 notes that even developed states such as New Zealand. All empirical models are vulnerable to problems of Ireland,and Israel have well-developed diaspora en- omitted variable bias.We choose a very conservative gagement strategies designed to encourage active eco- strategy and estimate a cross-sectional model of bilat- nomic behavior on the part of their external citizens eral investment that contains fixed effects for both the In a response to a question from a member of their source and destination countries.Although this estima- parliament about how Scotland intends to engage their tion strategy sacrifices some richness in that we cannot diaspora residing in Canada,Scottish Executive Min- explore variation in host and homeland domestic insti- ister for Parliament declared,"The Scottish Executive tutions and policies,it allows us to concentrate on bi- intends to engage with and mobilize the Scottish dias- lateral characteristics that influence dyadic investment. pora to further Scotland's interests for the long-term Concretely,we write our model as benefit of our economy and society.We aim to encour- age the diaspora's active participation and engagement log0y)sd=δs+δd+φlog(I)+βXd+csd, (1) in promoting Scotland as a great country to visit,live, learn,work,do business and invest."13 Because they increase familiarity and decrease in- where ysd is the level of either portfolio or FDI from formation asymmetries,we hypothesize that larger mi- a source country (s)into a destination country (d).15 grant networks,all other things being equal,will exert Because we are working with a cross-section of country a positive influence on cross-border investment.The dyads,we control for unmeasured source and destina- effect of migrant networks should increase when the tion country characteristics through the inclusion of migrants themselves are the entrepreneurs,but this two sets of dummy variables:δ，andδa.Our variable increase may also be due to their ability to provide in- of interest is the log of the migrant stock (Ids)-the formation about opportunities to the investment com- population of migrants-from the destination country munity.Following Rauch and Trindade (2002),we also residing in the source country. hypothesize that migrant networks will be more im- The vector (Xsd)includes variables other than mi- portant for investment in heterogeneous(as compared grant stock that potentially influence cross-border in- with homogeneous)assets.This is because informa- vestment.Drawing on gravity-type models of FDI and tion costs-both asymmetries and transactions costs- portfolio investment,16 we control for the dyad's mar- associated with heterogeneous assets are substantially ket size,the bilateral distance between countries,the larger than those associated with homogeneous assets. existence of a common border,and a shared common When it comes to cross-border investment,we argue language.17 Market size is measured as the log of the that FDI is far more heterogeneous than portfolio in- product of the two country's gross domestic products vestment.Whereas portfolio investors choose debt or (GDPs).Given that distance is an imperfect measure equity stakes that are offered by an issuing agency of information costs,following Portes and Rey(2005). on an organized market,foreign direct investors can we include the log of bilateral telephone traffic take innumerable different ownership stakes across a Trade-in financial assets may also be influenced by countless number of commodity classes.We hypothe- shared policy variables.As in Lane and Milesi-Ferretti size,therefore,that migrant networks will have a larger (2004),we include variables capturing whether the two substantive effect on FDI than on portfolio investment. countries share a common exchange rate peg,have a It may be the case that migrant networks do not dual taxation treaty,or are members of a preferential matter at all and that they just proxy for greater cul- trade agreement.Given the importance of portfolio tural or informational similarities among countries- diversification in ICAPMs,we proxy for common eco- the sorts of variables identified in empirical studies of nomic shocks by including the (lagged)correlation of ICAPMs.To decrease the risk of incorrect inferences, the two country's growth rates. we go to great lengths to control for a variety of ways As discussed previously,it is difficult to differentiate that countries are connected with one another. between investment due to cultural affinity,market fa- miliarity,or information asymmetry.Our approach is EMPIRICS 14 See Alfaro.Kelemli-Ozcan,and Volosovych (2006).Buthe and Model and Variables Milner (2008).and Jensen (2003)for a discussion of destination country factors that are correlated with capital inflows. In addition to migrant networks,there are scores of is Source and destination countries are listed in Appendix A while factors that conceivably effect investment between two variable sources and definitions are contained in Appendix B. 16 Studies of FDI employing a gravity model include Wei (2000)and countries.In the source country,capital outflows can Loungani,Mody,and Razin (2002).Portes,Rey,and Oh(2001),Lane be influenced by high tax rates,the lack of protection and Milesi-Ferretti(2004),Portes and Rey (2005),Lane (2005),and for private property,high savings rates,etc.A similar Eichengreen and Luengnaruemitchai(2006)use a gravity model to list can be made for recipient countries:countries that examine bilateral investment in equities and bonds. 17 Some gravity models also include a variable indicating whether the two countries were ever in a colonial relationship.When included in our models,this variable was never statistically significant.We omit 13 Quoted from www.martinfrost.ws/htmlfiles/gazette/scot_diaspora. it from our specification because we also include a measure of shared html. legal origin as it is highly correlated with colonial status. 588

Familiarity Breeds Investment August 2010 facilitating investment and in further integrating their countries into the global economy. Gamlen (2008) notes that even developed states such as New Zealand, Ireland, and Israel have well-developed diaspora en￾gagement strategies designed to encourage active eco￾nomic behavior on the part of their external citizens. In a response to a question from a member of their parliament about how Scotland intends to engage their diaspora residing in Canada, Scottish Executive Min￾ister for Parliament declared, “The Scottish Executive intends to engage with and mobilize the Scottish dias￾pora to further Scotland’s interests for the long-term benefit of our economy and society. We aim to encour￾age the diaspora’s active participation and engagement in promoting Scotland as a great country to visit, live, learn, work, do business and invest.”13 Because they increase familiarity and decrease in￾formation asymmetries, we hypothesize that larger mi￾grant networks, all other things being equal, will exert a positive influence on cross-border investment. The effect of migrant networks should increase when the migrants themselves are the entrepreneurs, but this increase may also be due to their ability to provide in￾formation about opportunities to the investment com￾munity. Following Rauch and Trindade (2002), we also hypothesize that migrant networks will be more im￾portant for investment in heterogeneous (as compared with homogeneous) assets. This is because informa￾tion costs—both asymmetries and transactions costs— associated with heterogeneous assets are substantially larger than those associated with homogeneous assets. When it comes to cross-border investment, we argue that FDI is far more heterogeneous than portfolio in￾vestment. Whereas portfolio investors choose debt or equity stakes that are offered by an issuing agency on an organized market, foreign direct investors can take innumerable different ownership stakes across a countless number of commodity classes. We hypothe￾size, therefore, that migrant networks will have a larger substantive effect on FDI than on portfolio investment. It may be the case that migrant networks do not matter at all and that they just proxy for greater cul￾tural or informational similarities among countries— the sorts of variables identified in empirical studies of ICAPMs. To decrease the risk of incorrect inferences, we go to great lengths to control for a variety of ways that countries are connected with one another. EMPIRICS Model and Variables In addition to migrant networks, there are scores of factors that conceivably effect investment between two countries. In the source country, capital outflows can be influenced by high tax rates, the lack of protection for private property, high savings rates, etc. A similar list can be made for recipient countries: countries that 13 Quoted from www.martinfrost.ws/htmlfiles/gazette/scot_diaspora. html. receive capital tend to have better institutions, more stable economic policies, and better governance.14 All empirical models are vulnerable to problems of omitted variable bias. We choose a very conservative strategy and estimate a cross-sectional model of bilat￾eral investment that contains fixed effects for both the source and destination countries. Although this estima￾tion strategy sacrifices some richness in that we cannot explore variation in host and homeland domestic insti￾tutions and policies, it allows us to concentrate on bi￾lateral characteristics that influence dyadic investment. Concretely, we write our model as log(y)sd = δs + δd + φ log(Ids) + βXsd + εsd, (1) where ysd is the level of either portfolio or FDI from a source country (s) into a destination country (d).15 Because we are working with a cross-section of country dyads, we control for unmeasured source and destina￾tion country characteristics through the inclusion of two sets of dummy variables: δs and δd. Our variable of interest is the log of the migrant stock (Ids)—the population of migrants—from the destination country residing in the source country. The vector (Xsd) includes variables other than mi￾grant stock that potentially influence cross-border in￾vestment. Drawing on gravity-type models of FDI and portfolio investment,16 we control for the dyad’s mar￾ket size, the bilateral distance between countries, the existence of a common border, and a shared common language.17 Market size is measured as the log of the product of the two country’s gross domestic products (GDPs). Given that distance is an imperfect measure of information costs, following Portes and Rey (2005), we include the log of bilateral telephone traffic. Trade-in financial assets may also be influenced by shared policy variables. As in Lane and Milesi-Ferretti (2004), we include variables capturing whether the two countries share a common exchange rate peg, have a dual taxation treaty, or are members of a preferential trade agreement. Given the importance of portfolio diversification in ICAPMs, we proxy for common eco￾nomic shocks by including the (lagged) correlation of the two country’s growth rates. As discussed previously, it is difficult to differentiate between investment due to cultural affinity, market fa￾miliarity, or information asymmetry. Our approach is 14 See Alfaro, Kelemli-Ozcan, and Volosovych (2006), Buthe and ¨ Milner (2008), and Jensen (2003) for a discussion of destination country factors that are correlated with capital inflows. 15 Source and destination countries are listed in Appendix A while variable sources and definitions are contained in Appendix B. 16 Studies of FDI employing a gravity model include Wei (2000) and Loungani, Mody, and Razin (2002). Portes, Rey, and Oh (2001), Lane and Milesi-Ferretti (2004), Portes and Rey (2005), Lane (2005), and Eichengreen and Luengnaruemitchai (2006) use a gravity model to examine bilateral investment in equities and bonds. 17 Some gravity models also include a variable indicating whether the two countries were ever in a colonial relationship. When included in our models, this variable was never statistically significant. We omit it from our specification because we also include a measure of shared legal origin as it is highly correlated with colonial status. 588

American Political Science Review Vol.104,No.3 to control-as completely as possible-for other mea- Sample and Methods sures of culture and familiarity.To that end,we include a number of measures that capture cultural similari- To examine the link between migrant networks and ties between the source and destination countries.The bilateral portfolio investment,we use data from the first-a measure of common legal origin-is more in- International Monetary Fund's (IMF's)Coordinated stitutional than cultural,but it captures the ability of Portfolio Investment Survey(CPIS).The CPIS collects investors from country s to invest in country d with information on the stock of cross-border investments minimal transaction costs because they will already in equities and in short-and long-term bonds broken be familiar with the rules and regulations.We expect down by issuer's country of residence.19 Due to data that country pairs with common legal origins will ex- constraints.we are able to use data on the investment perience higher levels of cross-border investment than portfolio of 56 source (reporting)countries and 154 those pairs with dissimilar legal origins. destination countries.20 The list of source and destina- Our second control for cultural similarity is a mea- tion countries is contained in Appendix C. sure of cultural proximity that is created through the Our data on FDI come from the OECD's Interna- creation of a dummy variable measuring whether the tional Direct Investment.This source is limited in that two countries have a common dominant religion.Com it only provides data for outflows from OECD coun- mon religion proxies for similar beliefs,values,and ex- tries.Therefore,when we look at bilateral FDI,our pectations regarding the existence of social norms and sample is restricted to one of 28 source countries and the internally imposed constraints that are important 158 destination countries. for a business partnership across borders. Our key independent variable-that of migrant The third cultural control is grounded in cultural networks-measures the stock (or total number)of mi- economics and operationalized as a measure of ge- grants from country d residing in country s.These data netic distance between countries.Based on the work come from a World Bank project on South-South mi- of Cavalli-Sforza,Menozzi,and Piazza (1994),schol- gration and remittances.They are based on data from ars have developed measures of genetic distances be- national statistical bureaus (censuses and population tween indigenous populations based on genetic or registers)and secondary sources(the OECD,the Inter- DNA polymorphism.18 This measure of genetic dis- national Labour Organization,and the UN).A 162 x tance has been used to proxy for culture in studies of 162 matrix of the migrant stock in country s from coun- international trade and FDI (Giuliano,Spilimbergo. try d classified according the migrant's country of birth and Tonon 2006:Guiso,Sapienza,and Zingales 2005). is constructed from these national sources (Ratha and economic development(Spolaore and Wacziarg 2008). Shaw 2007).Although some of the underlying data and state formation in Europe (Desmet et al.2007). are from the late 1990s,the majority correspond to Desmet et al.provide evidence that European coun- migrant stock for 2000 or 2001.Consequently,we are tries that are genetically alike have populations that restricted to working with cross-sectional and not time- provide similar answers to World Values Survey ques- series data. tions about cultural,religious,and moral issues. We estimate Equation (1)using ordinary least Finally,we include a more direct measure of cul- squares(OLS)and control for source-and destination tural similarity.Studies in international business find country-specific variables through the use of a double that greater cultural distance between countries is as- set of fixed effects.Inferences based on OLS standard sociated with larger transactions costs,higher uncer- errors may,however.be underestimated.This bias may tainty about business practices,and overall greater un be attributable to two related causes.First,investment ease regarding the prospects for doing business (e.g. by source countries may cluster geographically;conse- Habib and Zurawicki 2002;Kogut and Singh 1988; quently,we may need standard errors that are clustered Siegel,Licht,and Schwartz 2008).Some recent stud- by s.Second,some destination countries,for a multi- ies of international trade find that culturally similar tude of reasons,receive more investment than other countries engage in larger levels of transactions (e.g. Guiso,Sapienza,and Zingales 2005;Siegel,Licht,and Schwartz 2008;White and Tadesse 2008).Following 19 Lane and Milesi-Ferretti (2004)and Eichengreen and Lueng. this lead.we use a measure of cultural difference or naruemitchai(2006)point out some advantages and disadvantages of distance based on questions from the World Values the CPIS data.In designing the survey,the IMF attempted to ensure Survey.Unfortunately,these surveys are only given in comparability across countries;to that end,the surveys are structured to prevent double counting.With that said,the CPIS does not report 95 countries,so their use limits the size of our sample; the domestic holdings of investors,which makes testing theories of consequently,we include these measures as a robust- portfolio allocation and home bias difficult with these data,and it ness check is possible that there is some underreporting.Most significantly,for our purposes,the CPIS does not have data on the foreign holdings of a few large origin countries,including China and Saudi Arabia (although it does have these countries as destinations). 2 As in Rose and Spiegel (2008),we use the average of portfolio investment for 2002,2003,and 2004 because response rates for these 18 The details involved in the derivation of these measures in and of years differ broadly by country.Pooling these years allows us to themselves constitute a paper.The interested reader is directed to almost double the sample size.The correlation between portfolio Spolaore and Wacziarg(2008)for a discussion and application.We investment for 2002 and the average from 2002 to 2004 is 0.91.For are grateful to Spolaore and Wacziarg for generously sharing their the purpose of comparability,we construct the dependent variable data. for FDI in a similar manner. 589

American Political Science Review Vol. 104, No. 3 to control—as completely as possible—for other mea￾sures of culture and familiarity. To that end, we include a number of measures that capture cultural similari￾ties between the source and destination countries. The first—a measure of common legal origin—is more in￾stitutional than cultural, but it captures the ability of investors from country s to invest in country d with minimal transaction costs because they will already be familiar with the rules and regulations. We expect that country pairs with common legal origins will ex￾perience higher levels of cross-border investment than those pairs with dissimilar legal origins. Our second control for cultural similarity is a mea￾sure of cultural proximity that is created through the creation of a dummy variable measuring whether the two countries have a common dominant religion. Com￾mon religion proxies for similar beliefs, values, and ex￾pectations regarding the existence of social norms and the internally imposed constraints that are important for a business partnership across borders. The third cultural control is grounded in cultural economics and operationalized as a measure of ge￾netic distance between countries. Based on the work of Cavalli-Sforza, Menozzi, and Piazza (1994), schol￾ars have developed measures of genetic distances be￾tween indigenous populations based on genetic or DNA polymorphism.18 This measure of genetic dis￾tance has been used to proxy for culture in studies of international trade and FDI (Giuliano, Spilimbergo, and Tonon 2006; Guiso, Sapienza, and Zingales 2005), economic development (Spolaore and Wacziarg 2008), and state formation in Europe (Desmet et al. 2007). Desmet et al. provide evidence that European coun￾tries that are genetically alike have populations that provide similar answers to World Values Survey ques￾tions about cultural, religious, and moral issues. Finally, we include a more direct measure of cul￾tural similarity. Studies in international business find that greater cultural distance between countries is as￾sociated with larger transactions costs, higher uncer￾tainty about business practices, and overall greater un￾ease regarding the prospects for doing business (e.g., Habib and Zurawicki 2002; Kogut and Singh 1988; Siegel, Licht, and Schwartz 2008). Some recent stud￾ies of international trade find that culturally similar countries engage in larger levels of transactions (e.g., Guiso, Sapienza, and Zingales 2005; Siegel, Licht, and Schwartz 2008; White and Tadesse 2008). Following this lead, we use a measure of cultural difference or distance based on questions from the World Values Survey. Unfortunately, these surveys are only given in 95 countries, so their use limits the size of our sample; consequently, we include these measures as a robust￾ness check. 18 The details involved in the derivation of these measures in and of themselves constitute a paper. The interested reader is directed to Spolaore and Wacziarg (2008) for a discussion and application. We are grateful to Spolaore and Wacziarg for generously sharing their data. Sample and Methods To examine the link between migrant networks and bilateral portfolio investment, we use data from the International Monetary Fund’s (IMF’s) Coordinated Portfolio Investment Survey (CPIS). The CPIS collects information on the stock of cross-border investments in equities and in short- and long-term bonds broken down by issuer’s country of residence.19 Due to data constraints, we are able to use data on the investment portfolio of 56 source (reporting) countries and 154 destination countries.20 The list of source and destina￾tion countries is contained in Appendix C. Our data on FDI come from the OECD’s Interna￾tional Direct Investment. This source is limited in that it only provides data for outflows from OECD coun￾tries. Therefore, when we look at bilateral FDI, our sample is restricted to one of 28 source countries and 158 destination countries. Our key independent variable—that of migrant networks—measures the stock (or total number) of mi￾grants from country d residing in country s. These data come from a World Bank project on South–South mi￾gration and remittances. They are based on data from national statistical bureaus (censuses and population registers) and secondary sources (the OECD, the Inter￾national Labour Organization, and the UN). A 162 × 162 matrix of the migrant stock in country s from coun￾try d classified according the migrant’s country of birth is constructed from these national sources (Ratha and Shaw 2007). Although some of the underlying data are from the late 1990s, the majority correspond to migrant stock for 2000 or 2001. Consequently, we are restricted to working with cross-sectional and not time￾series data. We estimate Equation (1) using ordinary least squares (OLS) and control for source- and destination country–specific variables through the use of a double set of fixed effects. Inferences based on OLS standard errors may, however, be underestimated. This bias may be attributable to two related causes. First, investment by source countries may cluster geographically; conse￾quently, we may need standard errors that are clustered by s. Second, some destination countries, for a multi￾tude of reasons, receive more investment than other 19 Lane and Milesi-Ferretti (2004) and Eichengreen and Lueng￾naruemitchai (2006) point out some advantages and disadvantages of the CPIS data. In designing the survey, the IMF attempted to ensure comparability across countries; to that end, the surveys are structured to prevent double counting. With that said, the CPIS does not report the domestic holdings of investors, which makes testing theories of portfolio allocation and home bias difficult with these data, and it is possible that there is some underreporting. Most significantly, for our purposes, the CPIS does not have data on the foreign holdings of a few large origin countries, including China and Saudi Arabia (although it does have these countries as destinations). 20 As in Rose and Spiegel (2008), we use the average of portfolio investment for 2002, 2003, and 2004 because response rates for these years differ broadly by country. Pooling these years allows us to almost double the sample size. The correlation between portfolio investment for 2002 and the average from 2002 to 2004 is 0.91. For the purpose of comparability, we construct the dependent variable for FDI in a similar manner. 589

Familiarity Breeds Investment August 2010 countries;a phenomena that would call for clustering portfolio investment is positively influenced by reli- on d. gious similarity,but not by a common legal heritage We deal with this potential bias by estimating stan- or by genetic distance.Both latter variables are statis- dard errors that are robust to multiway clustering as tically insignificant.Column 4 includes a more direct developed by Cameron.Gelbach.and Miller (2006). indicator of cultural similarity by including the World Their approach allows for arbitrary correlations be- Values Survey-based measure of cultural difference tween errors that belong to "the same group (along As expected,increasing cultural difference decreases either dimension)"(p.7).As they point out,this esti- cross-border portfolio investment.Adding these mea- mator is applicable in situations when the errors exhibit sures of cultural affinity or institutional familiarity do spatial correlation.Consequently,we report standard not,however,significantly affect the parameter esti- errors that are clustered by both source and desti- mate for migrant stock. nation countries.21 It should be noted that Cameron. It is also possible that patterns of bilateral investment Gelbach,and Miller mention that multiway clustering reflect other economic relationships between countries. increases-by an order of magnitude-the size of stan- Rauch and Trindade (2002)were the first to report a dard errors.In the results reported here,the standard positive relationship between diaspora networks and errors are between 60%and 100%larger than tradi- bilateral trade.If investment follows trade and not mi- tional robust standard errors.Hence,our results are gration,then inclusion of this variable should render very conservative. migrant stock statistically insignificant-or at least de- crease its substantive impact.Consequently,in column EMPIRICAL FINDINGS 5,we include a measure of bilateral trade.Trade has a negative effect on bilateral investment,indicating that these flows are substitutes rather than complements Central Results and its inclusion does not decrease the statistical or Table 1 reports models of dyadic portfolio investment. substantive importance of migrant networks. The specification in column 1 is our benchmark model, Table 2 repeats this exercise,substituting FDI as the where we just control for the variables used in prior dependent variable.Note that due to data limitations, studies of portfolio investment.Consistent with a stan- the FDI models refer to a much smaller number of dard gravity model,portfolio investment is a positive source countries.For the sake of space,we summarize function of country size (as measured by the product rather than walk through the findings from Table 2 of GDPs)and a negative function of distance.Surpris- We find that the gravity specification is a reasonable ingly,common language and common border are sta- benchmark because economic size and distance are tistically insignificant,as is the proxy for diversification statistically significant and consistently signed.The log (correlation of growth rates).Shared policies-a com- of migrant stock has a positive and statistically signif- mon exchange rate peg,a shared dual taxation treaty, icant effect on bilateral FDI that does not go away and membership in a preferential trade agreement- once we use other variables to measure cultural and have a positive and statistically significant effect on institutional familiarity. portfolio investment.We fail to find evidence that bilat- eral telephone traffic-a measure of information costs in previous studies(Portes and Rey 2005)-influences Migrant Networks and cross-border portfolio investment. Heterogeneous Investments In column 2.we add our measure of diaspora networks-the size of the migrant stock from the desti- The findings thus far support the argument that migrant nation residing in the source country.Consistent with networks serve as a conduit for capital flows,and they our hypotheses,we find that migrant networks have a point to the importance of migrant networks in the positive and statistically significant effect on portfolio provision of information.In this section,we test the in- investment.Because both the portfolio investment and formational hypothesis more directly.Following Rauch migrant stock have been transformed into logs,we can and Trindade(2002),we argue that the informational interpret the coefficient as an elasticity.This means role of migrant networks should be more important that increasing the migrant stock from a destination for trade in heterogeneous commodities,where private in a source country by 1%results in 0.2%increase information has greater value.We view FDI opportuni- in portfolio investment.Evaluated at their means,this ties as more heterogeneous than portfolio investment translates to a contribution of $450 per migrant to his opportunities.Not only are there an infinite number or her home country. of FDI opportunities-ranging from joint ownership to The migrant stock,of course,could simply be cap- greenfield investments-they also differ in that their turing cultural affinity or institutional familiarity.In risk of expropriation is greater.Portfolio investment, column 3,we include additional variables to control in contrast,can only be made in assets that are publicly for this possibility.These results are surprising because issued by either governmental or corporate interests entities that provide relatively more information to markets.Because portfolio investment is more liquid 21 We use Cameron,Gelbach,and Miller's (2006)cgmreg ado file. it can more easily be moved from market to market version 3.0,downloaded on August 2,2009,from http://gelbach. and from asset to asset,something that requires rela- eller.arizona.edu/~gelbach/ado/cgmreg.ado. tively less information than FDI.We therefore expect 590

Familiarity Breeds Investment August 2010 countries; a phenomena that would call for clustering on d. We deal with this potential bias by estimating stan￾dard errors that are robust to multiway clustering as developed by Cameron, Gelbach, and Miller (2006). Their approach allows for arbitrary correlations be￾tween errors that belong to “the same group (along either dimension)” (p. 7). As they point out, this esti￾mator is applicable in situations when the errors exhibit spatial correlation. Consequently, we report standard errors that are clustered by both source and desti￾nation countries.21 It should be noted that Cameron, Gelbach, and Miller mention that multiway clustering increases—by an order of magnitude—the size of stan￾dard errors. In the results reported here, the standard errors are between 60% and 100% larger than tradi￾tional robust standard errors. Hence, our results are very conservative. EMPIRICAL FINDINGS Central Results Table 1 reports models of dyadic portfolio investment. The specification in column 1 is our benchmark model, where we just control for the variables used in prior studies of portfolio investment. Consistent with a stan￾dard gravity model, portfolio investment is a positive function of country size (as measured by the product of GDPs) and a negative function of distance. Surpris￾ingly, common language and common border are sta￾tistically insignificant, as is the proxy for diversification (correlation of growth rates). Shared policies—a com￾mon exchange rate peg, a shared dual taxation treaty, and membership in a preferential trade agreement— have a positive and statistically significant effect on portfolio investment.We fail to find evidence that bilat￾eral telephone traffic—a measure of information costs in previous studies (Portes and Rey 2005)—influences cross-border portfolio investment. In column 2, we add our measure of diaspora networks—the size of the migrant stock from the desti￾nation residing in the source country. Consistent with our hypotheses, we find that migrant networks have a positive and statistically significant effect on portfolio investment. Because both the portfolio investment and migrant stock have been transformed into logs, we can interpret the coefficient as an elasticity. This means that increasing the migrant stock from a destination in a source country by 1% results in 0.2% increase in portfolio investment. Evaluated at their means, this translates to a contribution of $450 per migrant to his or her home country. The migrant stock, of course, could simply be cap￾turing cultural affinity or institutional familiarity. In column 3, we include additional variables to control for this possibility. These results are surprising because 21 We use Cameron, Gelbach, and Miller’s (2006) cgmreg.ado file, version 3.0, downloaded on August 2, 2009, from http://gelbach. eller.arizona.edu/∼gelbach/ado/cgmreg.ado. portfolio investment is positively influenced by reli￾gious similarity, but not by a common legal heritage or by genetic distance. Both latter variables are statis￾tically insignificant. Column 4 includes a more direct indicator of cultural similarity by including the World Values Survey–based measure of cultural difference. As expected, increasing cultural difference decreases cross-border portfolio investment. Adding these mea￾sures of cultural affinity or institutional familiarity do not, however, significantly affect the parameter esti￾mate for migrant stock. It is also possible that patterns of bilateral investment reflect other economic relationships between countries. Rauch and Trindade (2002) were the first to report a positive relationship between diaspora networks and bilateral trade. If investment follows trade and not mi￾gration, then inclusion of this variable should render migrant stock statistically insignificant—or at least de￾crease its substantive impact. Consequently, in column 5, we include a measure of bilateral trade. Trade has a negative effect on bilateral investment, indicating that these flows are substitutes rather than complements, and its inclusion does not decrease the statistical or substantive importance of migrant networks. Table 2 repeats this exercise, substituting FDI as the dependent variable. Note that due to data limitations, the FDI models refer to a much smaller number of source countries. For the sake of space, we summarize rather than walk through the findings from Table 2. We find that the gravity specification is a reasonable benchmark because economic size and distance are statistically significant and consistently signed. The log of migrant stock has a positive and statistically signif￾icant effect on bilateral FDI that does not go away once we use other variables to measure cultural and institutional familiarity. Migrant Networks and Heterogeneous Investments The findings thus far support the argument that migrant networks serve as a conduit for capital flows, and they point to the importance of migrant networks in the provision of information. In this section, we test the in￾formational hypothesis more directly. Following Rauch and Trindade (2002), we argue that the informational role of migrant networks should be more important for trade in heterogeneous commodities, where private information has greater value. We view FDI opportuni￾ties as more heterogeneous than portfolio investment opportunities. Not only are there an infinite number of FDI opportunities—ranging from joint ownership to greenfield investments—they also differ in that their risk of expropriation is greater. Portfolio investment, in contrast, can only be made in assets that are publicly issued by either governmental or corporate interests, entities that provide relatively more information to markets. Because portfolio investment is more liquid, it can more easily be moved from market to market and from asset to asset, something that requires rela￾tively less information than FDI. We therefore expect 590

American Political Science Review Vol.104,No.3 TABLE 1.Determinants of Cross-border Portfolio Investment (1) (2) (3) (4) (5) Log(migrant stock from din s) 0.204* 0.189* 0.252* 0.209* (0.0584) (0.0608) (0.0660) (0.0596) Log(product of GDPs) 0.181* 0.178* 0.172* 0.0902* 0.157* (0.0511) (0.0481) (0.0501) (0.0449) (0.0476) Log(e) -1.421* -1.209* -1.191* -0.672* -1.337* (0.210) (0.225) (0.233) (0.282) (0.227) Common border -0.365 -0.832* -0.906* -0.802* -0.922* (0.407 (0.438) (0.438) (0.483) (0.427 Official common language 0.359 0.0387 -0.0903 0.0526 0.0265 (0.267 (0.275) (0.288) (0.327 (0.268) Correlation of growth rates 0.0926 0.0777 0.0564 -0.170 0.0458 (0.204) (0.199) (0.197) (0.274) (0.193) Common exchange rate peg 1.375 1.191* 1.147* 1.131* 1.169* (0.364) (0.347) (0.341) (0.475) (0.343) Dual taxation treaty 1.334* 1.264* 1.253* 0.795* 1.228* (0.347) (0.343) (0.341) (0.302) (0.338) Preferential trade agreement 0.806* 0.807* 0.810* 0.838* 0.810*4 (0.456) (0.423) (0.418) (0.404) (0.410) Log(bilateral telephone volume) 0.109 0.0935 0.0875 0.110 0.0822 (0.0736) (0.0702) (0.0712) (0.123) (0.0722) Common legal heritage 0.212 0.180 0.218 (0.216) (0.229) (0.211) Common dominant religion 0.550* 0.784* 0.569* (0.277) (0.272) (0.276) Genetic distance -0.000110 0.0000764 (0.000484) 0.0000707 (0.000510) (0.000493) Cultural difference -0.560* (0.196) Log(bilateral trade) -0.137* (0.0551) Constant -133.2* -134.6* -129.9* -61.02 -111.9* (44.76) (41.99) (43.82) (39.70) (41.48) Observations 4.980 4,980 4.980 2,971 4,980 Adjusted R2 0.783 0.786 0.787 0.800 0.788 GDP,gross domestic product. Robust standard errors in parentheses. Dependent variable:log(portfolio investment from origin to destination). Robust standard errors clustered by both origin and destination country. All models include both origin and destination dummy variables. p<.10:*p<.05. that migrant networks should be substantively more Table 3 contains the results of this analysis.It is im- important for FDI than for portfolio investment. portant to point out that we require complete obser- We examine this hypothesis in Table 3,where we vations for both portfolio and FDI.The overlapping estimate a seemingly unrelated regression of the de- sample results in investment from 24 source countries terminants of both portfolio investment and bilateral (all OECD)into 152 destination countries.We find that trade in commodities.This approach allows for testing larger migrant networks increase both portfolio invest- cross-equation restrictions-the null being that the ef- ment and FDI,but that the effect on FDI is substan- fect of migrant stock on portfolio investment is equal tively larger,and statistically different at the 95%con- to its effect on FDI.Rather than reporting standard fidence level,than the effect on portfolio investment. errors,we provide 95%confidence intervals.22 This is consistent with our expectations:as portfolio in- vestment represents a more homogenous opportunity 22 Hypothesis testing using seemingly unrelated regression assumes set,private information provided by migrant networks that the errors from both equations are asymptotically normal.In the becomes increasingly valuable for investors evaluating case of trade,the residuals are not due to a large number of zeros. more heterogeneous options.23 We therefore calculate standard errors and associated confidence in- tervals of Huber/White standard errors.Unfortunately,the approach of Cameron,Gelbach,and Miller(2006)is not directly applicable to 23 We note an important caveat to this result:if we divide the sam. seemingly unrelated regression. ple,as we do later,so that we are examining investments from rich 591

American Political Science Review Vol. 104, No. 3 TABLE 1. Determinants of Cross-border Portfolio Investment (1) (2) (3) (4) (5) Log(migrant stock from d in s) 0.204∗∗ 0.189∗∗ 0.252∗∗ 0.209∗∗ (0.0584) (0.0608) (0.0660) (0.0596) Log(product of GDPs) 0.181∗∗ 0.178∗∗ 0.172∗∗ 0.0902∗∗ 0.157∗∗ (0.0511) (0.0481) (0.0501) (0.0449) (0.0476) Log(e) −1.421∗∗ −1.209∗∗ −1.191∗∗ −0.672∗∗ −1.337∗∗ (0.210) (0.225) (0.233) (0.282) (0.227) Common border −0.365 −0.832∗ −0.906∗∗ −0.802∗ −0.922∗∗ (0.407) (0.438) (0.438) (0.483) (0.427) Official common language 0.359 0.0387 −0.0903 0.0526 0.0265 (0.267) (0.275) (0.288) (0.327) (0.268) Correlation of growth rates 0.0926 0.0777 0.0564 −0.170 0.0458 (0.204) (0.199) (0.197) (0.274) (0.193) Common exchange rate peg 1.375∗∗ 1.191∗∗ 1.147∗∗ 1.131∗∗ 1.169∗∗ (0.364) (0.347) (0.341) (0.475) (0.343) Dual taxation treaty 1.334∗∗ 1.264∗∗ 1.253∗∗ 0.795∗∗ 1.228∗∗ (0.347) (0.343) (0.341) (0.302) (0.338) Preferential trade agreement 0.806∗ 0.807∗ 0.810∗ 0.838∗∗ 0.810∗∗ (0.456) (0.423) (0.418) (0.404) (0.410) Log(bilateral telephone volume) 0.109 0.0935 0.0875 0.110 0.0822 (0.0736) (0.0702) (0.0712) (0.123) (0.0722) Common legal heritage 0.212 0.180 0.218 (0.216) (0.229) (0.211) Common dominant religion 0.550∗∗ 0.784∗∗ 0.569∗∗ (0.277) (0.272) (0.276) Genetic distance — −0.000110 — 0.0000764 (0.000484) 0.0000707 (0.000510) (0.000493) Cultural difference −0.560∗∗ (0.196) Log(bilateral trade) −0.137∗∗ (0.0551) Constant −133.2∗∗ −134.6∗∗ −129.9∗∗ −61.02 −111.9∗∗ (44.76) (41.99) (43.82) (39.70) (41.48) Observations 4,980 4,980 4,980 2,971 4,980 Adjusted R2 0.783 0.786 0.787 0.800 0.788 GDP, gross domestic product. Robust standard errors in parentheses. Dependent variable: log(portfolio investment from origin to destination). Robust standard errors clustered by both origin and destination country. All models include both origin and destination dummy variables. ∗p < .10; ∗∗p < .05. that migrant networks should be substantively more important for FDI than for portfolio investment. We examine this hypothesis in Table 3, where we estimate a seemingly unrelated regression of the de￾terminants of both portfolio investment and bilateral trade in commodities. This approach allows for testing cross-equation restrictions—the null being that the ef￾fect of migrant stock on portfolio investment is equal to its effect on FDI. Rather than reporting standard errors, we provide 95% confidence intervals.22 22 Hypothesis testing using seemingly unrelated regression assumes that the errors from both equations are asymptotically normal. In the case of trade, the residuals are not due to a large number of zeros. We therefore calculate standard errors and associated confidence in￾tervals of Huber/White standard errors. Unfortunately, the approach of Cameron, Gelbach, and Miller (2006) is not directly applicable to seemingly unrelated regression. Table 3 contains the results of this analysis. It is im￾portant to point out that we require complete obser￾vations for both portfolio and FDI. The overlapping sample results in investment from 24 source countries (all OECD) into 152 destination countries. We find that larger migrant networks increase both portfolio invest￾ment and FDI, but that the effect on FDI is substan￾tively larger, and statistically different at the 95% con- fidence level, than the effect on portfolio investment. This is consistent with our expectations: as portfolio in￾vestment represents a more homogenous opportunity set, private information provided by migrant networks becomes increasingly valuable for investors evaluating more heterogeneous options.23 23 We note an important caveat to this result: if we divide the sam￾ple, as we do later, so that we are examining investments from rich 591

Familiarity Breeds Investment August 2010 TABLE 2.Determinants of Cross-border Foreign Direct Investment (1) (2) (3) (4) (5) Log(migrant stock from din s) 0.174* 0.163* 0.157* 0.151# (0.0638) (0.0637) (0.0738) (0.0639) Log(product of GDPs) 0.163* 0.156* 0.152* 0.149* 0.162 (0.0484) (0.0479) (0.0486) (0.0454) (0.0488) Log(distance) -1.358* -1.173* -1.115* -1.122* -1.026* (0.243) (0.240) (0.238) (0.273) (0.242) Common border 0.907 0.700 0.693 0.835 0.676 (0.645) (0.603) (0.594) (0.612) (0.595) Official common language 2.096* 1.774* 1.667* 1.193* 1.619# (0.624) (0.560) (0.568) (0.645) (0.566) Correlation of growth rates 0.177 0.172 0.162 0.258 0.155 (0.207) (0.205) (0.205) (0.273) (0.203) Common exchange rate peg 0.820 0.774 0.739 0.320 0.738 (0.628) (0.603) (0.599) (0.822) (0.598) Dual taxation treaty 0.714 0.625* 0.629* 0.489 0.615* (0.311) (0.315) (0.316) (0.338) (0.316) Preferential trade agreement 0.155 0.147 0.127 0.362 0.0685 (0.410) (0.410) (0.411) (0.482) (0.414) Log(bilateral telephone volume) -0.0394 -0.0494 -0.0519 -0.119 -0.0509 (0.103) (0.103) (0.102) (0.152) (0.102) Common legal heritage 0.237 0.503* 0.213 (0.249) (0.259) (0.251) Common dominant religion 0.204 0.362 0.202 (0.277) (0.313) (0.278) Genetic distance -0.000504 -0.000161 -0.000528 (0.000366) (0.000486) (0.000365) Cultural difference -0.000885 (0.199) Log(bilateral trade) 0.0954 (0.0585) Constant -125.6* -122.7* -119.5* -116.8* -131.9# (42.89) (43.00) (43.60) (41.04) (43.80) Observations 3.573 3.573 3,573 2,366 3,573 Adjusted R2 0.639 0.641 0.642 0.624 0.643 GDP gross domestic product. Robust standard errors in parentheses. Dependent variable:log(FDI from origin to destination). Robust standard errors clustered by both origin and destination country All models include both origin and destination dummy variables. p<.10:*p<.05. Educated Migrants Real Estate(FIRE)sectors.24 We also include the log of The results thus far support the hypothesis that migrant the total number of migrants so that the former variable networks encourage cross-border investment because is not capturing the effect of the entire migrant stock. they provide investors with information about invest- The results in column 1 of Table 4 report results for ment opportunities across particular destinations.We portfolio investment using the share of migrants with expect that this effect will be more pronounced when higher education,column 2 replaces the higher edu- the migrant themselves are involved in the investment cation variable with the share of migrants employed process.Unfortunately,we cannot directly measure in FIRE.and column 3 includes both measures.The migrant-based investment,so we proxy for it using the results indicate that higher education per se does not percentage of migrants from country d living in coun- have a statistically significant influence on portfolio in- try s that have higher (tertiary)education and.more vestment,whereas the opposite is true for employment directly,those employed in the Finance,Insurance,or 24 We use education data from the OECD's Immigration and Expa- triate Database.This database only has information on immigrants countries into poor countries,the statistical significance of this result into OECD countries;consequently,the sample size is greatly re. goes away.Whether that is due to a difference in sample or because duced,but the most recent release (accessed January 15.2009)con- of different investment opportunities across these sets of countries tains migrants broken down by occupation based on the standard is difficult to sort out given our use of cross-sectional data with both classification of occupations(ISCO-88 codes).We use those codes to source and destination dummy variables. construct the measure of employment in FIRE. 592

Familiarity Breeds Investment August 2010 TABLE 2. Determinants of Cross-border Foreign Direct Investment (1) (2) (3) (4) (5) Log(migrant stock from d in s) 0.174∗∗ 0.163∗∗ 0.157∗∗ 0.151∗∗ (0.0638) (0.0637) (0.0738) (0.0639) Log(product of GDPs) 0.163∗∗ 0.156∗∗ 0.152∗∗ 0.149∗∗ 0.162∗∗ (0.0484) (0.0479) (0.0486) (0.0454) (0.0488) Log(distance) −1.358∗∗ −1.173∗∗ −1.115∗∗ −1.122∗∗ −1.026∗∗ (0.243) (0.240) (0.238) (0.273) (0.242) Common border 0.907 0.700 0.693 0.835 0.676 (0.645) (0.603) (0.594) (0.612) (0.595) Official common language 2.096∗∗ 1.774∗∗ 1.667∗∗ 1.193∗ 1.619∗∗ (0.624) (0.560) (0.568) (0.645) (0.566) Correlation of growth rates 0.177 0.172 0.162 0.258 0.155 (0.207) (0.205) (0.205) (0.273) (0.203) Common exchange rate peg 0.820 0.774 0.739 0.320 0.738 (0.628) (0.603) (0.599) (0.822) (0.598) Dual taxation treaty 0.714∗∗ 0.625∗∗ 0.629∗∗ 0.489 0.615∗ (0.311) (0.315) (0.316) (0.338) (0.316) Preferential trade agreement 0.155 0.147 0.127 0.362 0.0685 (0.410) (0.410) (0.411) (0.482) (0.414) Log(bilateral telephone volume) −0.0394 −0.0494 −0.0519 −0.119 −0.0509 (0.103) (0.103) (0.102) (0.152) (0.102) Common legal heritage 0.237 0.503∗ 0.213 (0.249) (0.259) (0.251) Common dominant religion 0.204 0.362 0.202 (0.277) (0.313) (0.278) Genetic distance −0.000504 −0.000161 −0.000528 (0.000366) (0.000486) (0.000365) Cultural difference −0.000885 (0.199) Log(bilateral trade) 0.0954 (0.0585) Constant −125.6∗∗ −122.7∗∗ −119.5∗∗ −116.8∗∗ −131.9∗∗ (42.89) (43.00) (43.60) (41.04) (43.80) Observations 3,573 3,573 3,573 2,366 3,573 Adjusted R2 0.639 0.641 0.642 0.624 0.643 GDP, gross domestic product. Robust standard errors in parentheses. Dependent variable: log(FDI from origin to destination). Robust standard errors clustered by both origin and destination country. All models include both origin and destination dummy variables. ∗ p < .10; ∗∗ p < .05. Educated Migrants The results thus far support the hypothesis that migrant networks encourage cross-border investment because they provide investors with information about invest￾ment opportunities across particular destinations. We expect that this effect will be more pronounced when the migrant themselves are involved in the investment process. Unfortunately, we cannot directly measure migrant-based investment, so we proxy for it using the percentage of migrants from country d living in coun￾try s that have higher (tertiary) education and, more directly, those employed in the Finance, Insurance, or countries into poor countries, the statistical significance of this result goes away. Whether that is due to a difference in sample or because of different investment opportunities across these sets of countries is difficult to sort out given our use of cross-sectional data with both source and destination dummy variables. Real Estate (FIRE) sectors.24 We also include the log of the total number of migrants so that the former variable is not capturing the effect of the entire migrant stock. The results in column 1 of Table 4 report results for portfolio investment using the share of migrants with higher education, column 2 replaces the higher edu￾cation variable with the share of migrants employed in FIRE, and column 3 includes both measures. The results indicate that higher education per se does not have a statistically significant influence on portfolio in￾vestment, whereas the opposite is true for employment 24 We use education data from the OECD’s Immigration and Expa￾triate Database. This database only has information on immigrants into OECD countries; consequently, the sample size is greatly re￾duced, but the most recent release (accessed January 15, 2009) con￾tains migrants broken down by occupation based on the standard classification of occupations (ISCO-88 codes). We use those codes to construct the measure of employment in FIRE. 592

American Political Science Review Vol.104,No.3 TABLE 3.Comparing Portfolio and Foreign Direct Investment Portfolio Investment FDI Log(migrant stock from din s) 0.225* 0.319 [0.146.0.304] [0.227,0.412] Log(product of GDPs) 0.138* 0.180* [0.0904,0.185) [0.117,0.243] Log(distance) -0.708* -0.967* [-1.015,-0.401) [-1.357,-0.577 Common border -0.475 0.363 [-1.191,0.240] [-0.541.1.266] Official common language 0.0154 1.406* [-0.477,0.5081 [0.684,2.129] Correlation of growth rates -0.423* 0.197 [-0.761,-0.0851] [-0.154,0.548] Common exchange rate peg 0.929* 0.296 「0.482,1.377J [-0.419,1.010] Dual taxation treaty 0.820* 0.656* [0.377.1.262] [0.150,1.162] Preferential trade agreement 0.262 0.706* [-0.392,0.916] [0.0103,1.401] Common legal heritage 0.333* 0.351* [0.0255,0.641] [-0.0189,0.721] Common dominant religion 0.448* 0.337 [0.109.0.786] [-0.0891,0.7631 Genetic distance -0.000301 -0.0000469 [-0.00101,0.000406] [-0.000834,0.0007401 Log(bilateral telephone volume) 0.233* -0.0803 [0.0620,0.404] [-0.269,0.109] Constant -104.0 -148.1* 【-145.8,-62.24 [-204.0,-92.30j Observations 2,337 Adjusted R2 FDI,foreign direct investment;GDP,gross domestic product. Seemingly unrelated regression estimated via maximum likelihood estimate (MLE);confidence intervals based on robust covariance matrix. 95%robust confidence intervals in brackets. Column 1 dependent variable:log(portfolio investment from origin to destination). Column 2 dependent variable:log(FDI). All models include both origin and destination dummy variables p<.10;*p<.05. in the FIRE sector.In columns 4,5,and 6,we repeat across different subsamples.25 This is especially im- this exercise using FDI as the dependent variable and portant as pointed out by Blonigen and Wang(2005), reach the opposite conclusion because the tertiary ed- who experience problems when pooling FDI data ucation variable is statistically insignificant,whereas across developed and developing countries.26 Table 5 employment in the FIRE sector is not.Although not breaks up the sample and examines flows of portfo- statistically consistent across all models.these results lio investment and FDI from rich countries into rich do point to the importance of specific information when it comes to portfolio investment-something associated with training in the FIRE sectors.The findings also support our inclination regarding different information 25 In an earlier version of this article,the models were estimated without fixed effects,and the findings reported were robust to the requirements of heterogeneous versus homogenous inclusion of a large number of source and destination country con- investments:heterogeneous investment opportunities trol variables.Those results were also robust when estimated via across a large number of sectors-like those associated instrumental variables using the source country's immigration policy with FDI-do not necessarily privilege those with spe- as the instrument.Unfortunately,it is not possible to repeat that cific knowledge of banking or financial assets. here because immigration policy is country specific and would be completely collinear with the set of source country fixed effects. Other plausible instruments for immigration policy(e.g.,common frameworks for labor mobility within the structure of the General Robustness of the Central Results Agreement on Trade in Services)would not meet the excludability criteria. We check the robustness of our results by estimat- ing the effect of the migrant stock on investment 26 Iam grateful to an anonymousreviewer for bringing this reference to my attention. 593

American Political Science Review Vol. 104, No. 3 TABLE 3. Comparing Portfolio and Foreign Direct Investment Portfolio Investment FDI Log(migrant stock from d in s) 0.225∗∗ 0.319∗∗ [0.146, 0.304] [0.227, 0.412] Log(product of GDPs) 0.138∗∗ 0.180∗∗ [0.0904, 0.185] [0.117, 0.243] Log(distance) −0.708∗∗ −0.967∗∗ [−1.015, −0.401] [−1.357, −0.577] Common border −0.475 0.363 [−1.191, 0.240] [−0.541, 1.266] Official common language 0.0154 1.406∗∗ [−0.477, 0.508] [0.684, 2.129] Correlation of growth rates −0.423∗∗ 0.197 [−0.761, −0.0851] [−0.154, 0.548] Common exchange rate peg 0.929∗∗ 0.296 [0.482, 1.377] [−0.419, 1.010] Dual taxation treaty 0.820∗∗ 0.656∗∗ [0.377, 1.262] [0.150, 1.162] Preferential trade agreement 0.262 0.706∗∗ [−0.392, 0.916] [0.0103, 1.401] Common legal heritage 0.333∗∗ 0.351∗ [0.0255, 0.641] [−0.0189, 0.721] Common dominant religion 0.448∗∗ 0.337 [0.109, 0.786] [−0.0891, 0.763] Genetic distance −0.000301 −0.0000469 [−0.00101, 0.000406] [−0.000834, 0.000740] Log(bilateral telephone volume) 0.233∗∗ −0.0803 [0.0620, 0.404] [−0.269, 0.109] Constant −104.0∗∗ −148.1∗∗ [−145.8, −62.24] [−204.0, −92.30] Observations 2,337 Adjusted R2 FDI, foreign direct investment; GDP, gross domestic product. Seemingly unrelated regression estimated via maximum likelihood estimate (MLE); confidence intervals based on robust covariance matrix. 95% robust confidence intervals in brackets. Column 1 dependent variable: log(portfolio investment from origin to destination). Column 2 dependent variable: log(FDI). All models include both origin and destination dummy variables. ∗ p < .10; ∗∗ p < .05. in the FIRE sector. In columns 4, 5, and 6, we repeat this exercise using FDI as the dependent variable and reach the opposite conclusion because the tertiary ed￾ucation variable is statistically insignificant, whereas employment in the FIRE sector is not. Although not statistically consistent across all models, these results do point to the importance of specific information when it comes to portfolio investment—something associated with training in the FIRE sectors. The findings also support our inclination regarding different information requirements of heterogeneous versus homogenous investments: heterogeneous investment opportunities across a large number of sectors—like those associated with FDI—do not necessarily privilege those with spe￾cific knowledge of banking or financial assets. Robustness of the Central Results We check the robustness of our results by estimat￾ing the effect of the migrant stock on investment across different subsamples.25 This is especially im￾portant as pointed out by Blonigen and Wang (2005), who experience problems when pooling FDI data across developed and developing countries.26 Table 5 breaks up the sample and examines flows of portfo￾lio investment and FDI from rich countries into rich 25 In an earlier version of this article, the models were estimated without fixed effects, and the findings reported were robust to the inclusion of a large number of source and destination country con￾trol variables. Those results were also robust when estimated via instrumental variables using the source country’s immigration policy as the instrument. Unfortunately, it is not possible to repeat that here because immigration policy is country specific and would be completely collinear with the set of source country fixed effects. Other plausible instruments for immigration policy (e.g., common frameworks for labor mobility within the structure of the General Agreement on Trade in Services) would not meet the excludability criteria. 26 I am grateful to an anonymous reviewer for bringing this reference to my attention. 593