The impact of regional integration on insider and outsider FDI.
* We develop hypotheses for the implications of regional economic integration on foreign direct investment (FDI) from insider and outsider countries contingent on member nations' country-specific characteristics.
* We find that following regional integration (1) on average there is an increase in inward FDI, and (2) structural determinants such as the host country's market size, cultural and geographic distances, and institutional efficiency have a significantly different impact than when the host nation was outside the integrated area. Common economic area membership is associated with greater FDI flows, with the larger members gaining more; furthermore, insider FDI is less sensitive to labor cost, whereas the host country's institutional efficiency is an even more positive determinant of insider FDI.
Keywords: Regional economic integration * RIA * Foreign direct investment * FDI * OECD
The centrality of foreign direct investment (FDI) in the modern economy cannot be overemphasized. The United Nations Conference on Trade and Development (UNCTAD) estimates that by 2005 there were about 77,000 multinational corporations (MNCs) with approximately 770,000 foreign affiliates world-wide (UNCTAD 2006, p. 10). Global FDI inflows reached a peak of nearly $1.5 trillion in 2000, before dropping to $ 558 billion in 2003, and subsequently recovering to $ 916 billion in 2005, and the share of value added by affiliates of MNCs now account for roughly 10% of world GDP. As countries liberalize their policy regimes to attract inward FDI, they may consider regional economic integration to serve as an additional location-specific advantage (Ethier 1998); in fact, the UN observes that "international investment rules are increasingly adopted as an essential part of free trade agreements and other treaties on economic cooperation" (UNCTAD 2006, p. 27). In addition, in recent years, the international business (IB) literature has drawn renewed attention to the regional nature of much of MNC activity (see, for example, Rugman and Verbeke 2004), thus the investigation of the relationship between regional integration and FDI flows is particularly apt.
A regionally integrated area (RIA) may attract more inward FDI for various reasons such as access to a larger market, defensive investments by firms from non-member countries to obtain similar treatment as firms from within the RIA, the lessening of the liability of foreignness within the RIA, and gains in economic efficiency. However, individual RIA members may not all observe gains (and indeed, some may even see losses) in FDI inflows, depending on the interaction between the motivations of the firms making the FDI and the variation in the location-specific advantages among the RIA members (Dunning 1997; Eden 2002; Ethier 1998; Feils and Rahman 2008; Rugman and Verbeke 2007). At present, "[t]he role of increasing regional integration" represents "a rich avenue for further research into the globalization/regionalisation debate" (Dunning et al. 2007, p. 187) in the IB literature using macro-level data, as is done in the present study.
We first develop hypotheses that explain which countries are expected to benefit most in terms of inward FDI from regional economic integration by examining the manner in which an RIA member's country-specific characteristics may materially affect firms' FDI decisions following integration. We then test these hypotheses by examining the aggregate FDI inflows into 24 OECD countries between 1980 and 2003. Our paper contributes to the globalization/regionalization debate in two ways: first, we provide evidence on which country characteristics are most important to attract FDI following accession to an RIA; second, we provide evidence on the country characteristics that attract FDI from inside the integrated region.
Our study extends the scope of existing studies by studying multiple RIAs and by examining the interaction effects of country characteristics and RIA membership. Specifically, we analyze the impact of joining one of the three RIAs--namely, the European Union (EU), the North American Free Trade Area (NAFTA), and Australia-New Zealand Closer Economic Relations (ANZCERTA)--that best consolidates the characteristics of a 'home region' within the OECD dataset for the firms that originate within the fellow member countries of an RIA. (1) Our key findings are as follows: joining an RIA increases inward FDI; host country market size and cultural affinity diminish in importance after accession, while the new member's institutional efficiency becomes more important. Insofar as intra-RIA flows are concerned, there are greater FDI flows among the RIA members, however, countries with relatively larger, more attractive markets receive a greater portion of the intra-RIA flows; and intra-RIA flows are less sensitive to labor cost, but even more sensitive to the host nation's institutional efficiency.
Most empirical analyses of the impact of regional integration on FDI tend to focus only on specific RIAs such as the EU or NAFTA (see, for example, Altomonte 2007; Bevan and Estrin 2004; Dunning 1997; Eden and Li 2004; Feils and Rahman 2008) while our study looks at multiple RIAs and thus allows for testing some of the questions raised in the globalization/regionalization debate. Recently, some studies have begun to address the effects of regional integration on FDI from insider countries using multiple RIAs with conflicting results. Daude et al. (2003) analyze the stock of outward FDI from 20 OECD countries into 60 host countries using panel data over the 1982-1999 period; their key finding is that intra-RIA FDI stock is significantly greater. Balasubramanyam et al. (2002) examine 381 bilateral FDI flows in 1995 to see if intra-RIA flows are greater for EU and NAFTA members; they do not find a significant impact of RIA membership. On the other hand, Hejazi (2005) finds that intra-RIA FDI is higher than expected for the EU, but not for NAFTA. None of these studies address the issue of changes in inward FDI flows for individual countries after they join an RIA, but only the differences in FDI flows between insider and outsider countries. Thus, these recent studies examining multiple RIAs cannot address the question of which countries benefit (or suffer) more when they join an RIA. We fill a gap in the literature by examining both the impact of accession to an RIA as well as any heterogeneity in post-accession FDI flows from insider and outsider countries based on an individual country's location determinants.
Dunning's (1993) OLI (Ownership-Location-Internalization) framework of international production shows the role of location in the overall FDI decision of a firm. According to Dunning (1993), three factors need to be present for firms to engage in FDI: ownership advantages, internalization benefits and location advantages. The firm must possess ownership advantages, such as a recognized brand name that cannot be exploited via arm's length transactions (e.g., trade or licensing). The firm may then seek to extract value by internalizing the market for its firm-specific advantage (FSA) across borders--i.e., FDI. The interaction of the particular FDI motivations of an MNC and the location-specific advantages provided by a potential host country will thus prove determinant as to where it seeks to make the investment.
The two main motivations of the MNC are the quest for increased sales and the quest for more efficient means of sourcing and producing a product. Following Rugman and Verbeke (2004), we refer to FDI in search for increased sales as downstream FDI and to FDI in search of increased production/sourcing efficiency as upstream FDI. (2) Depending on the type of its locational advantages, a country will be able to attract relatively more upstream or downstream FDI. Downstream FDI naturally searches for countries where the product can be marketed successfully; thus, downstream FDI will go to the region with the largest market potential and which is amenable to the MNC's FSAs (Rugman and Verbeke 2004). Upstream FDI, on the other hand, is concerned with reducing costs and will go to low-cost, more efficient countries (Rugman and Verbeke 2004, Sethi et al. 2003), absent extenuating factors (e.g., an RIA pact).
FDI Inflows Following Accession to an RIA
The formation of an RIA generally leads to an increase in inward FDI into the entire region (e.g., Dunning 1997). Downstream FDI by outsider MNCs is expected to increase since firms may want to take advantage of the increased market size (Buckley et al. 2001) and/or to offset the tariff advantages of intra-RIA firms (Eden 2002). Downstream FDI by insider MNCs is expected to increase if the RIA reduces transaction costs that serve as barriers to MNCs being able to exploit their FSAs (Rugman and Verbeke 2005). However, within an RIA the location of the FDI will depend on the relative location-specific advantages of a given country and not all countries may be able to attract additional FDI. For example, the liberalization of internal tariff regimes will cause upstream FDI shifts to locations with relatively greater efficiency (Buckley et al. 2001; Dunning 1997; Eden 2002); depending on the economies of scale and scope of the plants being switched to regional production centers, the net effect on inward FDI in any given country may or may not be positive. More efficient members of the RIA are expected to gain, while less efficient members of the RIA are expected to lose. Finally, although joining an RIA may lead to a reduction in certain types of FDI, if formation of RIAs leads to improvement in the members' economic effectiveness, overall FDI inflows should rise (Wolf et al. 2008). Therefore, though there may be variations on the impact of RIA accession contingent on the characteristics of individual countries, we can expect that accession leads to an increase in FDI inflows.
Hypothesis 1: Following accession to an RIA, new members, on average, observe an increase in inward FDI.
We next consider some of the potential variations in FDI inflows across the new member countries contingent on their country-specific characteristics; prior empirical evidence documents that joining an RIA can create a structural break in the existing FDI policy regime (Buckley et al. 2007a; Feils and Rahman 2008). We develop a series of corollary hypotheses on post-accession inward FDI that are contingent on the host's country-specific characteristics.
The FDI literature has shown that the host country market size is an important positive determinant of inward FDI (see, for example, Globerman and Shapiro 2003). However, following accession, the individual member's market size may no longer be a limiting factor since the RIA pact increases the "size of the country" (Buckley et al. 2001, p. 252) and the entire RIA may be accessed from any member country. Therefore, market size should be less important as a determinant of FDI inflows.
Hypothesis 1.1: Following accession to an RIA, an individual RIA member's market size becomes less important as a determinant for inward FDI.
The 'New Regionalism' school (e.g., Ethier 1998) suggests that following regional integration, some factors influencing the FDI decision, such as overcoming trade barriers, are no longer relevant. As some motivations for FDI fall away, the remaining motivations become more important, such as geographic distance (Wolf et al. 2008).
Hypothesis 1.2: Following accession to an RIA, an individual RIA member's geographic distance from the home country becomes more important as a determinant for inward FDI.
Along similar lines, following accession one can expect both insider and outsider MNCs to attempt rationalizing production within the RIA as firms migrate to relatively more efficient RIA members: those with better institutional efficiency and greater labor productivity (Buckley et al. 2001; Dunning 1997; Eden 2002).
Hypothesis 1.3: Following accession to an RIA, an individual RIA member's institutional efficiency becomes more important as a determinant for inward FDI.
Hypothesis 1.4: Following accession to an RIA, an individual RIA member's labor cost efficiency becomes more important as a determinant for inward FDI.
One exception to the increasing importance of the remaining factors once tariffs and other barriers are removed may be cultural distance. In the IB literature, the Uppsala School stressed the internationalization process whereby MNCs initiate their activities in markets with a low "psychic distance" (Johanson and Vahlne 1977), defined as the "the sum of factors preventing or disturbing the flows of information between firm and markets" (Johanson and Wiedersheim-Paul 1975, p. 308). Regional integration may serve to reduce psychic distance as it "brings an awareness between firms in both countries and facilitates the exchange of information between the countries" (Brewer 2007, p. 51). One major component of psychic distance is cultural distance. In fact, these two terms are often used interchangeably, even though psychic distance is a broader concept (Brewer 2007). If accession to an RIA serves to increase familiarity and knowledge about RIA members--thereby reducing the psychic distance to the new member--cultural distance may no longer be as important a determinant for FDI (Wolf et al. 2008).
Hypothesis 1.5: Following accession to an RIA, an individual RIA member's cultural distance from the home country becomes less important as a determinant for inward FDI.
Regionalization and Intra-RIA FDI Flows
In a series of studies Rugman and co-authors have drawn attention to the overwhelmingly regional, and, in particular, the home regional bias in the loci of activity of the world's largest MNCs (see Rugman and Verbeke 2004 and 2005 for a review of this stream of literature). While the subsequent debate has centered on the issues of the definition of the regions and/or the appropriateness of using macro-level vs. firm-level data (Dunning et al. 2007; Osegowitsch and Sammartino 2008; Rugman and Verbeke 2007), regional economic integration affords a complementary avenue for examining this phenomenon.
Rugman and Verbeke (2007, p. 201) suggest that the "liability of intra-regional expansion appears to be much lower than the liability of inter-regional expansion" since firms are better able to operate in more familiar countries in their home region. Wolf et al. (2008) develop additional hypotheses in support of the home-region orientation of firms based on a number of different streams of research: new regionalism (e.g., Ethier 1998), agglomeration forces (e.g., Fujita and Mori 2005), knowledge spillovers (e.g., McCann and Simonen 2005), competitive advantage (Porter 1990), psychic distance (e.g., Johanson and Vahlne 1977), and population ecology (e.g., Hannan and Freeman 1977). If regional integration reduces the additional cost of doing business abroad within the region relative to outside of the region, FDI by insider firms should be higher.
Hypothesis 2: Following accession to an RIA, FDI inflows are higher from MNCs from fellow RIA member countries.
The impact of RIA on the FDI decisions of the insider firms also depends on their motivations for FDI. We develop the following corollary hypotheses related to intra-RIA FDI flow: To the extent that downstream FSAs are regionally bound (Rugman and Verbeke 2005) and integration leads to greater liberalization in the investment sector (e.g., national treatment) and a lower psychic distance, intra-RIA market-seeking FDI may increase.
Hypothesis 2.1: Following accession to an RIA, an individual RIA member's market size becomes more important as a determinant of inward FDI from fellow RIA member countries.
Intra-RIA upstream FDI is expected to reshuffle and move to more efficient member countries as MNCs rationalize their production within the RIA (Eden 2002) as other determinants of FDI flows, such as trade barriers, are no longer relevant (Ethier 1998).
Hypothesis 2.2: Following accession to an RIA, an individual RIA member's institutional efficiency becomes more important as a determinant of inward FDI from fellow RIA member countries.
Finally, the move towards more efficient member countries (Eden 2002; Ethier 1998) also applies to labor cost efficiency. This effect may, however, be reduced by intra-RIA MNCs diverting production from lower labor cost, but non-member countries to higher labor cost fellow RIA member countries to take advantage of the lowered trade barriers following integration (Rugman and Gestrin 1993).
Hypothesis 2.3: Following accession to an RIA, an individual RIA member's labor cost becomes more important as a determinant of inward FDI from fellow RIA member countries.
Data and Methodology
In order to determine the impact of regional integration on inward FDI, we examine the annual flow of inward FDI (3) into OECD member countries between 1980 and 2003 (4). Table 5 in the Appendix contains the list of the host and home countries; the FD1 inflows are reported by the OECD in the SourceOECD database.
We use pooled, cross-sectional time-series data to test the hypotheses. Given that some of the independent variables are time-invariant (namely, geographic distance and cultural distance), the fixed effects model is not applicable; therefore, we use the random effects model (Wooldridge 2002). The dependent variable used in the analyses is the natural log of one plus the annual inflow of FDI from home country j to host country k in year t measured in USD. We add one to the FDI inflows so as not to lose the observations with a value of zero due to the logarithmic transformation. Our basic empirical model for the first set of hypotheses is specified as:
In(1 + FDI [Inflow.sub.j,k,t]) = [alpha] + [[beta].sub.1] In[(GDP Home).sub.j,t] + [[beta].sub.2]In[(GDP Host).sub.k,t] +[[beta].sub.3]In[(Geographic Distance).sub.j,k] +[[beta].sub.4][RIA.sub.k,t] +[[beta].sub.5]In[(Trade).sub.k,j,t] + [[beta].sub.2]Labor [Cost.sub.k,t] + [[beta].sub.7]Institutional [Quality.sub.k,t] +[[beta].sub.8]In[(Exchange Rate).sub.k,j,t] + [[beta].sub.9][NAFTA.sub.k] + [[beta].sub.10][ANZCERTA.sub.k] +[[beta].sub.11][EU1992.sub.k,t] + [[beta].sub.12]Cultural [Distance.sub.j,k] + [[epsilon].sub.j,k,t]
This specification includes the variables of interest from our hypotheses as well as customary control variables. Table 1 provides an overview of the variable definitions and their sources. We use an augmented gravity model commonly used for the analysis of international trade flows, which also appears to fit FDI flow data quite well (Balasubramanyam et al. 2002; Blonigen 2005; Daude and Stein 2007).
To test our first set of hypotheses (1-1.5) related to regional integration and the effect of accession on FDI flows, we restrict our sample to only those countries that joined either ANZCERTA, the EU or NAFTA in our sample period and we utilize an indicator variable: RIA. RIA is a zero-one indicator variable that takes on a value of one once the host country becomes a member of an RIA. Although existing evidence generally shows a positive impact from regional integration on FDI for the region as a whole, theory suggests that the impact on individual members may depend on their relative locational advantages within the RIA; however, on average, we expect a positive sign on RIA as predicted in Hypothesis 1. Also, as studies have shown that joining an RIA may result in a regime change in the structural determinants of inward FDI, we add interaction variables with RIA and the other country-specific determinants.
Host country market size is an important indicator of the attractiveness of an FDI location for downstream FDI; countries with larger GDP represent more profitable investment opportunities (Buckley et al. 2001; Globerman and Shapiro 1999). We measure host country market size using the host country GDP in USD. Consistent with existing evidence, we posit a positive relationship between host country market size and FDI inflows. To test Hypothesis 1.1., namely that a country's size matters less after it joins an RIA we add an interaction term for host country GDP and RIA in the regression analysis. A negative coefficient on the interaction term would be consistent with our predictions.
The size of the home country market is also expected to influence a country's propensity to make outward FDI, though the direction of the effect is ambiguous (Globerman and Shapiro 1999). On the one hand, firms from countries with a large home market could have greater cash flows for making FDI (Grosse and Trevino 1996) or possess more of the intangible assets necessary to succeed in FDI (Anand and Kogut 1997; Dunning 1997); however, the same home country characteristic can make domestic investments more attractive compared to FDI. We measure the home country market size by using the home country GDP expressed in USD. We do not posit any changes in this relationship post-accession.
In gravity models of trade, geographic distance is expected to have a negative sign. However, it is not possible to posit ex ante an unambiguous effect of geographic distance on FDI. As geographic distance increases, it is possible that FDI becomes more attractive relative to trade; however, greater geographical distance is expected to result in greater costs in information gathering and managing the affiliate (Hejazi and Safarian 2005). Following regional integration, geographic distance is expected to be a more important determinant of FDI inflows (Hypothesis 1.2) and, therefore, we expect an additional incremental effect of geographic distance on FDI inflows after a country joins an RIA.
Bilateral trading relations between the host and home countries have been found to be a significant determinant of FDI flows. Though, in theory, trade and FDI can be substitutes or complements, recent empirical studies on North America (Eden and Li 2004; Feils and Rahman 1998; Grosse and Trevino 1996) have generally found a positive relationship between the two; Globerman and Shapiro (1999) suggest that in markets where trade is relatively unrestricted, the two "are more likely to be complements given the propensity of multinational companies to engage in vertical and horizontal specialization of production within their affiliate networks" (p. 520). Therefore, we expect a positive relationship between trade and FDI inflows. Trade is measured as the sum of bilateral imports and exports in any given year measured in USD. (5) A priori, there is no reason to believe that trade will be more or less important of a predictor of FDI inflows following regional integration.
The efficiency of a nation's institutions is expected to be a determinant measure of its attractiveness as an FDI destination (Balasubramanyam et al. 2002; Daude and Stein 2007; Globerman and Shapiro 2003; Mina 2009). We include three different measures of the efficiency of a country's institutional environment.6 Our first measure is the quality of the legal and regulatory environment, as collected by the IMD for the Global Competitveness Report. Our next two measures are based on the World Governance Indicators (see Kaufmann et al. 2007). There are six indicators, namely voice and accountability, political stability and absence of violence, government effectiveness, regulatory quality, rule of law, and control of corruption, which are highly correlated. Following Globerman and Shapiro (2003) we use the first principal component of these indicators as a measure of a country's governance. As a check, we also report the results on using regulatory quality since this component of the world governance indicators has been shown to matter for FDI. We expect that all measures of institutional quality will be positively related to FDI. Once a country has accessed a regionally integrated area, we expect its institutional environment to matter more, as predicted in Hypothesis 1.3.
Countries with the highest labor force efficiency are more attractive to upstream FDI. We use the ratio of labor cost to productivity as a proxy for labor force efficiency. When firms choose the location for their FDI, they may consider the country's labor efficiency relative to that of the other countries in the RIA. Therefore, we use the productivity-adjusted labor cost of a country relative to the average productivity-adjusted labor cost of the integrated region. The higher the labor cost in a country relative to that of other countries in the RIA, the less FDI is expected to flow into a country. Thus, we expect a negative coefficient on this variable. Following accession to an RIA, labor cost should be a more important determinant of FDI inflows (Hypothesis 1.4), and therefore, we expect a negative sign on the interaction between RIA and labor costs.
Cultural distance is expected to affect firms' FDI decisions (e.g., Feils and Rahman 1998; Globerman and Shapiro 2003; Grosse and Trevino 1996; Sethi et al. 2003). Cultural distance--i.e., differences in language, religion, legal systems, etc.--are likely to make integration of the subsidiary harder and more costly and thus should have a negative effect on FDI. We use Kogut and Singh's (1988) measure of cultural distance based on Hofstede's dimensions of culture in the regression analyses. Following the accession to an RIA, we expect cultural distance to matter less as predicted in Hypothesis 1.5.
The foreign exchange rate may also affect FDI flows. A weak host country currency allows firms from the home country to acquire a larger investment (for example Blonigen 2005; Froot and Stein 1991). We expect a negative relationship between a strong host country currency and FDI inflows.
As explained in more detail in the Appendix, the three regional agreements we study in this paper differ in the extent to which they pursue integration. In order to control for the type of agreement, we included three indicator variables. NAFTA takes on a value of one if the host country is one of the NAFTA member countries, ANZCERTA takes on a value of one if the host country is one of the ANZCERTA member countries and EU 1992 takes on a value of one if the host country is one of the EU member countries and the time period is 1992 or thereafter.7
To test our second set of hypotheses related to regional integration and intra-RIA FDI we utilize an indicator variable: Intra-RIA. Intra-RIA is a zero-one indicator variable that takes on a value of one if the home and the host countries are members of the same RIA. We expect to see more FDI from insider countries as predicted in Hypothesis 2. In addition to the customary control variables for FDI inflows explained above, we use the following three interaction terms in the second set of regressions to test for hypotheses 2.1, 2.2, and 2.3.
To test Hypothesis 2.1--that an individual RIA member's market size is more important for fellow RIA member countries--we include the interaction between Intra-RIA and the host country GDP in the analysis. We expect a positive sign for the interaction term.
We propose that a country's institutional efficiency becomes a more important determinant of inward FDI from member countries (Hypothesis 2.2). We include the interaction between Intra-RIA and institutional efficiency in the analysis and we expect a positive sign on the interaction term.
In order to examine whether a member country's labor efficiency is viewed differently by insider MNCs (Hypothesis 2.3), we include the interaction between Intra-RIA member and the productivity-adjusted labor cost in the regressions. A negative coefficient would suggest that insider MNCs tend to value members' labor efficiency more than outsider MNCs.
The Pearson correlation coefficients for the independent variables are reported in Table 2. The signs of the individual correlation coefficients are consistent with expectations. The various measures of institutional quality are very highly correlated. Therefore, we repeat the analysis for each of the three measures of institutional quality. In order to mitigate potential multicollinearity for the interaction variables, we center the variables by subtracting the sample mean of each variable from the variable series and use these centered variables in our regressions (Hamilton 2004; Wooldridge 2005).
We run two separate groups of pooled, cross-sectional regression tests, one restricted to countries that joined an RIA during the sample period, and the other for the entire sample of countries. The focus of the first group is to examine whether joining an RIA has an impact on inward FDI, and whether the impact diverges based on country-specific characteristics. The focus of the second set of tests is on examining whether common RIA membership has an additional impact on inward FDI.
FDI Inflows Following Accession to an RIA
In order to capture the effect, if any, of accession to an RIA, we ran our basic regression model restricting the sample to countries that joined an RIA during the 1980-2003 period (about 30% of the full sample). Models 1-3 in Table 3 differ in the alternative measures of institutional efficiency, namely, indicators of quality of the institutions of governance, the legal environment, and the regulatory environment. The core gravity model variables, GDP Host and GDP Home have the expected positive sign and are significant; Geographic Distance has a negative sign and is also significant. These results are consistent with most prior studies on FDI flows.8 RIA has a positive and significant coefficient in all the models: on average, joining an RIA leads to a secular increase in inward FDI. This is, of course, consistent with prior evidence, as incorporated in Hypothesis 1, that regional integration generally increases FDI into the region. Also, members with lower relative labor costs and greater institutional efficiency receive larger FDI inflows. Exchange rate, though barely significant (only in Model 3), has the expected negative sign, as higher host country currency would be expected to act as a deterrent. Cultural affinity between home and host nations appear to matter, in that MNCs direct greater investments to host nations that are culturally closer to themselves. Finally, we checked to see whether being a member of a specific 'home region' RIA has any additional effect, and find that in all three cases there was an additional positive effect to joining an RIA relative to the base case (the EU prior to 1992).
To see if joining an RIA changes the underlying structural regime of FDI determinants, Models 4-6 repeat the regressions, adding the interaction terms with RIA for all the control variables. The results indicate that a nation's FDI policy regime undergoes significant changes following accession into an RIA. Looking at the interaction terms, it appears that host country GDP is less important once the host country provides access to the larger market of an RIA, which is consistent with Hypothesis 1.1. Geographic distance, as expected, becomes a bigger deterrent post-accession, whereas lack of cultural affinity diminishes in importance, also as expected, after accession. Following accession to an RIA, the proxies for institutional efficiency have the expected positive sign and are marginally significant; the results seem to indicate that once an RIA pact provides alternative access to the member country's markets, MNCs favor locations with relatively higher quality institutional infrastructure. However, contrary to expectations, labor cost efficiency is less important after accession. Finally, trade and exchange rates increase in saliency in the post-accession period; however, after accession there is an increase in FDI inflows from smaller market economies. We did not posit any a priori expectations for these variables based on our analysis of the FDI literature, and thus present these, at this juncture, as stylized facts regarding the change in the regime of FDI determinants following regional integration.
Regionalization and Intra-RIA FDI Flows
Table4 presents the results from the pooled, cross-sectional regressions for the full sample. Models 1-3 are similar to Models 1 3 in Table 3 except for the replacement of RIA by an indicator variable, Intra-RIA, which takes the value of 1 when the home and host countries are members of the same RIA. Results on the control variables are similar to those for the restricted sample. In all three models, the indicator variable, Intra-RIA, is positive and highly significant, which is consistent with Hypothesis 2; it appears that common RIA membership increases FDI flows, controlling for the typical determinants of FDI. This result mirrors the firm-level evidence provided on the largest MNCs. As in the restricted sample, trade has a positive coefficient indicating that trade and FDI complement each other. The host country's productivity adjusted relative labor costs has the expected negative coefficient, and is significant; a nation's labor cost appears to be a relevant locational advantage. Also, the country's institutional efficiency has the expected positive effect on FDI. Just like in the restricted sample of Table 3, both cultural distance from the home country and higher-valued host currency reduces FDI inflows. However, in the full sample, being a member of NAFTA or ANZCERTA, did not lead to additional FDI inflows; in fact, compared to the other nations, NAFTA members appeared to have received less FDI.
Models 4-6 replicate the respective regressions with the addition of three interaction terms to investigate whether regional integration changes how insider MNCs perceive the attractiveness of fellow RIA member host countries, namely, Intra-RIA*GDP Host, Intra-RIA*Labor Cost, and Intra-RIA*Institutions. These variables have been included to explore the reasons behind the higher FDI flows observed among common RIA members and yield some interesting results. First, all the other control variables retain the same sign and significance as in Models 1-3. The indicator variable, Intra-RIA, is still positive and highly significant. Two of our three key interactive variables, Intra-RIA*GDP Host and Intra-RIA*Institutions, have the expected signs and are significant. Intra-RIA* Labor Cost is also significant but has a positive rather than the expected negative sign, indicating that labor costs are less important for insider FDI following accession.
These results illuminate and highlight the complex impact of regional economic integration on FDI flows. The secular, positive relationship evidenced in the Intra-RIA FDI flows in the macro-level data is consistent with the various 'home bias' explanations within FDI theory (Wolf et al. 2008). The fact that the insider FDI flows are greater for members with larger market size can be viewed as sustaining the Rugman and Verbeke (2005) thesis in that market-seeking FDI is more feasible when an RIA pact lowers the liability of foreignness. We also observe that MNCs from fellow RIA members value the host country's institutional efficiency above and beyond the fact of sharing a home RIA. Finally, and curiously, insider MNCs do not appear to value labor cost efficiency; this result, which is also similar to our findings from our examination of the effects of accession using the restricted sample as presented in Table3, maybe indicating investment diversion to higher labor cost RIA-members as MNCs seek the advantages of producing within the RIA boundaries (Rugman and Gestrin 1993).
Economic performance following regional integration is a key matter of interest to policy makers and businesses. In this paper we examine the impact of regional economic integration on FDI, using a large sample of bilateral FDI inflows into 24 OECD countries; the dataset includes three RIAs which encapsulate the 'home' region characteristic in IB literature. First, we consider whether accession to an RIA leads to both a secular increase in FDI inflows, as well as changing the structural regime of the determinants of FDI, and find the answer to be affirmative on both counts. On average, host country market size and lack of cultural affinity are less important determinants once the host country joins an RIA, suggesting MNCs value the access to the larger market that joining an RIA provides to individual countries. Also, we find that following accession, the host country's institutional efficiency becomes a more important determinant, suggesting that prior to integration, MNCs may have been more willing to trade off market access for institutional efficiency. However, contrary to expectations, labor costs become less important post-accession, perhaps due to high labor costs proxying for larger market potential for downstream FDI and/or reflecting investment diversion.
Focusing on the intra-regional relationships, we find that common RIA membership increases FDI inflows, in that, on average, FDI by insider MNCs is higher. The observed increase in FDI among and between the RIA members is even greater for the larger economies within the RIA, which is consistent with prior findings on the EU and NAFTA. This finding also parallels the firm-level studies that suggest that market-seeking FDI benefits from the lowering of the liability of foreignness from a regional integration pact. In addition, intra-RIA FDI seeks out the fellow RIA host countries that provide a more efficient institutional framework, which is consistent with discussion in the IB literature of firms engaging in locational reshuffling within the home region following the formation of an RIA. Furthermore, we find some evidence that is consistent with RIA members diverting FDI into less labor cost efficient, fellow RIA member countries.
We believe this evidence, based on an expansive dataset of bilateral FD1 flows across several RIAs, adds to our understanding of the dynamics of regional integration in the FDI decision, as well as providing complementary insights into the regionalization vs. globalization debate. The current findings should be useful to policy-makers at both the regional and national levels, in both pre- and post-integration periods, and both in terms of implementation strategies, as well as in guiding public expectations. As regions such as the EU continue to widen their integration area, we naturally see increasing disparity within the membership in terms of market size, labor productivity, and relative economic effectiveness. While on average integration leads to an increase in FDI inflows, there are significant differences related to the characteristics of the destination countries.
Limitations and Future Research
There are a number of limitations of our research which point to future research directions. First, there are more RIAs beyond those investigated here, and future research should investigate whether the kind of 'home region' bias seen in these three RIAs also obtain among others. While we find the EU, NAFTA, and ANZCERTA to be particularly apt in capturing the 'home region' criteria insofar as the regionalization/globalization debate is concerned (see, for example, Fratianni and Oh 2009), clearly other RIAs should also be tested to see whether similar results obtain. In addition, each RIA contains unique provisions. A study examining the different characteristics of RIAs, such as specific trade and investment provisions, would also be useful in explaining which facets of the RIA serve to attract FDI.
Second, this study uses macro-level data though the IB literature on FDI used to formulate the hypotheses is based on firm-level strategies of MNCs. The observed heterogeneity in the gains/losses related to member country characteristics introduces an important dimension to the study of FDI decisions. The literature on FDI suggests such divergent effects across member countries within an RIA contingent on firm-level motivations for FDI. We have attempted to capture the differential effects using macro-level data, relying on the expected differences in country-level indicators corresponding to country-specific characteristics associated with different types of FDI motivations. Although this is precisely the kind of study--i.e., using RIAs to investigate the globalization/regionalization debate--that the IB literature has called for, the absence of direct measures of firm-level decision-making is a natural limitation of this stream of research. For example, the limitations of our dataset do not allow us to distinguish between FDI through acquisitions vs. greenfield investments, horizontal vs. vertical FDI, or export-oriented vs. local market based FDI. This research needs to be conducted at the firm level and would complement the results of our study.
While we have striven to link our empirical findings to the theoretical literature as apposite, some of our other findings--e.g., the observed increase in FDI inflows from smaller home countries in the post-accession period--must necessarily be viewed, at this juncture, as stylized facts regarding the change in the regime of FDI decision-making following regional integration in the host country. Future research that examines FDI patterns using firm- and/or industry-level data would be helpful in further developing our understanding of the effects of regional integration on FDI decisions.
Appendix: The Main Elements of the RIAs
ANZCERTA: The Australia New Zealand Closer Economic Relations Trade Agreement came into effect January 1, 1983. The two member countries are Australia and New Zealand. Prior to ANZCERTA, Australia and New Zealand had signed other preferential trade agreements that resulted in removing restrictions on about 80% of the bilateral trade. However, the remaining restrictions could not be dealt with under the existing agreements and ANZCERTA was introduced. ANZCERTA "has been described by the World Trade Organization (WTO) as 'one of the most comprehensive, effective and mutually compatible trade agreements in the world'" (Aorere 2008).
Originally, ANZCERTA provided for free trade in goods, and by 1990 all tariffs and quotas were removed. In 1988, services were also included under the ANZCERTA provisions. In addition, ANZCERTA contains provisions on the mutual recognition of goods and occupations, i.e. allows also for labour mobility (Aorere 2008). A number of agreements were added to ANZERTA over time, such as an open skies agreement, government procurement agreement and the Trans-Tasman Mutual Recognition Agreement (Closer Economic Relations 2009). The agreement has no specific investment provisions, however, the two countries are working towards creating a Single Economic Market by coordinating business laws, bank supervision and accounting and auditing standards (Aorere 2008).
NAFTA: The North American Free Trade Agreement can into effect January 1, 1994 and includes Canada, Mexico and the United States. NAFTA extended the Canada-U.S. Free Trade Agreement to include Mexico and added additional provisions to the agreement. The objectives of NAFTA as stated in article 102 of the agreement are free trade in goods and services, increased investment opportunities, protection of intellectual property rights and the establishment of a dispute settlement mechanism (NAFTA 2009). By 2008, tariffs and quotas on 99% of goods and services were eliminated. While the focus of NAFTA is trade liberalization, the agreement does contain investment provisions in the well-known Chap. 11. Specifically, Chap. 11 calls for (1) national treatment of foreign investments with a number of industry exceptions; (2) most favoured nation status for investments; (3) "stipulates that disputes between NAFTA investors and any of the parties to the agreement can go to binding, enforceable international arbitration" (Rugman and Gestrin 1999, p. 2); and (4) a prohibition of performance requirements with some exceptions and greater transparency in the discriminatory measures (see Rugman and Gestrin 1999; Wilkie 2002).
EU: The European Union is the oldest and most dynamic regionally integrated area of the three areas in our study. Since the Treaty of Rome was signed by six European countries in 1957 establishing the European Economic Community, the number of member countries has increased to 27 at the current time. The integration changed its focus from an "economic community" to a "new structure which is political as well as economic": the European Union (EUROPA 2009) with the Treaty of Maastricht on the European Union, that was signed in 1992. The three pillars of the EU are the European Community, the common foreign and security policy, and field of justice and home affairs. The treaty also set the stage for the economic and monetary union, which ultimately led to the introduction of the euro in 1999. Thus, 1992 represents a year in which the EU decided to integrate much further than previously envisioned.
Acknowledgements: We would like to thank Florin Sabac, the reviewers and the editor for their helpful comments and suggestions. We are grateful for the financial support provided by the Western Centre for Economic Research, the Centre for International Business Studies, and the J. D. Muir fund at the Alberta School of Business, University of Alberta.
Altomonte, C. (2007). Regional integration and the location of multinational firms. Review of World Economics, 143(2), 277-305.
Anand, J., & Kogut, B. (1997). Technological capabilities of countries, firm rivalry and foreign direct investment. Journal of International Business Studies, 28(3), 445465.
Aorere, M. (2008). CER: A closer economic relationship. New Zealand Ministry of Foreign Affairs & Trade. http://www.mfat.govt.nz/Foreign-Relations/Australia/1-CER/index.php. Accessed 24 Apr 2009.
Balasubramanyam, V., Sapsford, D., & Griffiths, D. (2002). Regional integration agreements and foreign direct investment: Theory and preliminary evidence. The Manchester School, 70(3), 460-482.
Baldwin, R., & Venables, A. (1995). Regional economic integration. In G. Grossman & K. Rogoff (Eds.), Handbook of international economics (Vol. III, pp. 1597-1644). Amsterdam: North-Holland.
Bevan, A., & Estrin, S. (2004). The determinants of foreign direct investment into European transition economies. Journal of Comparative Economics, 32(4), 775-787.
Blomstrom, M., & Kokko, A. (1997). Regional integration and foreign direct investment. NBER Working Paper 6019.
Blonigen, B. (2005). Review of the empirical literature on FDI determinants. Atlantic Economic Journal, 33(3), 383-403.
Brewer, P. (2007). Operationalizing psychic distance: A revised approach. Journal of International Marketing, 15(1), 44-66.
Buckley, P. J., Clegg, J., Forsans, N., & Reilly, K.T. (2001). Increasing the size of "country": Regional economic integration and foreign direct investment in a globalized world economy. Management International Review, 41(3), 251-274.
Buckley, E J., Clegg, J., Forsans, N., & Reilly, K. T. (2007a). A simple and flexible dynamic approach to foreign direct investment growth: The Canada-united states relationship in the context of free trade. World Economy, 30(2), 267-291.
Buckley, P., Clegg, L., Cross, A., Liu, X., Voss, H., & Zheng, E (2007b). The determinants of Chinese outward foreign investment. Journal of International Business Studies, 38(4), 499-518.
Closer Economic Relations (CER). (2009). Australian High Commission New Zealand. http://www. australia.org.nz/wltn/CloseEconRel.html. Accessed 24 Apr 2009.
Daude, C., & Stein, E. (2007). The quality of institutions and foreign direct investment. Economics & Politics, 19(3), 317-344.
Daude, C., Levy-Yeyati, E., & Stein, E. (2003). Regional integration and the location of foreign direct investment. Working Paper 492, Research Department, Inter-Amercian Development Bank, Washington, DC.
Dunning, J. H. (1993). The globalization of business. London: Routledge.
Dunning, J. H. (1997). The European internal market programme and inbound foreign direct investment. Journal of Common Market Studies, 35(1), 1-30.
Dunning, J. H., Fujita, M., & Yakova, N. (2007). Some macro-data on the regionalisation/globalisation debate: A comment on the Rugman/Verbeke Analysis. Journal of International Business Studies, 38(1), 177-199.
Eden, L. (2002). Regional integration and foreign direct investment: Theory and lessons from NAFTA. In M. Kotabe & P. Aulakh (Eds.), Emerging issues in international business research (pp. 15-36). Cheltenham: Elgar.
Eden, L., & Li, D. (2004). Insiders, outsiders and foreign direct investment after NAFTA. Paper presented at 9th Annual Conference, Texas A&M International University, "Trade & Investment in the Americas: The Challenge of Globalization" Laredo, Texas, July 2004.
Ethier, W. J. (1998). The new regionalism. Economic Journal, 108(449), 1149-116l.
EUROPA. (2009). Treaty of Maastricht on European Union. http://europa.eu/scadplus/treaties/ maastricht_en.htm. Accessed 15 Apr 2009.
Feils, D., & Rahman, M. (1998). Foreign direct investment in Canada. Global Business and Finance Review, 3(2), 55-62.
Feils, D., & Rahman, M. (2008). Regional economic integration and foreign direct investment: The case of NAFTA. Management International Review, 48(2), 147-163.
Fratianni, M., & Oh, C. H. (2009). Expanding RTAs, trade flows, and the multinational enterprise. Journal of International Business Studies, 40(7), 1206-1227.
Froot, K., & Stein, J. (1991). Exchange rates and foreign direct investment: An imperfect capital markets approach. Quarterly Journal of Economics, 106(4), 119l-1218.
Fujita, M., & Mori, T. (2005). Frontiers of the new economic geography. Papers in Regional Science, 84(3), 377-409.
Globerman, S., & Shapiro, D. M. (1999). The impact of government policies on foreign direct investment: The Canadian experience. Journal of International Business Studies, 30(3), 513-532.
Globerman, S., & Shapiro, D. M. (2003). Governance infrastructure and US foreign direct investment. Journal of International Business Studies, 34(1), 19-39.
Grosse, R., & Trevino, L. (1996). Foreign direct investment in the United States: An analysis by country of origin. Journal of International Business Studies, 2 7(1), 139-155.
Hamilton, L. (2004). Statistics with STATA. Belmont: Brooks/Cole--Thomson Learning.
Hannan, M., & Freeman, J. (1977). The population ecology of organizations. The American Journal of Sociology, 82(5), 929-964.
Hejazi, W. (2005). Are regional concentrations of OECD exports and outward FDI consistent with gravity? Atlantic Economic Journal, 33(4), 423-436.
Hejazi, W., & Safarian, A. E. (2005). NAFTA effects and the level of development. Journal of Business Research, 58(12), 1741 1749.
Hofstede, G. (1980). Culture's consequences: International differences in work-related values. Beverly Hills: Sage.
Johanson, J., & Vahlne, J. E. (1977). The internationalization process of the firm--a model of knowledge development and increasing foreign market commitments. Journal of International Business Studies, 8(1), 23-32.
Johanson, J., & Wiedersheim-Paul, F. (1975). The internationalisation of the firm: Four Swedish case studies. Journal of Management Studies, 12(3), 305-322.
Kaufmann, D., Kraay, A., & Mastruzzi, M. (2007). Governance matters VI: Aggregate and individual governance indicators 1996-2006. The World Bank, WPS World Bank Policy Research Working Paper Series 4280, 1-93.
Kogut, B., & Singh, H. (1988). The effect of culture on the choice of entry mode. Journal of International Business Studies, 19(3), 411-432.
McCann, E, & Simonen, J. (2005). Innovation, knowledge spillovers and local labour markets. Papers in Regional Science, 84(3), 465-485.
Mina, W. (2009). External commitment mechanisms, institutions, an FDI in GCC countries. Journal of International Financial Markets, Institutions, and Money, 19(2), 371-386.
NAFTA: Part One: General Part. (2009). Foreign affairs and international trade Canada. http:// www.international.gc.ca. Accessed 20 Apr 2009.
Osegowitsch, T., & Sammartino, A. (2008). Reassessing (home-)regionalisation. Journal of International Business Studies, 39(2), 184-196.
Porter, M. (1990). The competitive advantage of nations. New York: Free.
Rugman, A., & Gestrin, M. (1993). The strategic response of multinational enterprises to NAFTA. Columbia Journal of World Business, 28(4), 18-29.
Rugman, A., & Gestrin, M. (1999). The investment provisions of NAFTA. In S. Globerman & M. Walker (Eds.), Assessing NAFTA: A trinational analysis. Vancouver: Fraser Institute.
Rugman, A., & Verbeke, A. (2004). A perspective on regional and global strategies of multinational enterprises. Journal of International Business Studies, 35(1), 3-18.
Rugman, A., & Verbeke, A. (2005). Towards a theory of regional multinationals: A transactions cost economics approach. Management International Review, 45(SI 1), 5-17.
Rugman, A., & Verbeke, A. (2007). Liabilities of regional foreignness and the use of firm-level versus country-level data: A response to dunning et al. (2007). Journal of International Business Studies, 38(1), 200-205.
Sethi, D., Guisinger, S. E., Phelan, S. E., & Berg, D. M. (2003). Trends in foreign direct investment flows: A theoretical and empirical analysis. Journal of International Business Studies, 34(4), 315-326.
Stephan, M., & Pfaffmann, E. (2001). Detecting the pitfalls of data on foreign direct investment: Scope and limitations of FDI data. Management International Review, 41(2), 189-218.
te Velde, D., & Bezemer, D. (2006). Regional integration and foreign direct investment in developing countries. Transnational Corporations, 15(2), 41-70.
Thomas, D., & Grosse, R. (2001). Country-of-origin determinants of foreign direct investment in an emerging market: The case of Mexico. Journal of International Management, 7(1), 59-79.
UNCTAD. (2006). World Investment Report 2006. New York: United Nations.
Wilkie, C. (2002). The origins of NAFTA investment provisions: Economy and policy considerations. In: L. Dawson (Ed.), Whose rights? The NAFTA chapter 11 debate (pp. 5-41), Ottawa: Centre for Trade Policy and Law, Carleton University.
Wolf, J., Dunemann, T., & Egelhoff, W. G. (2008). Economic, psychological, and sociological theories for the explanation of home-region oriented MNCs. Academy of Management, Best Paper Proceedings.
Wooldridge, J. (2002). Econometric analysis of cross section and panel data. Cambridge: MIT Press.
Wooldridge, J. (2005). Introductory econometrics: A modern approach. Mason: Thomson South-Western.
Wu, S. (2006). Corruption and cross-border investment by multinational firms. Journal of Comparative Economics, 34(4), 839-856.
(1) Note that there are different degrees of regional integration (e.g., Baldwin and Venables 1995) and that the term is used differently in the literature (e.g., Altomonte 2007; Blomstrom and Kokko 1997; Fratianni and Oh 2009; te Velde and Bezemer 2006). In our study, RIAs refer to those areas of adjacent countries that are characterized not only by trade liberalization but also include investment liberalization and/or other institutional changes (BlomstrOm and Kokko 1997). The degree of investment liberalization and the type of additional institutional changes are unique to each RIA. The major provisions of the three RIAs in our sample are summarized in the Appendix. Our selection of RIAs is consistent with Ethier's (1998) characterization of the New Regionalism trend. He explicitly discusses the EU and NAFTA as two of the regions that fit his description; ANZCERTA also fulfills his criteria for a 'home' region. Note that in all regions some or all of the new member countries had already some free trade agreements with each other. In this paper, we do not test the effect a free trade agreement on FDI but rather that of regional integration in the sense proposed by Ethier (1998), Rugman and Verbeke (2004, 2005), and Wolf et al. (2008).
(2) Note, the literature also classifies downstream FDI as market-seeking, and upstream FDI as efficiency-seeking and resource-seeking (e.g., Dunning 1993).
(3) Changes in FDI stock may be used as an alternative dependent variable. Stephan and Pfaffmann (2001) provide a discussion of data limitations for both FD1 stock and flow data. Some recent studies that use the FDI flow data are Buckley et al. (2007b), Thomas and Grosse (2001), and Wu (2006).
(4) We dropped Belgium and Luxembourg from the sample because the information on these two countries is often combined and the individual country effects cannot be determined. Information on the member countries of the Central European Free Trade Agreement was often missing Therefore we dropped these countries as well from the final sample. Due to missing observations the panels are not balanced.
(5) Exports and imports are highly correlated. Therefore, we use the sum of exports and imports to measure the bilateral trade intensity.
(6) The world governance indicators and the world competitiveness report are not available for the entire time horizon of our sample period. We extrapolated the missing observations for the legal and regulatory environment and we extra- and intrapolated the missing observations for the world governance indicators.
(7) We used EU 1992 to take into account the deepening of the European Union after 1992; we thank an anonymous referee for the suggestion.
(8) Although we did not posit a sign for Geographic Distance, Daude and Stein (2007) and Balasubramanyam et al. (2002) found a negative and significant coefficient for it.
Table 1: Variable definitions Variable Definition Source Pred. sign GDP GDP of the home country measured in IMF home current USD GDP host GDP of the host country measured in IMF + current USD Geograph- Geographic distance (from capital to CEPII is distance capital) measured in kilometres RIA Indicator variable: 1 if the host + country is a member of an RIA, zero otherwise Intra-RIA Indicator variable: 1 if the home + country and the host country belong to the same RIA, 0 if the home country is from outside the RIA Labor cost Ratio of the productivity-adjusted US dept. - labor cost of the host country of labor; [hourly manufacturing labor rate divided by the host country hourly GGDC output adjusted for PPP (both measured in US dollars)] to the average productivity-adjusted labor cost of the host RIA Govern- First principal component of the six World Bank + ance governance measures provided by the world bank, with values ranging approximately from -2.5 to +2.5. A higher value indicates better governance (see Kaufmann et al. 2007) Legal Measures "the legal and regulatory IMD + framework encourages the competitiveness of enterprises" (IMD) on a scale of 0-10. This index constitutes part of the global competitiveness index developed by the IMD. A higher score indicates a better legal and regulatory environment Regulato- One of the six governance indicators World Bank + ry quality provided by the world bank which measures the "ability of the government to formulate and implement sound policies and regulations that permit and promote private sector development" (Kaufmann et al. 2007, p.4) on a scale from approximately -2.5 to +2.5. A higher score indicates better regulatory quality NAFTA Indicator variable that takes on a value of one if the host country is one of the NAFTA countries ANZCER- Indicator variable that takes on a TA value of one if the host country is one of the ANZCERTA countries EU 1992 Indicator variable that takes on a value of one if the host country is one of the EU countries and the year is 1992 or later Exchange The natural log of the average UN - rate exchange rate between the home and the host country in a given year expressed as home currency/host currency Cultural Adapted from Kogut and Singh (1988) Hofstede - distance (1980) [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] [CD.sub.jk] refers to the cultural distance of home country j to the host country k based on Hofstede's (1980) four dimensions of culture: power distance (PD), uncertainty avoidance (UA), individualism (IND), and masculinity/femininity (MF). VAR measures the variance of the different dimensions of culture as defined by Hofstede (1980) Table 2: Variable description and pearson correlation coefficients Mean SD 1 2 3 1. GDP home (In) 25.37 1.51 1.0000 2. GDP host (In) 26.19 1.54 0.0670 * 1.0000 3. Geographic 8.38 1.06 -0.0429 * 0.0379 * 1.0000 dist. (In) 4. Trade (In) 5.90 2.39 0.5538 * 0.6163 * -0.3169 * 5. RIA 0.67 0.47 0.0734 * -0.0255 * -0.0616 * 6.Intra-RIA 0.09 0.29 0.2283 * 0.0553 * -0.4265 * 7. Labor cost 1.00 0.31 0.0000 0.2557 * -0.0643 * 8. Governance 3.19 1.37 0.0095 -0.0337 * -0.0799 * 9. Legal 5.89 1.38 0.0088 -0.2719 * 0.0307 * 10. Regulatory 1.06 0.39 0.0594 * 0.0561 * -0.0495 * quality, 11. Foreign 0.59 3.68 -0.0925 * 0.1056 * 0.0567 * exchange (In) 12. NAFTA 0.13 0.33 0.0000 0.3438 * 0.2040 * 13.ANZCERTA 0.08 0.28 0.0000 -0.1509 * 0.3037 * 14. EU 1992 0.27 0.44 0.1227 * 0.1547 * -0.1812 * 15. Cultural 2.07 1.42 -0.1572 * -0.0622 * 0.1697 * distance 4 5 6 7 1. GDP home (In) 2. GDP host (In) 3. Geographic dist. (In) 4. Trade (In) 1.0000 5. RIA 0.0238 * 1.0000 6.Intra-RIA 0.3125 * 0.2222 * 1.0000 7. Labor cost 0.2117 * -0.0920 * -0.0302 * 1.0000 8. Governance 0.0742 * 0.3081 * 0.0537 * 0.6736 * 9. Legal -0.1310 * 0.2518 * 0.0094 0.3281 * 10. Regulatory 0.1083 * 0.3606 * 0.1137 * 0.3334 * quality 11. Foreign -0.0283 * 0.3336 * -0.0539 * -0.0816 * exchange (In) 12. NAFTA 0.0901 * -0.0906 * -0.0882 * 0.0000 13.ANZCERTA -0.1601 * 0.1457 * -0.0639 * 0.0000 14. EU 1992 0.1803 * 0.3546 * 0.2281 * 0.0000 15. Cultural -0.1527 * 0.0474 * -0.0692 * 0.1288 * distance 8 9 10 11 1. GDP home (In) 2. GDP host (In) 3. Geographic dist. (In) 4. Trade (In) 5. RIA 6.Intra-RIA 7. Labor cost 8. Governance 1.0000 9. Legal 0.6949 * 1.0000 10. Regulatory 0.7387 * 0.7313 * 1.0000 quality, 11. Foreign 0.3461 * 0.2151 * 0.4040 * 1.0000 exchange (In) 12. NAFTA -0.1949 * -0.0639 * -0.0555 * 0.1037 * 13.ANZCERTA 0.1873 * 0.3483 * 0.2607 * 0.0718 * 14. EU 1992 0.1520 * 0.0427 * 0.3149 * 0.2570 * 15. Cultural 0.2025 * 0.1346 * 0.1542 * 0.0671 * distance 12 13 14 15 1. GDP home (In) 2. GDP host (In) 3. Geographic dist. (In) 4. Trade (In) 5. RIA 6.Intra-RIA 7. Labor cost 8. Governance 9. Legal 10. Regulatory quality, 11. Foreign exchange (In) 12. NAFTA 1.0000 13.ANZCERTA -0.1140 * 1.0000 14. EU 1992 -0.2304 * -0.1838 * 1.0000 15. Cultural -0.0740 * -0.0061 0.0439 * 1.0000 distance * Significant at the 5% level Table 3: Random effects model regression results. The dependent variable is the log of the foreign direct investment inflows. The sample is restricted to those countries that changed from non-RIA member to RIA member during the sample period Model 1 Model 2 Model 3 governance legal regulatory quality GDP home (ln) 0.569 0.532 0.515 (10.42) *** (9.83) *** (9.57) *** GDP host (ln) 0.336 0.400 0.383 (4.52) *** (5.30) *** (5.19) *** Geographic -0.862 -0.822 -0.836 distance (ln) (9.06) *** (8.64) *** (8.82) *** trade (ln) 0.355 0.381 0.365 (7.70) *** (8.47) *** (8.18) *** RIA 0.348 0.312 0.283 (5.13) *** (4.59) *** (4.24) *** Labor cost -0.641 -0.518 -0.729 (2.57) ** (2.42) ** (3.41) *** Institutions 0.165 0.171 1.254 (2.28) ** (5.20) *** (10.79) *** Exchange rate -0.002 -0.01 -0.029 (In) -0.11 -0.56 (1.67) * NAFTA 1.948 1.730 1.721 (7.20) *** (7.01) *** (6.99) *** ANZCERTA 1.905 1.786 1.597 (8.29) *** (7.74) *** (6.94) *** EU 1992 0.527 0.526 0.345 (7.62) *** (7.71) *** (4.93) *** Cultural -0.063 -0.055 -0.078 distance (1.27) (1.11) (1.58) RIA*GDP home (In) RIA*GDP host (In) RIA*Geographic distance (ln) RIA*Trade RIA*Labor cost RIA*Institutions RIA *Exchange rate RIA*Cultural distance constant 1.600 1.584 1.589 (19.41) *** (19.31) *** (19.40) *** Observations 3930 3930 3930 Time control No No No R-squared 61.74% 62.01% 62.07% Wald's Chi 2342.74 *** 2372.66 *** 2502.37 *** Square Model 4 Model 5 Model 6 governance legal regulatory quality GDP home (ln) 0.457 0.455 0.439 (8.34) *** (8.43)*** (8.01) *** GDP host (ln) 0.336 0.355 0.365 (4.54) *** (4.74)*** (4.96) *** Geographic -0.617 -0.614 -0.629 distance (ln) (6.59) *** (6.53)*** (6.65) *** trade (ln) 0.474 0.478 0.461 (10.26) *** (10.64)*** (10.11) *** RIA 0.412 0.396 0.345 (4.42) *** (4.08)*** (3.05) *** Labor cost -0.296 -0.176 -0.395 (1.21) (0.83) (1.82) * Institutions 0.089 -0.009 0.616 (1.24) (0.19) (3.27) *** Exchange rate -0.056 -0.052 -0.06 (In) (3.05) *** (2.91)*** (3.31) *** NAFTA 1.146 0.959 1.107 (4.20) *** (3.85)*** (4.46) *** ANZCERTA 1.412 1.461 1.340 (6.17) *** (6.38)*** (5.83) *** EU 1992 0.227 0.248 0.174 (3.05) *** (3.35)*** (2.34) ** Cultural -0.092 -0.091 -0.100 distance (1.91) * (1.88)* (2.05) ** RIA*GDP home -0.115 -0.13 -0.118 (In) (2.18) ** (2.51)** (2.23) ** RIA*GDP host -0.115 -0.087 -0.112 (In) (1.86) * -1.20 (1.90)* RIA*Geographic -0.224 -0.217 -0.228 distance (ln) (3.17) *** (3.03)*** (3.26)*** RIA*Trade 0.166 0.183 0.171 (3.08) *** (3.50)*** (3.17)*** RIA*Labor cost 0.616 0.739 0.473 (2.45) ** (3.46)*** (2.47)** RIA*Institutions 0.102 0.136 0.462 (1.69) * (1.30) (1.26) RIA *Exchange -0.091 -0.092 -0.092 rate (4.59) *** (4.55)*** (4.68) *** RIA*Cultural 0.091 0.106 0.093 distance (2.90) *** (3.27)*** (3.01) *** constant 1.695 1.670 1.664 (20.82) *** (19.72) *** (19.62) *** Observations 3930 3930 3930 Time control No No No R-squared 65.50% 65.34% 65.23% Wald's Chi 2758.65 *** 2745.04 *** 2798.12 *** Square Absolute value of z statistics in parentheses * significant at 10%; ** significant at 5%; *** significant at 1% Table 4: Random effects model regression results. The dependent variable is the log of the foreign direct investment inflows Model 1 Model 2 Model 3 governance legal regulatory quality GDP home (ln) 0.665 0.622 0.616 (16.20) *** (15.33) *** (15.33) *** GDP host (ln) 0.347 0.466 0.470 (7.67) *** (10.46) *** (10.89)*** Geographic -0.197 -0.224 -0.302 distance (In) (3.35) *** (3.88)*** (5.22)*** Intra-RIA 0.713 0.710 0.587 (7.43) *** (7.48)*** (6.22)*** Trade (In) 0.412 0.408 0.346 (10.45) *** (10.71)*** (9.16)*** Labor cost -0.08 -0.28 -0.519 (0.52) (2.13) ** (3.98) *** Institutions 0.077 0.297 1.734 (1.57) (11.98) *** (20.53) *** NAFTA -0.745 -0.741 -0.566 (3.02) *** (3.03) *** (2.36) ** ANZCERTA 0.008 0.424 0.652 (0.01) (0.48) (0.73) EU 1992 0.277 0.211 0.047 (6.25) *** (4.87) *** (1.07) Exchange rate (ln) -0.027 -0.034 -0.06 (1.93) * (2.49) ** (4.36) *** Cultural distance -0.131 -0.14 -0.171 (3.55) *** (3.88) *** (4.73) *** Intra-RIA*GDP host (In) Intra-RIA*Labor cost Intra- RIA*Institutions Constant 1.122 1.001 0.998 (17.74) *** (16.21) *** (16.40) *** Observations 7680 7680 7680 Time control No No No R-squared 52.63% 54.84% 55.08% Wald's Chi Square 3927.08 *** 4147.65 *** 4538.52 *** Model 4 Model 5 Model 6 governance legal regulatory quality GDP home (ln) 0.653 0.617 0.614 (15.95) *** (15.24) *** (15.33) *** GDP host (ln) 0.338 0.441 0.451 (7.46) *** (9.83) *** (10.44) *** Geographic -0.175 -0.205 -0.279 distance (In) (3.00) *** (3.56) *** (4.84) *** Intra-RIA 0.604 0.581 0.434 (5.46) *** (5.51) *** (4.31) *** Trade (In) 0.432 0.426 0.362 (10.96) *** (11.17) *** (9.59) *** Labor cost -0.19 -0.33 -0.536 (1.21) (2.48) ** (4.07) *** Institutions 0.087 0.267 1.580 (1.75) * (10.14) *** (17.49) *** NAFTA -0.757 -0.884 -0.634 (2.60) *** (3.40) *** (2.47) ** ANZCERTA 0.120 0.429 0.628 (0.13) (0.48) (0.70) EU 1992 0.241 0.186 0.021 (5.42) *** (4.26) *** (0.47) Exchange rate (ln) -0.03 -0.036 -0.057 (2.17) ** (2.60) *** (4.16) *** Cultural distance -0.116 -0.126 -0.157 (3.17) *** (3.51) *** (4.35) *** Intra-RIA*GDP 0.175 0.194 0.152 host (In) (3.38) *** (3.66) *** (3.01) *** Intra-RIA*Labor 0.771 0.636 0.387 cost (2.96) *** (3.15) *** (1.99) *** Intra- 0.135 0.102 0.602 RIA*Institutions -1.38 (2.04) ** (3.69) *** Constant 1.112 1.016 1.014 (17.58) *** (16.39) *** (16.68)*** Observations 7680 7680 7680 Time control No No No R-squared 52.62% 54.73% 55.18% Wald's Chi Square 4005.14 *** 4211.12 *** 4607.57 *** Absolute value of z statistics in parentheses * significant at 10%; ** significant at 5%; *** significant at 1% Table 5: List of host and home countries Host countries Regionally Creation Members integrated area European Union (EU) 1958 Austria (1995), Denmark, Finland (1995), France, Germany, Greece (1981), Ireland, Italy, Netherlands, Portugal (1986), Spain (1986), Sweden (1995), United Kingdom North American Free 1989 Canada, Mexico (1994), United States Trade Agreement (NAFTA) Australia-New 1983 Australia, New Zealand Zealand Closer Economic Relations (ANCERTA) None Iceland, Japan, Norway, South Korea, Switzerland, TurkeyHome countries Algeria, Argentina, Australia, Austria, Brazil, Bulgaria, Canada, Chile, China, Colombia, Costa Rica, Czech Republic, Denmark, Egypt, Finland, France, Germany, Greece, Hong Kong, Hungary, Iceland, India, Indonesia, Iran, Ireland, Israel, Italy, Japan, Kuwait, Libya, Malaysia, Mexico, Morocco, Netherland Antilles, Netherlands, New Zealand, Norway, Panama, Philippines, Poland, Portugal, Romania, Russia, Saudi Arabia, Singapore, Slovak Republic, Slovenia, South Africa, South Korea, Spain, Sweden, Switzerland, Thailand, Turkey, Ukraine, United Arab Emirates, United Kingdom, United States, Venezuela
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||RESEARCH ARTICLE|
|Author:||Feils, Dorothee J.; Rahman, Manzur|
|Publication:||Management International Review|
|Date:||Jan 1, 2011|
|Previous Article:||Entrepreneurship in emerging markets: strategies for new venture creation in uncertain institutional contexts.|
|Next Article:||Time to success in offshoring business processes: a multi level analysis.|