Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Varieties of Capitalism and institutional comparative advantage: A test and reinterpretation

Varieties of Capitalism and institutional comparative advantage: A test and reinterpretation Michael A. Witt and How do national-level institutions relate to national comparative advantages? Gregory Jackson We seek to shed light on this question by exploring two different sets of hypotheses based on the Varieties of Capitalism and other branches of INSEAD, 1 Ayer Rajah Avenue, comparative capitalisms literature. Applying fuzzy-set qualitative comparative Singapore 138676, Singapore; Free University analysis to data from 14 industries in 22 countries across 9 years, we find that of Berlin, Berlin, Germany comparative advantages in industries with radical innovation emerge in specific configurations mixing coordinated and liberal institutional features. Correspondence: Michael A. Witt, INSEAD, 1 Ayer Rajah Institutional comparative advantage in industries with radical innovation may Avenue, Singapore 138676, Singapore. thus be based on the ‘‘beneficial constraints’’ of opposing institutional logics Tel: +65-6799-5388; rather than on the self-reinforcing institutional coherence envisioned in much Fax: +65-6799-5399; of the Varieties of Capitalism literature. By contrast, we find that coordinated e-mail: [email protected] market economies may have comparative advantages in industries with incremental innovation, as envisioned in the Varieties of Capitalism literature. Our article contributes to our understanding of the ‘‘so what?’’ related to capitalist diversity and its implications for location decisions of multinational enterprises. We further present a coordination index going beyond Hall and Gingerich (Br J Polit Sci 39:449–482, 2009) with annual values for 22 OECD countries from 1995 through 2003. Journal of International Business Studies (2016) 47, 778–806. doi:10.1057/s41267-016-0001-8 Keywords: Varieties of Capitalism; economic sociology; comparative advantage; insti- tutional context; fuzzy-set methods The online version of this article is available Open Access INTRODUCTION What explains national comparative advantages? Almost 200 years after the publication of David Ricardo’s On the Principles of Political Economy and Taxation in 1817, the question has not conclusively been answered. Traditional economics explanations, including Ricardo’s and the later Heckscher–Ohlin model of trade, have emphasized the importance of inherited natural endowments with production factors such as labor and land. Empirical performance of these models, however, is weak (Trefler & Zhu, 2000). Received: 7 November 2014 More recent alternative explanations are based on the diverse Revised: 18 April 2016 institutional characteristics of national economies. Comparative Accepted: 18 May 2016 studies have argued that different types of institutions constrain Online publication date: 22 July 2016 VOC and Comparative Advantage Michael A. Witt and Gregory Jackson and enable different forms of economic activity of action, rather than coherence. We hypothesize (Jackson & Deeg, 2008). Most prominent among that certain combinations of liberal market and these has been the Varieties of Capitalism frame- coordinated logics across two or more institutional work (Hall & Soskice, 2001), which contrasted domains may enable institutional comparative ‘‘liberal market economies’’ and ‘‘coordinated mar- advantage by compensating for institutional weak- ket economies.’’ Meanwhile, other theories of nesses inherent in ‘‘pure’’ configurations. For exam- comparative capitalism categorize countries in dif- ple, liberal corporate governance may provide an ferent ways based on diverse governance modes important external monitoring of strongly coordi- (Crouch, 2005; Crouch & Streeck, 1997) or the nated and otherwise insider-oriented governance concepts of National Business Systems (Whitley, institutions (Aoki, 2010). This view posits the 1999). The common objective of this comparative potential for ‘‘beneficial constraints’’ based on capitalisms literature has been to shed light on how institutional arrangements with conflicting logics. the institutional diversity of advanced capitalist The article explores these ideas empirically by economies shapes economic and business comparing the trade patterns of countries during outcomes. the period of 1995–2003. In particular, we use A core argument of the literature is that institu- fuzzy-set qualitative comparative analysis (fsQCA) tions may generate distinct profiles of institutional to test whether different configurations of institu- comparative advantage in production, which mani- tions are sufficient for high performance in differ- fest themselves in nationally distinct patterns of ent sectors, characterized by either radical or economic performance and specialization across incremental forms of innovation. We find that different industrial sectors. For example, Schneider, pure CMEs have comparative advantages in indus- Schulze-Bentrop, & Paunescu (2010) explained rel- tries with incremental innovation. However, we do ative export performance in high-tech and med- not find that LMEs have comparative advantages in ium-tech industries based on different institutional industries with radical innovation. Rather, the characteristics of countries. Rather than a single results suggest that institutional comparative ‘‘best’’ set of institutional arrangements, different advantage involves very specific combinations of types of institutions give rise to distinct forms of both liberal and coordinated types of institutions, comparative advantage. Strong evidence in support which is in line with our own hypotheses. of institutional comparative advantage would have We conclude with a discussion of implications important implications for government policy as and limitations of this study. In particular, our well as the location choices of multinational findings suggest the need to distinguish conceptu- enterprises (MNEs) seeking to avoid home country ally between complementarity and coherence of disadvantages, project competitive advantages institutions and pay greater attention to how related to their home country, or exploit comple- tensions resulting from opposing institutional log- mentary resources and knowledge related to host ics can result in beneficial outcomes. By exploring country institutional environments (cf. Jackson & the linkage between institutional configurations Deeg, 2008; Singh, 2007; Witt & Lewin, 2007). and economic outcomes, our article contributes to In this article we explore the question of institu- our understanding of the ‘‘so what?’’ of variety in tional comparative advantage in relation to two capitalisms. Finally, our analysis shows that a arguments stemming from different branches of number of countries have, over time, diverged the comparative capitalisms literature. First, we test from the commonly held notions about their the well-known but widely contested hypothesis by institutional make-up. Germany, for instance, has Hall and Soskice (2001) that liberal market econo- evolved away from the pure-type coordinated mar- mies (LMEs) have an institutional comparative ket economy that it is commonly believed to advantage in industries featuring radical innova- represent. tion, whereas coordinated market economies (CMEs) have an institutional comparative advan- tage in industries with incremental innovation. LITERATURE REVIEW AND HYPOTHESES This argument posits strong complementarities among relatively coherent sets of institutions that Comparative Capitalisms follow similar logics across all institutional Renewed interest in institutions has led to a large and complex literature on comparative capitalisms domains of the economy. Second, we develop an focused on institutional diversity across national alternative view that is based on conflictual logics Journal of International Business Studies VOC and Comparative Advantage Michael A. Witt and Gregory Jackson contexts (Jackson & Deeg, 2008). Considerable 1999), no index is available. Some efforts have agreement exists that national diversity exists sought to overcome these shortcomings by drawing across several core institutional domains such as directly on the underlying institutional dimensions education and skills formation, employment rela- (e.g., Judge, Fainshmidt, & Brown III, 2014; Sch- tions, financial system, interfirm networks, internal neider & Paunescu, 2012). However, obtaining dynamics of the firm, ownership and corporate institutional measures with construct validity has governance, and the institutions of the state itself proved difficult, especially for countries outside the (Witt & Redding, 2013). Many studies explore how OECD (Witt & Redding, 2013). Together, these such differences cluster into distinct types of insti- obstacles have limited the wider application of tutional configurations. Several competing typolo- comparative institutional analysis in International gies exist. For instance, Hall and Soskice (2001) Business. A substantial research agenda remains to distinguished CMEs and LMEs as two distinct forms understand how institutions shape the comparative of capitalism. Amable (2003) identified five types: institutional advantage of different countries in the market-based, Asian, Continental European, social- world economy. democratic, and Mediterranean. Whitley (1999) As a step to overcoming this challenge, this paper describes six major types of business systems: examines the key arguments from the Varieties of fragmented, coordinated industrial district, com- Capitalism framework that link institutions to partmentalized, state-organized, collaborative, and comparative institutional advantage. Despite much highly coordinated. Looking beyond the advanced criticism (Allen, 2004; Blyth, 2003) and alternative industrialized countries, further types have been theoretical frameworks for comparing institutions proposed (cf. Witt, Kabbach de Castro, Amaeshi, (Amable, 2003; Crouch & Streeck, 1997; Whitley, Mahroum, Bohle, & Saez, 2015). 1999), the Varieties of Capitalism framework con- The challenge of creating theoretically rich and tinues to have a uniquely powerful hold on the valid typologies is complicated by institutional field. Critics have made various important claims change. While most comparative capitalism schol- about the categorization of countries and how this ars reject the possibility of institutional conver- changes over time: the parsimony of Varieties of gence (Hall & Soskice, 2001; Whitley 1999; Capitalism may overlook substantial institutional Yamamura & Streeck, 2003), many types of capi- variety within the LME and CME categories (Yama- talism have undergone substantial liberalization. mura & Streeck, 2003), emerging and transition Thus a growing literature now focuses on the economies cannot be understood using this mechanisms of institutional change (Mahoney & approach (No ¨ lke & Vliegenthart, 2009; Redding & Thelen, 2009; Streeck & Thelen, 2005) and the Witt, 2007), and its emphasis on complementari- diverging trajectories in different types of capital- ties and path dependence systematically overstates ism (Hall & Thelen, 2009; Streeck, 2008; Witt & institutional stability and underestimates institu- Lewin, 2007). tional change (Crouch, 2005). These criticisms Meanwhile, research linking institutional differ- notwithstanding, recent evidence supports the ences to specific economic outcomes has remained notion that advanced industrialized nations do surprisingly underdeveloped. First, a lack of con- indeed cluster very broadly into LMEs and CMEs sensus continues about relevant typologies. Conse- (Witt & Redding, 2013). More importantly, how- quently, scholars have continued to put energy ever, most critiques of Varieties of Capitalism do into criticizing specific typologies (Allen, 2004; not center on its core claims about institutional Blyth, 2003), validating typologies with empirical comparative advantage. data (Brewster, Wood, & Brookes, 2006) or extend- Next, we turn to the main hypothesis of the ing these to new geographic contexts (Bohle & Varieties of Capitalism approach and review empir- Greskovits, 2009; Carney, Gedajlovic, & Yang, ical literature that tests these claims. After this, we 2009). Second, suitable measures of institutional will explore and develop an alternative hypothesis diversity have limited availability. For instance, drawing on some alternative strands of the com- Varieties of Capitalism scholars have only produced parative capitalism literature. an index of institutional diversity for 22 OECD The Varieties of Capitalism Hypothesis: countries for a single time point, the mid-1990s Comparative Advantage Through Coherence (Hall & Gingerich, 2009). No time series is avail- In developing the Varieties of Capitalism approach, able, nor does the index cover countries outside the Hall and Soskice (2001) proposed that the advanced OECD. For the business systems approach (Whitley, Journal of International Business Studies VOC and Comparative Advantage Michael A. Witt and Gregory Jackson industrialized societies fall into two main types: manifest in comparative strengths and weaknesses LMEs and CMEs. Firms in LMEs tend to rely more in different types of industries. By contrast, coun- on liberal market mechanisms, while firms in CMEs tries with lower levels of coherence – ‘‘mixed types’’ tend to coordinate business transactions through – are argued to lack the institutional complemen- non-market relationships. The prototypical LME, tarities needed for institutional comparative such as the US, features a market-driven financial advantage. system, flexible use of external labor markets, Based on this logic, Hall and Soskice (2001) generalist education and training systems, low proposed that LMEs and CMEs would show distinct levels of networks and alliances among firms, and patterns of institutional comparative advantage for management-driven, top-down decision-making radical or incremental innovations. Hall and Sos- structures inside firms. The classic CME, such as kice (2001: 38f.) defined radical innovation as Germany around 1995, has a bank-led financial ‘‘entail[ing] substantial shifts in product lines, the system providing patient capital, stronger internal development of entirely new goods, or major labor markets based on employment protection, changes to the production process,’’ while incre- skills formation systems conducive to the develop- mental innovation is ‘‘marked by continuous but ment of specialized skills, high levels of networks small-scale improvements to existing product lines and alliances among firms, and consensual deci- and production processes.’’ They argued that the sion-making inside firms bringing together man- combination of patient capital, long-term employ- agement and labor. These institutional differences ment, and firm-specific skills in CMEs would enable encourage firms to invest in and utilize transferable more efficient production in industries with incre- assets to a greater extent in LMEs, where institu- mental patterns of innovation because the relative tions do not bind economic actors to long-term immobility of labor and capital in CMEs enabled commitments but support their ‘‘keeping options and constrained firms to focus their efforts on open’’ to using the external market. By contrast, the improving existing lines of products. Meanwhile, use of relational assets in CMEs implies investments fluid capital markets with short-term employment whose value is specific to the continuation of long- and general skills in LMEs would enable more term relationships among company stakeholders. efficient production in industries with radical pat- Such investments require different sorts of institu- terns of innovation, as these conditions support tional support – such as protection of investments firms using external markets to mobilize risky in firm-specific human capital and contracting equity finance and workers with different skill sets, arrangements, or mechanisms to govern collective and thereby take advantage of new technological action problems (e.g., free rider problems) and breakthroughs. Since comparative advantage is support wider patterns of cooperation across a defined as the ability to make one product more network or industry. efficiently than another, this suggests the presence The Varieties of Capitalism framework posits that of comparative advantage in industries drawing on LMEs and CMEs exhibit strong complementarities incremental innovation in CMEs and radical inno- by being organized around a coherent institutional vation in LMEs. By the Law of Comparative logic (Deeg, 2007). Complementarities exist where Advantage (Dearsdorff, 1980), this advantage features of institutional structures reinforce each should be visible in international trade patterns. other by mutually generating increasing returns. This leads to the following hypothesis: Meanwhile, coherence is present where institutions H1 LMEs show trade patterns consistent with follow the same market or coordinated logic across institutional comparative advantage in industries domains of the economy, such as corporate gover- featuring radical innovation, whereas CMEs show nance and employment relations. Since coherent trade patterns consistent with institutional com- institutions are hypothesized to uniformly support parative advantages in industries with incre- either transferrable or relational assets, the Varieties mental innovation. of Capitalism approach claims that institutions in different domains will complement one another by Prior Tests mutually reinforcing such investments. Based on Despite its centrality, some fifteen years after the their capacity to enable these different kinds of publication of Hall and Soskice’s piece, H1 remains investments, Varieties of Capitalism argues that open territory. Attempts to test it have faced several complementary sets of institutions favor certain interrelated conceptual and methodological patterns of economic activity over others that Journal of International Business Studies VOC and Comparative Advantage Michael A. Witt and Gregory Jackson challenges. First, mapping institutional diversity Paunescu (2012) used trade outcomes to claim onto the two LME and CME categories is less qualified support of the Hall and Soskice hypoth- straightforward than commonly assumed (see more esis. Both papers relied on exports as a measure of generally Jackson & Deeg, 2008). Different studies comparative advantage. However, as the case of use different sets of indicators, or benchmark electronics in China illustrates (e.g., Redding & countries against assumed ideal–typical countries Witt, 2009), high levels of exports in an industry like Germany for CMEs (Allen & Aldred, 2009; may occur on the back of high levels of imports of Schneider et al., 2010). Second, the concept of high value added inputs in the same industry, in radical and incremental innovation has proven which case only a small proportion of the value of slippery. Operationalizing radicality requires a valid the exports in these industries are produced in the way of classifying industries according to different respective country. These imports need taking into types of innovation. Most existing studies have account to establish whether a comparative advan- relied on categorizations supplied by the OECD to tage exists. The two papers also relied on OECD describe high-tech or medium high-tech industries. classifications of research intensity to express rad- This classification is based on R&D intensity, which icality. However, radicality is an output of research is a measure of investment or inputs into the activity, while research intensity is an input mea- innovation process rather the output of actual sure based on R&D spending relative to industry innovation. Third, empirical work has focused size. In sum, prior works have either sought to either on the influence of institutions on innova- explain innovation patterns rather than compara- tion patterns or the influence of institutions on tive advantage, or suffered from issues with con- export success, but missed what we see as the core struct validity of key variables. As will be explained of the Varieties of Capitalism hypothesis – namely, in the methodology section, this article will offer a the idea that the influence of institutions on trade more comprehensive test of H1 than the previous outcomes is mediated by or contingent upon types literature. of innovation. Cross-country differences exist in Alternative Hypothesis: Complementarities the patterns of radical or incremental innovation Through Beneficial Constraints (Furman, Porter, & Stern, 2002), and this capacity The Varieties of Capitalism approach has assumed does relate to institutional features of the economy, that complementarities require coherence of insti- such as employment relations (Bassanini & Ernst, tutions (e.g., pure LME types, where all institutions 2002). However, we see this as a separate debate. In are liberal). The underlying logic stems from the our view, the core of the Varieties of Capitalism concept of complementarities as developed in hypothesis actually revolves around comparative economics (Milgrom & Roberts, 1994, 1995), advantage in products as the dependent variable, whereby investments in transferable assets in one rather than R&D spending or patents. The Varieties institutional domain are complemented by trans- of Capitalism argument stresses the ability of firms ferable assets in other domains. For example, to draw on these innovations in the productive portable skills are expected to increase in value for process, not their ability to make radical innova- workers if firms also have corporate governance tions. Firm-level studies using both approaches arrangements whereby capital investment is shifted have found these to be potentially complementary, rapidly from one firm to another. Conversely, long- but in fact very distinct, phenomena (Herrmann & term relational employment relationships are pre- Peine, 2011). sumed to benefit from ‘‘patient capital’’ of coordi- To our knowledge, only four major published nated blockholders, whose long-term strategic studies have attempted to explicitly test the Vari- investment in the firm will protect the firm-specific eties of Capitalism hypothesis. Taylor (2004) and skill investments of workers. Akkermans, Castaldi, & Los (2009) examined inno- Based on this reasoning, Varieties of Capitalism vation patterns as visible in patent data. While implies a very restrictive view of how institutions Taylor found no support of the hypothesis, Akker- combine. Comparative advantage relates to the mans and colleagues claimed qualified support. coherence of incentives provided by the set of However, both papers tested only the ability of institutions within a country, and thus comple- economies to produce radically innovative patents mentarities only form when institutions cluster and not actual institutional comparative advan- into only two ideal–typical patterns. We believe tage, which is the outcome of interest in this paper. that these assumptions about complementarities Schneider et al. (2010) as well as Schneider and Journal of International Business Studies VOC and Comparative Advantage Michael A. Witt and Gregory Jackson are problematic and at odds with empirical work To elaborate this idea, we focus on the two most linking comparative advantage to more complex or salient stakeholders of the firm: shareholders and hybrid combinations of coordinated and liberal employees. Here we see a potential counterbalanc- institutions (Boyer, 2004). We thus draw on a ing role to be played by each of these groups. stream of the comparative capitalisms literature Shareholders may benefit from liberal corporate grounded in economic sociology to develop an governance and establish this as a counterweight to alternative set of hypotheses regarding comple- more coordinated institutions. Likewise, employ- mentarities based on opposing institutional logics. ees may favor more coordinated employment Our intuition here is to see complementarities as relations and benefit from coordination as a coun- essentially grounded within political projects. terweight to otherwise liberal institutions. These Crouch (2005) argued that institutions may have mixed configurations reflect a compromise or positive effects by compensating for the weaknesses balancing between opposing logics, whereby the of other institutions. For example, political systems interests of one key stakeholder are addressed often utilize competing institutional logics that within an overall context that is more hostile to serve as checks and balances on one another – such their interests. as when the rule of law in the judiciary acts to In developing these hypotheses, we turn to the balance the logic of majority rule in the legislature. concept of ‘‘beneficial constraints’’ in economic Along these lines, sociological theories of compar- sociology (Streeck, 1997). Streeck argues that ative capitalism have observed that opposing logics socially institutionalized constraints on the or principles of social organizations may be useful rational voluntarism associated with markets may complements (Crouch, Streeck, Boyer, Amable, be economically beneficial. Normative constraints Hall, & Jackson, 2005;Ho ¨ pner, 2005). Here the and social obligation are needed to correct market political economy of capitalism involves not simply failures, and thus also enable economic actors to a choice between coordination or liberal markets, develop and protect collective goods or commit- but reflects political coalitions and compromises ments to stakeholders that would otherwise fall that seek a balance between them. The interests victim to hyper-rational economizing. Rather than and relative powers of different actors groups, allowing economic actors to pursue their given particularly corporate stakeholders, come into play. preferences, institutions may transform the identi- Liberal markets produce winners and losers, also ties and interests of actors by constraining their generating counter-movements seeking to contain pursuit of short-term aims and leading them to the influence of markets through efforts of coordi- search for new and alternative strategies that are nation and social protection (Block & Somers, legitimate in their institutional context. In doing 2014). In terms of the CME and LME typology, this so, institutional constraints can turn institutions perspective implies that mixed cases or hybrids may into beneficial resources by simultaneously have beneficial characteristics. For example, Boyer enabling new forms of coordination and coopera- (2004) showed that Nordic economies achieved tion, thus potentially attaining higher levels of economic growth in emerging ITC sectors based on economic performance than might be possible in a specific combination of liberal markets and their absence. By constraining certain market coordinated institutions. However, whether or not behaviors, institutions may help create or protect mixed cases work well together depends on the critical resources based on coordinated investments extent to which the specific combination of market by stakeholders. As Streeck argued: ‘‘Beneficial and coordinated logics across two or more domains constraint…is a dialectical concept, suggesting a may help compensate for one another. Unfortu- relationship of both mutually subversive and mutu- nately, the comparative capitalism literature offers ally supportive conflict between the economic only scant theorizing about specific hybrid and the social, ruling out any lasting harmony combinations. between the two’’ (Streeck, 1997: 207). Importantly, Based on these considerations, we develop two this view of institutions goes beyond the notion hypotheses whereby coordinated institutions help of property rights or embeddedness in social net- to prevent or remedy market failures in liberal works that facilitate market exchange. Rather, settings or, following the inverse logic, liberal institutions act as constraints on markets, on market institutions help to remedy problems of the one hand, but serve to stabilize them, on the rigidity in highly coordinated settings. Here oppos- other – thus suggesting their Durkheimian charac- ing institutional logics complement one another. ter (Jackson & Muellenborn, 2012). Journal of International Business Studies VOC and Comparative Advantage Michael A. Witt and Gregory Jackson Viewed from this perspective, the coherent set of institutions that regulate pay at sectoral or even market-oriented institutions characterizing LMEs national level. Likewise, Japanese firms utilize may have inherently self-destructive tendencies. As strong norms of lifetime employment for their core markets across different domains of the economy workforce while limiting external labor markets to reinforce one another by creating incentives for a very narrow range of specialists or to temporary greater market-orientation, these markets may employees. While economists usually see these become overheated or hyper-rational. Time restrictions as politically motivated, by the logic horizons may become too short and market failures of beneficial constraints, they may be economically may occur in relation to long-term financial invest- beneficial by providing an institutional basis on ments, development of cooperative supplier which companies can develop and retain necessary relationships, or fostering of firm-specific skills in resources. productive capabilities (Campbell, 2011; Jackson & Rather than complementarities based on coher- Petraki, 2010). Positively, firms in LMEs may ence, we expect that firms facing radical innovation respond quickly to radical innovations by shifting in countries with liberal market institutions in their investments in markets for capital and labor. most domains may benefit from complementary However, left unconstrained, firms may also be institutions that support coordination in employ- likely to focus on predatory responses, such as ment relations. A combination of liberal and coor- using mergers and acquisitions to obtain new dinated institutions may be beneficial for dealing technologies or protect market shares or aggres- with radical innovation by providing for both sively restructure corporate hierarchies by shedding flexibility in restructuring of economic organiza- employees and boosting shareholder returns. As a tion as well as trust and coordination to solve result, firms may be unable to capitalize on radical problems of asymmetric information or hold-up. innovations by developing related firm-specific Applying the same logic as Hall and Soskice (2001) advantages that lead to sustained comparative linking these benefits to comparative advantage, we advantages, as suggested by the resource-based thus posit the following: view and other theories (Newbert, 2007). H2 Countries with coordinated institutions in These deficits of purely liberal markets are likely employment relations and liberal institutions in to be most acute in the area of employment other domains will show trade patterns consis- relations. Employee skills and organizational capa- tent with comparative advantage in industries bilities based upon them are essential for compar- with radical innovation. ative advantage during times of discontinuous change (Aoki & Jackson, 2008; Rajan & Zingales, Beneficial constraints also extend to a second 2000). In particular, firms entering and expanding argument. Similar to the inherent problems with new lines of business as a result of radical innova- unconstrained market-oriented transactions, high tion will need skilled employees with sufficient levels of coordination may lead to an overextension understanding of the workings of the firm (i.e., of relational logics. To the extent that coordination firm-specific skills) to capitalize on the opportuni- becomes strongly self-reinforcing, firms may lack ties offered by new technology. While active exter- incentives to take risks and enter into new lines of nal labor markets may help firms gain access to new business – employment is long-term, firm-specific skills, left fully to the logic of the market, firms are skills are well developed, workers participate in firm also likely to suffer from substantial problems to decision-making, inter-firm relations rely on stable partners, and owners take a long-term strate- retain key employees through periods of dynamic change. Economic sociology has thus suggested gic view. These patterns may be beneficial in that markets need to be embedded within social stable industry environments characterized by low relations, not only for moral or political reasons, levels of innovation or even incremental patterns, but for economic ones. Institutional constraints as hypothesized by Varieties of Capitalism. How- ever, it seems unlikely to these institutional imposed by coordination may be beneficial for markets precisely because they curtail the oppor- arrangements can support adjustments to more tunism and hyper-rationality that lock actors into radical and discontinuous forms of innovation potential market failures due to collective action (Witt, 2006). problems. Many countries in Europe thus restrict The potential deficits of high coordination may create particular problems for shareholders, who competition between firms based on wages through Journal of International Business Studies VOC and Comparative Advantage Michael A. Witt and Gregory Jackson are disadvantaged by low competitiveness. Conse- METHODOLOGY, DATA AND CALIBRATION quently, we see more liberal forms of corporate Methodology: fsQCA governance as having potential to exert a counter- We used fsQCA in Stata (StataCorp, 2013)as balancing effect on high coordination and thus to provided by Longest and Vaisey (2008). Qualitative represent an economically beneficial compromise comparative analysis draws on Boolean algebra to between competing logics. Here firms benefit from determine which configurations of causal condi- high coordination across different domains, includ- tions are related to an observed outcome (Ragin, ing employment relations, and develop strong 2000). The method is increasingly used in the capabilities based on their human assets and rela- business and management literature (e.g., Bell, tional linkages to other firms. However, a literature Filatotchev, & Aguilera, 2013; Crilly, 2011; Fiss, on hybrid forms of corporate governance suggests 2011; Judge et al., 2015; Schneider et al., 2010). that increasing pressures from capital markets may We adopted fsQCA because of its distinct advan- lead stakeholder-oriented firms to take bolder and tages for studying institutional complementarities faster adjustments to changes in technology and (Jackson & Ni, 2013). First, fsQCA is well suited to product markets (Vitols, 2002, 2004). For example, capture conjunctural types of causation that under- Aoki (2010) argues that external monitoring of lie theories of complementarity. Complementari- internal linkages of corporate organization may be ties are based on specific configurations where beneficial in giving corporate insiders important multiple causal factors combine to produce an signals about their overall business strategy and outcome. Traditional statistical methods model help benchmark their efforts against competitors. ‘‘net effects’’ of single variables, while holding other Unlike monitoring by ‘‘patient’’ investors, a more factors constant. While interaction effects can liberal market-oriented process may be a beneficial capture specific conjunctions, such as radical inno- resource for investors seeking to constrain the vation with institutional coordination, these meth- tendencies toward stagnation among corporate ods are not well suited to more complex insiders, such as employees and managers. While conjunctions. Consequently, most work on com- commitments to corporate stakeholders remains parative capitalism has used factor analysis to strong, some pressure from capital markets and simplify institutional configurations into highly influence from owners may help to catalyze more simplified one-dimensional constructs (Hall & Gin- dynamic, albeit ‘‘negotiated’’ responses to radical gerich, 2009) or group countries into a limited innovations through more discontinuous manage- number of types (Amable, 2003), at the risk of ment strategies or changes in production technolo- losing important distinctions in the process. fsQCA gies. Indeed, a growing literature on the relation does not require such simplifying steps. Second, between corporate governance and employment fsQCA can identify how multiple different combi- patterns suggest more complex relationships nations of independent variables can produce the between these variables than found in the Varieties same outcome (equifinality). Third, fsQCA evalu- of Capitalism literature (Gospel & Pendleton, ates set theoretical relationships, rather than corre- 2003, 2005;Ho ¨ pner, 2005). Despite their potential lations among different factors. This under- to destabilize commitments to stakeholders, liberal appreciated facet allows for the possibility of market institutions in corporate governance may asymmetrical relationships, whereby high and low also potentially help counterbalance risk averse values of the outcomes are driven by different tendencies of corporate insiders, thus paradoxically causal conditions. For example, if low coordination helping to catalyze their dynamic potential in and radicality are jointly sufficient for high perfor- terms of functional flexibility and ability to foster mance outcomes, it does not follow that coordina- trust, but here in the service of adjustments to tion is sufficient for outcomes of low performance. radical innovation. Applying the same logic about Finally, the method is not sensitive to outliers the linkage to comparative advantage and trade, we because it does not assume an underlying proba- thus posit: bility distribution (Fiss, 2011). H3 Countries with liberal institutions in corpo- rate governance and coordinated institutions in Sample and Data Structure other domains will show trade patterns consis- Our dataset consists of 2772 observations, made up tent with comparative advantage in industries of 14 industries at the 2-digit ISIC-3 level in 22 with radical innovation. OECD countries, with annual observations from Journal of International Business Studies VOC and Comparative Advantage Michael A. Witt and Gregory Jackson 1995 through 2003. Each case is a particular to other industries. This implies a need to compare industry for a given country and year. The choice within countries, as our measure does, rather than of countries, industries, and years was affected by across countries, as the Balassa index does. For data availability as explained below. example, assume that country A has two industries, X and Y, and is better at producing goods in these industries than any other country in the world. Outcome Measure: Contribution to the Trade This would give it a competitive (absolute) advan- Balance Our outcome measure was the contribution to the tage relative to other countries, which the Balassa trade balance for manufacturing industries. It is index would pick up as an advantage in both industries. However, if country A is better at based on the ‘‘Law of Comparative Advantage’’ in producing X than at Y, we can say that it has a economics (Dearsdorff, 1980), which states that comparative advantage at X over Y. The Balassa comparative advantages in manufacturing should be revealed in trade patterns. Much of the eco- index would miss this, while our preferred measure nomics literature has measured revealed compara- would pick this up. For robustness testing, we will later check our results against those of the Balassa tive advantage using the Balassa (1965) index, index. which compares a country’s export market share In choosing our industries, we omitted low in a given industry with the average export market share of all countries. We followed an alternative technology industries, such as food and beverages, calculation used by the OECD (2003) that we textiles, and wood and cork. Low technology industries are defined by the lowest research activ- believe to have higher construct validity, as we will ity, which implies a weak link between innovation explain below. and resultant comparative advantage. Our analy- For each country, the contribution to the trade balance (CTB in the formula) in the OECD concep- sis thus focused on low-medium technology, tualization and as used in this article is computed as medium–high technology, and high technology sectors. Since the contribution to the trade balance follows: is likely to manifest itself with a delay after an ðÞ X þ M i i invention occurred, our analysis used a lead-time of CTB ¼ðÞ X  MðÞ X  M ; ð1Þ i i ðÞ X þ M one year relative to the causal conditions (i.e., the contribution to the trade balance in year 2004 will where (X - M ) is the observed trade balance in i i be analyzed with respect to institutional factors for industry i in the given country and the remainder 2003). for the formula is the theoretical trade balance for this same industry i in the same country (OECD, Causal Conditions: Indicators of Radicality 2003: 150). To make the results comparable across and Institutional Coordination countries, they are expressed a percentage of total trade for the respective country. A positive value Indicators of Radicality denotes the presence of a comparative advantage, a Following Akkermans et al. (2009), we calculated negative value, of a comparative disadvantage for three distinct, patent-based measures to gauge the that country in a particular industry (OECD, 2013). extent of radical innovation in a given society. In We obtained these data from the OECD (2013). line with the prior literature, incremental innova- This measure improves over the Balassa index in tion is operationalized as the inverse of radical two important ways. First, it takes into account the innovation – innovations that are not radical are level of imports in a given industry. While the assumed to be incremental. While we will later Balassa index focuses entirely on exports and thus question this assumption, we retained it since no cannot distinguish whether a country has a high established measure of incremental innovation export market share because it produces much of exists in the literature. the value-added itself (a sign of comparative advan- In our first step, we obtained all entries from tage) or because it imports and re-exports (as in the 1990 through 2005 of the NBER Patent-Citations case of China already discussed), our measure can Data File (Hall, Bessen, & Thoma, 2011). This distinguish between these two cases. Second, the database contains the patent number and three Balassa index is a measure of competitive advantage indicators of radicality for each US patent issued rather than comparative advantage. Comparative during this period: number of citations received, advantage exists within each country with respect generality, and originality. The number of citations Journal of International Business Studies VOC and Comparative Advantage Michael A. Witt and Gregory Jackson received operationalizes the concept of radicality contents of the NBER database with the PATSIC- by measuring the impact on future technological CONAME database using patent numbers, which development, as expressed in subsequent citations are contained in both databases. by other patents. Generality refers to the propensity We calculated the relative proportions of radi- of basic innovation to diffuse to many different cally innovative patents by industry and country industries and areas of technology, that is, their for each of the three measures of radicality. Similar versatility. Originality expresses the idea that truly to Akkermans et al. (2009), we define a patent to be radical innovations are likely to cite patents from radically innovative if it scores higher than the different areas of technology. Further explanations 95th percentile for a given measure. For instance, if of these measures are contained in the works of the 95th percentile for citations received in the Hall, Jaffe, & Trajtenberg (2001) and Akkermans pharmaceuticals industry was 15, we would count et al. (2009). as radical all patents with more than 15 citations For all three measures, the innovation literature received. We consequently calculated the threshold has interpreted higher numbers to indicate higher separately for each year because the year of publi- levels of radicality (Akkermans et al., 2009). We cation of a patent affects how many citations it has believe that of the three measures, the number of received, with younger patents on average receiv- citations received is likely to be closest to a valid ing fewer citations than older ones. indicator of radicality. Similar to citations to In the final step, we calculated the proportion of scholarly articles, citations to patents indicate their patents above the threshold in each country, impact. Just as highly cited articles are relatively industry and year. To generate a more stable picture, more likely to be in some way seminal than less we assigned to each observation the three-year cited pieces, patents with higher citations counts average of these proportions from t - 2 years to t. are likely to have a greater impact on the course of We discarded data from 2004 and 2005 because the technology development and thus represent radical database ends in 2005, which means that there was steps in technological development. We are less insufficient time for citations from these two years convinced of the construct validity of the other two to receive sufficient numbers of citations for dis- measures. Breakthrough ideas in one area do not criminating between radically and incrementally necessarily have to be of interest to other areas, nor innovative patents. Given the strong interest in do they need to assemble insights from many innovation patterns in the Varieties of Capitalism different preexisting areas. Despite our doubts, we debate, Table 1 shows the cross-tabulation of rad- retained both generality and originality for robust- icality, using the citations received indicator, by ness tests. country and industry. The raw data required two further steps for our Institutional Indicators purposes. First, the generality and originality mea- The starting point in understanding institutions sures are defined as Herfindahl indexes and there- was past work categorizing countries into either fore not given for patents that received no citations LME or CME types of capitalism. However, different or contained no citations. In these cases, we studies based classifications on different institu- assigned a value of 0 for each measure. Second, tional domains and have made different assump- both measures are biased downward for patents tions about how institutions combine into these involving small numbers of citations (Hall, Jaffe, & broader types. For example, Hall and Gingerich Trajtenberg, 2001). We corrected for this bias as (2009) used confirmatory factor analysis to derive a proposed in Hall, Jaffe, & Trajtenberg (2001). coordination index for 20 advanced industrialized Since our outcome measure is categorized by economies based on three corporate governance industries following the ISIC 3 classification and three employment relations variables. How- scheme, we next matched each patent with one ever, this approach covered only two of the five or several ISIC codes. The United States Patent and institutional domains discussed by Varieties of Trademark Office (USPTO) assigns each patent to Capitalism (Hall & Soskice, 2001). In addition, the one or several industries to which the patent is construction of a single index assumes that com- likely to be relevant. While the original assignment plementarities are achieved through cohesion (all scheme is incompatible with ISIC classifications, domains being either fully coordinated or fully the USPTO identifies the SIC codes for each patent liberal), an assumption this paper seeks to relax and in its PATSIC-CONAME database. We obtained the test (H2, H3). year 2008 version of the database and matched the Journal of International Business Studies VOC and Comparative Advantage Michael A. Witt and Gregory Jackson Table 1 Proportion of radical patents by citations received, by country and industry, 1995–2003 Industry AT AU BE CA DK FI FR DE GR IE IT Aircraft, spacecraft 0.012 0.051 0.056 0.064 0.127 0.000 0.007 0.043 0.000 0.000 0.031 Basic metals 0.000 0.094 0.019 0.027 0.000 0.000 0.034 0.036 0.000 0.056 0.058 Chemicals other than pharmaceuticals 0.020 0.037 0.020 0.046 0.027 0.032 0.028 0.030 0.000 0.105 0.018 Electrical machinery 0.007 0.046 0.034 0.095 0.008 0.034 0.026 0.020 0.000 0.016 0.019 Machinery 0.017 0.036 0.028 0.053 0.015 0.023 0.017 0.022 0.000 0.042 0.012 Medical, precision, optical equipment 0.007 0.042 0.009 0.052 0.043 0.015 0.019 0.018 0.000 0.099 0.008 Metal products 0.039 0.038 0.038 0.066 0.025 0.016 0.030 0.029 0.000 0.000 0.024 Motor vehicles 0.041 0.018 0.028 0.084 0.028 0.019 0.024 0.037 0.000 0.053 0.020 Non-metallic products 0.006 0.079 0.016 0.067 0.023 0.026 0.030 0.023 0.000 0.000 0.018 Office, accounting, computing equipment 0.000 0.051 0.026 0.065 0.000 0.026 0.011 0.010 0.000 0.018 0.006 Other transport equipment 0.015 0.037 0.000 0.041 0.000 0.000 0.024 0.060 0.000 0.000 0.048 Pharmaceuticals 0.018 0.022 0.018 0.043 0.023 0.037 0.025 0.021 0.000 0.100 0.013 Radio, TV, communication equipment 0.008 0.048 0.023 0.072 0.002 0.066 0.019 0.015 0.000 0.044 0.004 Rubber, plastic 0.036 0.038 0.024 0.059 0.031 0.026 0.033 0.027 0.000 0.035 0.027 Average 0.016 0.046 0.024 0.060 0.025 0.023 0.023 0.028 0.000 0.041 0.022 Industry JP NL NO NZ PO KR ES SE CH UK US Aircraft, spacecraft 0.073 0.026 0.000 0.000 0.000 0.008 0.000 0.021 0.002 0.015 0.065 Basic metals 0.038 0.073 0.000 0.000 0.000 0.010 0.111 0.000 0.037 0.055 0.081 Chemicals other than pharmaceuticals 0.031 0.027 0.032 0.033 0.037 0.025 0.035 0.052 0.041 0.033 0.072 Electrical machinery 0.038 0.020 0.021 0.082 0.000 0.021 0.022 0.021 0.015 0.035 0.076 Machinery 0.057 0.023 0.015 0.009 0.000 0.025 0.018 0.018 0.014 0.029 0.077 Medical, precision, optical equipment 0.017 0.033 0.024 0.005 0.000 0.010 0.000 0.022 0.029 0.025 0.074 Metal products 0.054 0.038 0.006 0.000 0.000 0.034 0.024 0.031 0.022 0.035 0.077 Motor vehicles 0.049 0.014 0.065 0.000 0.000 0.010 0.016 0.032 0.060 0.022 0.078 Non-metallic products 0.042 0.052 0.000 0.000 0.000 0.033 0.009 0.063 0.024 0.026 0.080 Office, accounting, computing equipment 0.011 0.021 0.139 0.037 0.000 0.008 0.037 0.038 0.005 0.027 0.077 Other transport equipment 0.076 0.000 0.000 0.000 0.000 0.004 0.000 0.007 0.000 0.031 0.071 Pharmaceuticals 0.011 0.055 0.093 0.028 0.000 0.019 0.025 0.057 0.052 0.045 0.074 Radio, TV, communication equipment 0.026 0.015 0.058 0.019 0.000 0.021 0.022 0.058 0.014 0.035 0.074 Rubber, plastic 0.039 0.036 0.005 0.000 0.000 0.033 0.034 0.063 0.034 0.030 0.080 Average 0.040 0.031 0.033 0.015 0.003 0.019 0.025 0.034 0.025 0.032 0.075 We thus considered all five domains: corporate original indicator derived by LLSV (La Porta, Lopez- governance, inter-firm relations, hierarchies within de-Silanes, & Shleifer, 1999) and an updated mea- firms, employment relations, and education. For sure for 2003 (Botero, Djankov, La Porta, Lopez-de- each institutional domain, we conceptualized of Silanes, & Shleifer, 2004), interpolating annual institutions as varying along a single dimension values based on the movement between these two spanning between liberal institutions characterized periods. Shareholder protection indicates the by a high use of market relationships or transferable degree to which business firms are likely to have a assets and coordinated institutions characterized by liberal orientation toward shareholder control. high use of long-term strategic relationships or Second, we obtained stock market valuation as a relationship-specific assets. We adopted a separate percentage of GDP from OECD statistics to capture indicator for each and later will explore the differ- the salience of the stock market for company ent ways in which they combine into a number of financing. Third, we measured the dispersion of complex configurations. share ownership by the percentage of large firms in In terms of corporate governance, we used three each country that have a blockholder with an indicators. First, an index of legal rules for share- ownership stake of 10 % or greater. We used the holder protection measures whether laws in a LLSV indicator of ownership dispersion based on country adopt particular protections related to the percentage of large firms without an owner disclosure, voting rights, and so forth (La Porta, holding a 10 % stake or more, and calculated our Lopez-de-Silanes, Schleifer, & Vishney, 1998). Since own data for 2003 using ownership data from the no alternative suitable measures exist, we took the ownership module of Thomson Banker One. We Journal of International Business Studies VOC and Comparative Advantage Michael A. Witt and Gregory Jackson interpolated annual data based on the year-to-year Thus we examined only the proportion of employ- changes between these two dates. High blockhold- ees with short job tenure of less than one year. ing is associated with coordination. While all countries have some new hires, a high For inter-firm relationships, we used two indica- proportion of new hires indicate a more liberal tors based on merger and acquisition activity from market-oriented to employment practices, whereas the Capital IQ database. First, the number of a low proportion indicates that employment may merger and acquisition deals by acquiring firms in be more long-term and coordination. Second, we each country indicates the use of the market in examined the strictness of employment protection governing relationships between firms. While we for regular employees based on well-known OECD lack internationally comparable measures of coor- indices. Greater difficulty in firing workers indicates dination such as that achieved through long-term again a higher degree of coordination in employ- relational contracts in buyer–supplier relations, ment relations. Finally, we measured the degree of prior literature (Schneider & Paunescu, 2012) sug- coordination in wage bargaining across firms, gested that mergers and acquisitions indicate that industries and national levels. This indicator shows firms seek to obtain assets from target firms the degree to which firms engage in individualized through a market-driven acquisition of ownership market transactions in setting wages, or whether rights. Second, the proportion of merger and these are coordinated with other employers in the acquisition deals that take the form of a full-scale same industry or even nationally (Visser, 2011). merger indicates the use of strong market relation- Finally, we used two indicators to examine the ships. Mergers and acquisitions may involve arms’ characteristics of education and training institu- length transactions, where control is purchased and tions. We drew on OECD statistics on the propor- leads to the dissolution of a target firm, or acqui- tion of graduates from different types of sitions of minority ownership stakes. In the latter educational institutions within the wider popula- case, the target firms remain legally independent tion. First, we took the share of graduates from but enter into a coordinated long-term relationship upper secondary or post-secondary non-tertiary as with the acquiring firm. a broad indicator for occupationally-based voca- For firm hierarchies, we examined the degree of tional training (Schneider et al., 2010). Second, we employee participation at the level of corporate used university training, as measured by the boards and through works councils and similar number of graduates from university as a share labor-management consultation bodies. Board- of the population in the typical graduation age level employee representation has been categorized range, as an indicator of general training (Schnei- based the percentage of employees on the board der et al., 2010). Taken together, these indicators and on whether codetermination requirements show the importance of two pathways toward apply broadly to all large private companies, are skilled workforces based either on more occupa- restricted to public corporations, or do not exist at tionally based and thus relationship-specific train- all (Jackson, 2005). Strong legal rights to employee ing in particular occupations, which reflect higher representation on boards indicate long-term rela- degrees of coordination, or on university-based tional coordination of firm operations between education that offers more general set of skills that management and employees. In addition, many tend to be portable across firms or very broad countries anchor employee participation at the range of jobs according to individual career level of the establishment or even workplace in trajectories. the form of works councils. We measured work Table 2 presents an overview of the descriptive council rights by the rights to information, consul- statistics for all indicators used in the study. tation or even codetermination of managerial Calibration decisions, as well as whether these rights pertain fsQCA requires calibration of all data into set to social or also economic matters of the enterprise membership scores that fall between 0 (full absence (Visser, 2011). of the given causal condition) to 1 (full presence) In terms of employment relations, we used three (Ragin, 2008). Calibration combines information indicators. First, we considered the duration of about qualitative differences in kind (e.g., countries tenure with a particular employer. Unfortunately, with or without employee participation in firm existing data from the OECD do not permit detailed hierarchies) with information about differences in cross-national comparisons of average job tenure or degree (e.g., different degrees of participation). the proportion of employees with very high tenure. Journal of International Business Studies VOC and Comparative Advantage Michael A. Witt and Gregory Jackson Table 2 Descriptive statistics Indicator N Mean SD Min Max 1 Shareholder protection 198 3.24 1.1 0 5 2 Dispersion of control 198 0.33 0.24 0 0.9 3 Size of stock market 198 78.5 54.19 8.92 317.03 4 Mergers and acquisitions 198 0.24 0.27 0 1.1 5 Mergers 198 0.38 0.19 0 1 6 Wage coordination degree 198 0.18 0.06 0.08 0.41 7 Short-term employment 198 3.07 1.25 1 5 8 Employment protection 198 2.06 0.86 0.17 4.33 9 Board-level codetermination 198 0.43 0.39 0 1 10 Works council rights 198 1.19 1.1 0 3 11 Occupational training 198 43.89 19.25 2 82 12 University training 198 29.5 9.7 9 53.3 13 Radicality, citations received 198 0.03 0.04 0 0.42 16 Radicality, generality 198 0.04 0.04 0 0.48 17 Radicality, originality 198 0.07 0.06 0 0.67 14 Contribution to the trade balance 198 -0.12 1.72 -7.33 10.9 15 Balassa index 198 0.92 0.73 0 6.32 Ragin (2008) stresses using external benchmarks set of countries with coordinated corporate gover- rather than sample variance in defining thresholds nance was based on three indicators: the absence of for set membership. Rather than assuming that dispersed ownership and either weak stock markets being ‘‘above average’’ or ‘‘below average’’ on a or weak shareholder rights. Before combining with particular indicator is sufficient for having high or the ownership condition, we first took the maxi- low membership in a particular set, the extent of mum score for membership in the set of countries membership depends on theoretical criteria and with an absence of stock market activity or weak substantive knowledge of the cases at hand. While shareholder rights. We treat these as functionally we adopt this qualitative approach to data calibra- equivalents for coordination, since the absence of tion, the results of our calibration scoring correlate either condition means that managers are unlikely at levels of 95 % or above with scores derived from to feel strong pressures from shareholders. Turning more automated coding procedures such as using a to employment relations, high coordination is pre- standardized rank ordering of cases. sent in countries with coordinated wage bargaining Table 3 presents a summary of all data sources, and either long employee tenures or strong the calibrations of each institutional indicator and employment protection. Again, the outcomes of construction of higher-order constructs used in the high tenure or legal protection of regular workers empirical analysis. are treated as functionally equivalent indicators of We combined the indicators into five higher- long-term employment relations. order constructs, one for each domain, taking the Table 4 presents a summary overview of our minimum score among the selected indicators for calibrated measures for each country, showing each country (on set operators logical AND and OR, whether they fall into more coordinated or more see Ragin, 2008), whereby higher membership were liberal ‘‘types’’ for each institutional domain based coded to indicate greater coordination. Firm hier- on the average scores across the period 1995–2003. archy is the membership in the set of countries with In our analysis, we did not use these averages but both high coordination for board codetermination yearly indicators, since many countries actually and works councils. Inter-firm relations are highly changed configurations during this time period. coordinated in countries with both low mergers To calibrate set membership for radicality, we and acquisitions per capita and where full mergers considered membership in the set of industries are not the predominant form of mergers and with high radicality based on the proportion of acquisitions. Education is highly coordinated in patents that are highly cited. Similar to the contri- countries that have high occupational training bution to the trade balance measure, we calibrated and low university education. Membership in the radicality in terms of the concentration of radical Journal of International Business Studies VOC and Comparative Advantage Michael A. Witt and Gregory Jackson Journal of International Business Studies Table 3 Data sources and calibration of measures for institutional coordination Coordination in Indicator Explanation of the measure Anchor points for Years Sources institutional calibration of fuzzy-set available domain membership Corporate Membership in set of countries with low dispersion and either low shareholder protection or low stock market size (all items reverse coded) governance Shareholder Legal rights of shareholders (0–6 scale) Fully-out: 1.5 1995, 2003 Botero et al. (2004) and La Porta et al. protection Cross-over: 2.5 (1999) Fully-in: 5 Dispersion of Proportion of largest 20 firms by market Fully-out: 0 1995, 2003 La Porta et al. (1999) and own control capitalization without a blockholder (10 % stake Cross-over: 0.4 calculations from Thomson Banker One or greater) Fully-in: 0.8 database Size of stock Stock market capitalization as a percentage of Fully-out: 10 Continuous OECD statistics market GDP Cross-over: 60 Fully-in: 110 Inter-firm Membership in set of countries with low M&A and low mergers (all items reverse coded sets) relations Mergers and Number of mergers and acquisitions by domestic Fully-out: 0 Continuous S&P Capital IQ acquisitions acquiring firms divided by GDP (billion USD) Cross-over: 0.25 (M&A) Fully-in: 0.5 Mergers Proportion of M&A transactions by domestic Fully-out: 0 Continuous S&P Capital IQ acquiring firms involving full-scale merger of Cross-over: 0.4 target firm Fully-in: 0.8 Employment Membership in set of countries with high wage coordination and either low short-term employment (reverse coded) or high employment protection relations Wage 5 = economy-wide Fully-out: 1 Continuous Jelle Visser, http://www.uva-aias.net/ coordination 4 = industry and economy-wide Cross-over: 3.5 208 degree 3 = industry Fully-in: 5 2 = industry and firm level 1 = mostly at company level Short-term Proportion of employment with less than 1 year Fully-out: 0.1 Continuous OECD Statistics employment of tenure Cross-over: 0.15 Fully-in: 0.25 Employment Strictness of legal rules concerning dismissal (0–6 Fully-out: 0 Continuous OECD Statistics protection scale) Cross-over: 1.5 Fully-in: 3 Firm hierarchy Membership in set of countries with high codetermination and high works council rights Board-level Legal rights for employee representation on Fully-out: 0 2003 Jackson (2005) codetermination corporate board Cross-over: 0.5 1 = parity representation Fully-in: 1 0.8 = 1/3 representation 0.6 = information rights 0.4 = representation in public companies only 0.2 = constitutional protection 0 = no representation VOC and Comparative Advantage Michael A. Witt and Gregory Jackson innovations in particular industries relative to the mean level of radical innovations for each country. We considered citations in the bottom 25 % and top 75 % to be fully out or fully in the set of highly cited patents, whereas zero difference with the country mean would indicate a crossover point between comparative disadvantage or advantage for that industry relative to overall country performance. While this indicators looks at relative advantages of countries across indus- tries, the results of our fsQCA analysis proved identical to an alternative calibration that directly compared the proportion of radical patents in each industry between different countries. In order to further check the robustness of this radicality measure, we applied the same procedure to mea- sures of radicality based on generality or original- ity dimensions, and then constructed a higher- order set membership scores for radical innovation based on citations received, originality or generality. Finally, our outcome condition of contribution to the trade balance was calibrated based on -2as being fully out, 2 as being fully in, and zero as the cross-over point. Substantively, the contribution to the trade balance score of zero offers an intuitive crossover point, since this score repre- sents no comparative advantage or disadvantage. The anchors for no and full membership are slightly greater than two standard deviations around the crossover point. This calibration cor- relates highly with the rank ordering of cases, and our results proved robust to alternative anchor points. Data Analysis Our analysis used fsQCA to evaluate which of the 2 possible combinations of causal conditions (where 6 represents 5 institutional conditions plus indus- try-level radicality) are sufficient conditions for the outcome of high contribution to the trade balance. In evaluating the data, our solutions needed to meet several thresholds (e.g., Fiss, 2011; Judge et al., 2015; Ragin, 2008). First, we accepted only solutions with a consistency score (a measure of reliability) that showed a statistically significant level above the common benchmark of 0.80. Second, we excluded any solutions that were simultaneously consistent with the negation of the outcome. This problem of simultaneous subsets is possible where set membership in causal condi- tions is very low (Schneider & Wagemann, 2012). Third, valid solutions required a threshold of 40 Journal of International Business Studies Table 3 (Continued) Coordination in Indicator Explanation of the measure Anchor points for calibration of Years Sources institutional domain fuzzy-set membership available Works council 3 = economic and social rights, including codetermination on Fully-out: 0 Continuous Jelle Visser, rights some issues Cross-over: 1.5 http://www. 2 = economic and social rights, consultation (advice) only Fully-in: 3 uva-aias.net/ 1 = social rights 208 0 = no rights (sanctions), only information. Education and Membership in set of countries with high occupational training and low university training training Occupational Graduates from upper secondary or post-secondary non-tertiary Fully-out: 20 Continuous OECD statistics training training as percentage of graduation age population Cross-over: 40 Fully-in: 60 University Graduates from tertiary education as percentage of graduation age Fully-out: 10 Continuous OECD statistics training population Cross-over: 25 Fully-in: 40 VOC and Comparative Advantage Michael A. Witt and Gregory Jackson Table 4 Set membership of countries in highly coordinated institutions, 1995–2003 average Note: Set membership scores 0.5–1 indicate medium to full membership in the category of high coordinated, whereas scores 0 to 0.5 indicate zero to medium membership (shaded cells) in the set of coordinated countries and thereby displaying liberal institutional features. cases. Standard applications of fsQCA are usually assumptions and interpret the results of the fsQCA small-N, with a common threshold of two cases complex solutions only. (e.g., Fiss, 2011). Since we are using fsQCA in a large N setting, we chose a stricter threshold on the consideration that any valid solution would have to RESULTS contain observations from about two years or two Table 5 shows the results of our analysis. It suggests industries per country. Solutions with fewer cases three pathways to high contribution to the trade than 40 were treated as being inconsistent. We balance, with a coverage of 0.482 (i.e., they jointly confirmed robustness of this choice by increasing explain 48.2 % of membership in our dataset). and decreasing the threshold. Finally, our analysis Consistency levels (similar to statistical signifi- does not contain any strictly directional hypothe- cance) are all high throughout at levels of 0.914 ses linking institutional factors to comparative and higher, which is considerably better than the advantage outcomes, since our hypotheses suggest commonly used minimum threshold of 0.800. that both liberal and coordinated institutions may Solution 1 combines low radicality with high lead to comparative advantage under certain con- coordination across all institutional spheres. This ditions. Thus while we identify the ‘‘core condi- corresponds to incremental innovation producing tions’’ as provided by the parsimonious solutions a comparative advantage in CMEs and thus offers (cf. Fiss, 2011), we do not postulate any ‘‘easy partial support of H1. Austria follows this CME counterfactuals’’ to be used as simplifying pattern, achieving comparative advantage across a Journal of International Business Studies VOC and Comparative Advantage Michael A. Witt and Gregory Jackson Table 5 Sufficient conditions for high revealed comparative advantage Solution 1 Solution 2 Solution 3 Radicality Corporate governance coordination Employment relations coordination Education and skills coordination Firm hierarchy coordination Interfirm coordination Raw coverage 0.170 0.376 0.177 Unique coverage 0.057 0.248 0.026 Consistency 0.939 0.928 0.949 Total coverage, all solutions 0.482 Solution consistency, all solutions 0.914 Note: = presence of condition, = absence of condition, blank = absence or presence does not matter; large symbols show core conditions (parsimonious solution). range of industries such as office equipment, comparative advantage in motor vehicles, chemicals, aircraft, and metals. Notably, Austrian industries pharmaceuticals, machinery and metal products. Dur- experienced only incremental forms of innovation, ing the 1990s, Germany underwent substantial liber- while radical innovation was essentially absent. alization of its corporate governance arrangements, However, this result only gives partial support for despite relative continuity of its coordinated employ- H1, as the second pattern predicted by H1, high ment relations and strong occupational training insti- radicality with low coordination across all institu- tutions (Jackson, 2003). During this period, changes in tional spheres, is absent from the set of solutions. corporate governance exerted substantial pressure on Solution 2 combines radicality with low coordina- organized labor to cooperate in firm restructuring and tion in all institutional spheres except employment support changes in business models. This, we argue, relations. Interestingly, inter-firm relations are irrele- ultimately served here as a type of ‘‘beneficial con- vant to this solution and may be either liberal or straint’’ on coordinated patterns of business organiza- coordinated. This result is consistent with H2 and tion in the sense of balancing the logic of coordination describes Japan (e.g., motor vehicles, aircraft), Switzer- with a stronger market-orientation. The importance of land during the late 1990s (e.g., chemicals), and this ‘‘opposing’’ institutional logic is clear if we com- Ireland (e.g., chemicals, medical equipment sectors) pare Germany to its neighboring Austria, which as key cases. While Japan is often seen as an exemplar represents a pure CME-type of economy and excelled of CMEs (e.g., Witt, 2006; Yamamura & Streeck, 2003), only in incrementally innovative industries with a albeit with unusual characteristics, our measures notable absence of success with radical innovation. present a different picture. Japan is characterized by We conducted a number of robustness tests. First, a fairly liberal set of institutions – education is focused we used a more stringent calibration of the dependent more on university graduates with general skills, variable measuring contribution to the trade balance employee voice in firm hierarchies is not strongly by changing the anchor points for low and high rooted in law, and after substantial changes to corpo- contribution to the trade balance from -2to -4and rate governance during the 1990s, shareholder rights ?2to ?4, respectively. This produced precisely the and patterns of corporate ownership became fairly same three solutions as our main model. Coverage liberal. Nonetheless, Japanese firms sustained ‘‘micro- dropped minimally from 0.485 to 0.482, while con- corporatist’’ arrangements within the domain of sistency improved from 0.914 to 0.933. Second, we employment relations based on strong employment used the Balassa index instead of contribution to the protection and low labor turnover. Unlike more pure trade balance as a dependent variable. This produced a LMEeconomies,stableemploymentis coreto the single solution that is identical with Solution 3 in our success of Japanese manufacturing, such as Toyota. main model(describing thecaseofGermany). Cover- Solution 3 combines high radicality with high age was 0.194, consistency, 0.925. This lends support coordination in all institutional spheres except for to H3 only. At the same time, given the issues with corporate governance, which is consistent with H3. construct validity of the Balassa index as noted earlier, Germany falls into this category and achieved high we are not very confident that this support, or the lack Journal of International Business Studies VOC and Comparative Advantage Michael A. Witt and Gregory Jackson Table 6 Sufficient conditions for low revealed comparative advantage Solution 1 Solution 2 Solution 3 Solution 4 Radicality Corporate governance coordination Employment relations coordination Education and skills coordination Firm hierarchy coordination Interfirm coordination Raw coverage 0.345 0.199 0.283 0.408 Unique coverage 0.008 0.048 0.120 0.042 Consistency 0.880 0.921 0.914 0.854 Total coverage, all solutions 0.595 Solution consistency, all solutions 0.844 Note: = presence of condition, = absence of condition, blank = absence or presence does not matter; large symbols show core conditions (parsimonious solution). of support of the other hypotheses, is particularly relations and education. This configuration is consis- meaningful. Third, we used a more comprehensive tently associated with low contribution to trade measure of radicality. The three measures of radicality balance in industries if radical innovation is absent. discussed earlier – number of patents received, origi- Unlike pure CMEs, coordination is absent in certain nality, and generality – represent functional substi- domains such as employment relations, being rather tutes in that all three have been described as potential focused on a more liberal flexicurity employment expressions of radicality in innovation. Following model. At the same time, coordinated corporate Ragin’s (2008) approach for functional substitutes, we governance does not act as a beneficial constraint vis- combined all three variables using an OR operator – a-vis market-oriented employment patterns, since that is, the highest value of any of these three variables concentrated share ownership is not sufficient in itself became the radicality measure for each given data to induce investments in firm-specific skills and point. This model produced two solutions, which are capacities needed for incremental innovation. In equivalent to Solutions 2 and 3 in our main model. addition, Solution 3 showsaconfiguration forlow Joint coverage was 0.362, and all consistency scores contribution to the trade balance describing Greece were above 0.908. This provides additional support of and Portugal, as well as some years for New Zealand H2 and H3, but no support of H1. and Spain. This configuration shows that radicality is Overall, these results offer strong support of H3 associated with poor performance where coordinated and, given our doubts about the Balassa index, H2. institutions for corporate governance and inter-firm Partial support of H1 is confined to our main model relations are mixed with liberal patterns for firm and thus seems more tentative. hierarchies and education. Despite the mix of coordi- Unlike in correlational methods, solutions consis- nated and liberal elements, the configuration lacks tent with high and low contribution to the trade complementarities based on beneficial constraints. balance are not necessarily symmetrical in fsQCA Again, coordinated corporate governance, such as (Schneider & Wagemann, 2012). The additional results highly concentrated ownership patterns, or weak in Table 6 are helpful in interpreting our findings for M&A activity in inter-firm relations do not act as a the main model. Solutions 1 and 4 show that low constraining counterbalance to liberal market-ori- contribution to the trade balance is consistent with ented patterns of education and employment. As countries with liberal institutions across 3 or 4 insti- such, these particular ‘‘mixed market economies’’ fail tutional domains and the absence of radical innova- to obtain comparative advantage in radical industries tion. This finding is essentially a mirror of Solution 1 in (Hancke ´, Rhodes, & Thatcher, 2007). Table 5, but shows that countries approximating LME- type configurations do not achieve comparative advantage in industries with low radicality. Solution DISCUSSION In this article we used fsQCA to examine how 2 describes several Nordic cases, such as Norway, country-level institutional configurations relate to Denmark and Finland that have coordinated corporate governance and firm hierarchies with liberal inter-firm institutional comparative advantage across Journal of International Business Studies VOC and Comparative Advantage Michael A. Witt and Gregory Jackson different industrial sectors characterized by either theorize how complementarities may alternatively more radical or incremental patterns of innovation. be based on the combination of conflicting insti- Our main finding is that radical innovation con- tutional logics. Here complementarities may result sistently leads to high comparative advantage in from opposing logics, in which liberal and coordi- countries with institutions that combine specific nated institutions compensate for one another. liberal and coordinated elements: Radical innova- Specifically, coordinated employment relations tion leads to comparative advantage in economies institutions may constrain the influence of liberal with predominantly liberal institutions but coordi- markets, thereby enabling firms to make and nated employment relations and, conversely, in protect investments in core human resources. economies with predominantly coordinated insti- Meanwhile, in highly coordinated economies, lib- tutions but liberal corporate governance. We inter- eral corporate governance may act as an external pret these ‘‘mixed’’ or hybrid combinations of monitor that constrains the behavior of company liberal and coordinated institutions as having stakeholders, but in ways that also enable more complementarities derived from beneficial con- rapid and successful adaptation to radical innova- straints, whereby an institutional logic of either tion. Nonetheless, our analysis shows clearly that market or coordinated exchange is counter-bal- not all hybrid combinations of markets and coor- anced by an opposing institutional logic. dination achieve these sorts of beneficial con- One implication of our results relates to the straints. In mixed market economies, such as validity of the Varieties of Capitalism hypothesis Greece of Portugal, neither does liberal corporate (H1). While we find some evidence that CMEs have governance act as an external monitor for highly an institutional comparative advantage in indus- coordinated patterns of organization, nor does tries with incremental innovation, the data show coordinated employment relations serve as a con- no indication that LMEs have an institutional straint of liberal market organization. Of course, comparative advantage in industries with radical our analysis leaves a number of cases unexplained, innovation. Additional regression analysis, which suggests a research agenda to better map and reported in the Appendix, also shows no clear test hybrid institutional logics based on beneficial support of H1. This lack of strong and consistent constraints. support based on state-of-the-art measures and Complementarities based on opposing principles methodology suggests that the Varieties of Capital- of social organization have a further and poten- ism framework needs substantial reinterpretation tially wide-ranging implication: that complemen- and further theoretical development. tarities should be understood as political projects, At the same time, our results are broadly consis- not equilibrium outcomes. For example, as Ger- tent with earlier studies linking performance in many faced economic turmoil in the early 1990s, certain sectors with institutional configurations the political process of liberalization triggered a that go beyond the LME and CME dichotomy. For process where more coordinated employment rela- example, Boyer (2004) explored what institutional tions were defended and adapted to new con- configurations are consistent with a technology-led straints of more liberal corporate governance. growth regime based on information and commu- While this institutional dynamism may have ener- nications technology. He identified three different gized the economy, one consequence has been configurations, none consistent with Hall and increasing inequality and the political temptation Soskice’s (2001) notion of complementarities to liberalize institutions further, which may undo through coherence. Similarly, Schneider et al. the success (Streeck, 2009). Thus new hybrid con- (2010) linked hybrid economies with mixed insti- figurations of institutions are no more stable equi- tutional logics to strong export performance in libria than more coherent configurations; rather, high technology industries. Our article lends addi- institutions change dynamically all the time. The tional support to the argument that certain types of Celtic Tiger or Danish flexicurity are other good hybrids may be more successful than the pure types examples. These ‘‘success models’’ of one era even- envisioned by Hall and Soskice (2001). tually run their course, often because the winners The key contribution here is to go beyond these and losers of changing institutional arrangements previous studies by offering an ex ante prediction of push for further processes of institutional change the kinds of results we expect to see, drawing on the (Thelen, 2014). As countries travel past certain concept of beneficial constraints. This approach, windows, they may gain comparative advantages, which builds on Streeck (1997), allowed us to but these combinations of institutional Journal of International Business Studies VOC and Comparative Advantage Michael A. Witt and Gregory Jackson arrangements may eventually exhaust themselves the potential benefits of using fsQCA to elucidate as other countries and innovation patterns change the conjunctural effects of different combinations in unison (Boyer, 2004). Here a key finding of our of institutional indicators (Ragin, 2008). While study is that the competitive advantage of Ger- previous studies have applied fsQCA methods to many was based not simply on its presumed the Varieties of Capitalism literature, our results are character as an ideal–typical CME country; rather, broadly consistent with past results that have it did better than pure CMEs like Austria by linked more complex institutional configurations incorporating hybrid institutional logics. Similarly, that go beyond the simple LME and CME types we can go beyond other country stereotypes in the with stronger economic growth or export perfor- literature to compare other unexpected pairs, such mance (Kogut & Ragin, 2006; Schneider et al., as Ireland and Japan. Rather than being opposed as 2010). The multidimensionality evident in our LME and CME types, we can further investigate the analysis calls for international business scholars to parallel ways in which these countries combine pay greater attention to differences in kinds of liberal institutions with high coordination in institutions and their combination, rather than employment relations. conceptualizing of institutions as one-dimensional Our findings also add to the still relatively sparse differences in degrees, as done widely in literatures literature linking the comparative capitalism liter- on institutional distance, institutional develop- ature to economic outcomes. There is little doubt ment or even Varieties of Capitalism (Jackson & that there are institutional differences, but the ‘‘so Deeg, 2008). It also suggests that benefits usually what?’’ – whether and where these variations associated with social structures may be contingent matter – has been much less explored. Our study on the presence of other, supportive institutional finds that institutional comparative advantage may structures. For instance, while our analysis gener- indeed exist, though possibly not in the manner ally supports the widely accepted notion that previously envisaged. Importantly, this suggests interorganizational networks may produce benefi- adopting research designs that allow for causal cial outcomes (e.g., Dyer & Singh, 1998; Sako, complexity and equifinality, whereby different 1992), Solution 3 in Table 6 suggests that in some institutional structures can lead to similar out- configurations, they may be associated with weak comes. The original Varieties of Capitalism performance. approach (Hall & Soskice, 2001) suggested equifi- Our findings may further have important impli- nality between CMEs and LMEs with respect to cations for understanding the location choices of overall economic wealth, but posited a more simple multinational enterprises (MNEs) seeking to avoid relationship between LMEs and the ability to home country disadvantages, project competitive leverage radical innovation. By contrast, our results advantages related to their home country, or suggest that at least two different institutional exploit complementary resources and knowledge configurations are associated with comparative related to host country institutional environments advantage in the presence of radical innovation. (cf. Jackson & Deeg, 2008; Singh, 2007; Witt & Furthermore, these configurations are dramatically Lewin, 2007). The literature has previously identi- different: either all elements but corporate gover- fied host country institutions as sources of loca- nance are coordinated, or all but employment tional advantages (Dunning and Lundan 2008), relations are liberal. which suggests the possibility of institutional arbi- Our conceptualization of complementarities trage (Jackson & Deeg, 2008; Witt & Lewin, 2007): based on opposing logics and the empirical results Firms may move their operations, in part or in on equifinality challenge some existing conceptu- whole, to institutional contexts that better support alizations of institutional differences in Interna- these operations. The Varieties of Capitalism liter- tional Business. For example, recent works have ature would have suggested that firms should locate already noted that the impact of distances may not activities involving radical innovations in LMEs be symmetric or even indicative of difficulty oper- such as the United States. Our results suggest a ating abroad (Cuervo-Cazurra & Genc, 2012). In different picture: Production involving radical this context, we hope that our approach can help innovation seems to thrive in economies pairing provide a basis for future empirical work on the either liberal corporate governance with coordina- economic impact of Varieties of Capitalism in tion in the other spheres of the political economy, particular and of institutional differences in Inter- or coordinated employment relations with formally national Business more generally. Our study shows liberal institutions in other areas. These patterns are Journal of International Business Studies VOC and Comparative Advantage Michael A. Witt and Gregory Jackson far more complex than connections between insti- In terms of radicality, we faced the same problem tutions and locational choices of MNEs previously that has plagued all prior attempts at exploring the explored in the literature, such as political risk or hypothesis: how to measure it. Our operationaliza- corruption (Henisz, 2002; Rodriguez, Uhlenbruck, tion follows procedures used and accepted by the & Eden, 2005). At the same time, it is possible that specialist literature on R&D. However, patent data experience and evolutionary processes have shaped and even patent citations have well-known limita- MNE investment patterns accordingly. To the tions. Not all innovations can be patented (e.g., extent this is the case, one would expect that MNEs new software), and firms do not necessarily patent relying more on radical innovations locate rela- their innovations for cost or strategic reasons tively more of their activities in these economies. (Archibugi & Planta, 1996). Moreover, radical One would also expect that the performance of innovation may also be associated with process MNEs relying more on radical innovations should innovations driven by changes in organization for correspond to the extent to which they locate their which new technology is a necessary but not a operations in economies with suitable institutional sufficient condition. Our results analyze the impact configurations. Exploring these hypotheses would of radicality across a wide range of industries, but seem to be a fruitful avenue for future IB research. different dynamics may occur and different insti- tutional features may be salient within specific Limitations industries. As such, our article does not exhaust this Though our empirical findings are robust to a large agenda. number of alternative specifications, they have The notion of radicality can also be questioned at important methodological limitations. a deeper level. It remains unclear whether radical First, our results may be limited to the specific time innovation is properly understood at the level of frame and country sample. Our study is based on a patents, meaning proprietary inventions, or 9-year window from 1995 through 2003, for reasons whether it entails shifts in broader technological of data availability as explained earlier. This period paradigms independent of particular inventions or covers the time period for which the Varieties of applications. For example, major shifts in informa- Capitalism approach claimed validity (cf. Hall & tion technology are not centered on single radical Gingerich, 2009) and constitutes a strong test of the inventions, but may imply a radical shift in busi- original hypothesis. Nonetheless, as newer data ness models or the locus of competitive advantage. become available, future studies might examine These broader phenomena, like the information whether the hypothesized comparative advantages technology revolution, may give firms in LMEs new manifest themselves post-2003. For example, the full opportunities to leverage institutions for fast entry effects of the Internet revolution, shifts in biotech- and exit from lines of business, or to use external nology or expansion of global value chains may be labor markets to retool their human capital and clearer over a longer time span. Moreover, while our organizational capabilities. But we cannot discount sample covers all countries discussed by Hall and the idea that firms in CMEs will leverage their Soskice, and while Varieties of Capitalism theory institutions to make incremental innovations focuses on advanced industrialized economies, it is around this same context, whereby information possible that a larger range of countries might lead to technologies are integrated successfully into estab- more comprehensive insights. lished product lines in ways leading to quality or A second limitation concerns the construct valid- process improvements. Put differently, comparative ity of the coordination and radicality measures. advantage may be reconceived as being less cen- Concerning the former, Hall and Soskice discussed tered on advantages related to a fixed boundary of the notion of transferable versus relational assets an industry or technology. Rather, innovation may across the five institutional domains of corporate be refracted through the lens of institutions in ways governance and finance, employment relations, that are leveraged for more radical or incremental education, managerial hierarchies within firms, shifts in business strategy and organization. Here and inter-firm relations. However, little consensus multinational firms with capabilities for institu- exists about which institutional indicators should tional bricolage may have distinctive advantages. be used to best capture ‘‘coordination’’ in each Indeed, research on Central and Eastern Europe has domain. While we believe that our variables max- started to find competitive advantage in high-tech imized construct validity to the extent possible, industries based on strong FDI with complex insti- better measures may become available. tutional hybrids that involve combinations of Journal of International Business Studies VOC and Comparative Advantage Michael A. Witt and Gregory Jackson liberal elements with either strong employee par- NOTES ticipation or longer-term employment relations Porter’s (1990) work on the competiveness of (Allen & Aldred, 2011). As such, we think the nations has been an important influence in the agenda of understanding how coordinated rela- development of these new approaches. We exclude tions act as beneficial constraints on markets has a it from the discussion because Porter’s focus is on long future ahead. competitiveness and thus on absolute advantage rather than on comparative advantage. On the notion of institutions as resources more generally, see also Hall and Thelen (2009) and Jackson ACKNOWLEDGMENTS and Deeg (2008). We thank Jaeyong Song for his constructive editorial This original index has also important empirical guidance and the three anonymous reviewers for their limitations, please refer to the Appendix. insightful feedback. We are further grateful for com- Generally, raising thresholds for full membership ments and input received from Robert Boyer, Daniel produced very similar results, whereas lowering W. Gingerich, Bronwyn H. Hall, Sebastian Koos, and thresholds to just one standard deviation produced Gerald A. McDermott. Rashid Ansari at INSEAD and fewer consistent results, as the resultant calibration no So¨ren Salzwedel at FU Berlin provided valuable longer distinguishes between cases of high and very research assistance, and we thank the INSEAD libraries high contributions to the trade balance. for their support. Michael Witt also thanks the Hum- boldt Foundation for funding two 3-month research stays in Berlin to undertake this project. REFERENCES Akkermans, D., Castaldi, C., & Los, B. 2009. Do ‘liberal market Blyth, M. 2003. Same as it never was: Temporality and typology economies’ really innovate more radically than ‘coordinated in the varieties of capitalism. Comparative European Politics, market economies’? Hall and Soskice reconsidered. Research 1(2): 215–225. Policy, 38(1): 181–191. Bohle, D., & Greskovits, B. 2009. Varieties of capitalism and Allen, M. 2004. The varieties of capitalism paradigm: Not capitalism ‘‘tout court’’. European Journal of Sociology, 50(3): enough variety? Socio-Economic Review, 2(1): 87–108. 355–386. Allen, M. M. C., & Aldred, M. L. 2009. Varieties of capitalism, Botero, J. C., Djankov, S., La Porta, R., Lopez-de-Silanes, F., & varieties of innovation? A comparison of old and new EU Shleifer, A. 2004. The regulation of labor. Quarterly Journal of member states. Journal of Contemporary European Research, Economics, 19(4): 1339–1382. 5(4): 581–596. Boyer, R. 2004. New growth regimes, but still institutional Allen, M. M. C., & Aldred, M. L. 2011. Varieties of capitalism, diversity. Socio-Economic Review, 2(1): 1–32. governance, and high-tech export performancea fuzzy-set Brewster, C., Wood, G., & Brookes, M. 2006. Varieties of analysis of the new EU member states. Employee Relations, capitalism and varieties of firm. In P. James, & G. Wood (Eds), 33(4): 334–355. Institutions, production, and working life. Oxford: Oxford Amable, B. 2003. The diversity of modern capitalism. Oxford: University Press. Oxford University Press. Campbell, J. L. 2011. The US financial crisis: Lessons for theories of Aoki, M. 2010. Corporations in evolving diversity: Cognition, institutional complementarity. Socio-Economic Review, 9(2): governance, and institutions. Oxford: Oxford University Press. 211–234. Aoki, M., & Jackson, G. 2008. Understanding an emergent Carney, M., Gedajlovic, E., & Yang, X. 2009. Varieties of Asian diversity of corporate governance and organizational archi- capitalism: Toward an institutional theory of Asian enterprise. tecture: An essentiality-based analysis. Industrial and Corporate Asia Pacific Journal of Management, 26(3): 361–380. Change, 17(1): 1–27. Crilly, D. 2011. Predicting stakeholder orientation in the Archibugi, D., & Planta, M. 1996. Measuring technological multinational enterprise: A mid-range theory. Journal of change through patents and innovation surveys. Technova- International Business Studies, 42(5): 694–717. tion, 16(9): 451–519. Crouch, C. 2005. Capitalist diversity and change. Recombinant Balassa, B. 1965. Trade liberalisation and ‘‘revealed’’ compara- governance and institutional entrepreneurs. Oxford: Oxford tive advantage. The Manchester School, 33(2): 99–123. University Press. Bassanini, A., & Ernst, E. 2002. Labour market regulation, Crouch, C., & Streeck, W. 1997. Political economy of modern industrial relations and technological regimes: A tale of capitalism: Mapping convergence and diversity. London: Sage. comparative advantage. Industrial and Corporate Change, Crouch, C., Streeck, W., Boyer, R., Amable, B., Hall, P. A., & 11(3): 391–426. Jackson, G. 2005. Dialogue on ‘institutional complementar- Bell, G., Filatotchev, I., & Aguilera, R. 2013. Corporate gover- ity and political economy. Socio-Economic Review, 2(4): nance and investors’ perceptions of foreign ipo value: An 359–382. institutional perspective. Academy of Management Journal, Cuervo-Cazurra, A., & Genc, M. E. 2012. Categories of distance 57(1): 301–320. and international business. In G. Wood, & M. Demirbag (Eds), Block, F. L., & Somers, M. R. 2014. The power of market Handbook of institutional approaches to international business: fundamentalism: Karl Polanyi’s critique. Cambridge, MA: Har- 219–235. Cheltenham: Edward Elgar. vard University Press. Journal of International Business Studies VOC and Comparative Advantage Michael A. Witt and Gregory Jackson Dearsdorff, A. V. 1980. The general validity of the law of and methods in organizational research: 129–158. Bingley: comparative advantage. Journal of Political Economy, 88(5): Emerald Group Publishing. 941–957. Jackson, G., & Petraki, A. 2010. Understanding short-termism: Deeg, R. 2007. Institutional complementarity and change in The role of corporate governance. Stockholm: Glasshouse capitalist systems. Journal of European Public Policy, 14(4): Forum. 612–623. Judge, W. Q., Fainshmidt, S., & Brown, J. L, I. I. I. 2014. Which Dunning, J. H., & Lundan, S. M. 2008. Institutions and the oli model of capitalism best delivers both wealth and equality? paradigm of the multinational enterprise. Asia Pacific Journal of Journal of International Business Studies, 45(4): 363–386. Management, 25(4): 573–593. Judge, W. Q., et al. 2015. Configurations of capacity for change Dyer, J. H., & Singh, H. 1998. The relational view: Cooperative in entrepreneurial threshold firms: Imprinting and strategic strategy and sources of interorganizational competitive advan- choice perspectives. Journal of Management Studies, 52(4): tage. Academy of Management Review, 23(4): 660–679. 506–530. Fiss, P. C. 2011. Building better causal theories: A fuzzy set Kogut, B., & Ragin, C. C. 2006. Exploring complexity when approach to typologies in organization research. Academy of diversity is limited: Institutional complementarity in theories of Management Journal, 54(2): 393–420. rule of law and national systems revisited. European Manage- Furman, J. L., Porter, M. E., & Stern, S. 2002. The determinants ment Review, 3(1): 44–59. of national innovative capacity. Research Policy, 31(6): La Porta, R., Lopez-de-Silanes, F., Schleifer, A., & Vishney, R. W. 899–933. 1998. Law and finance. Journal of Political Economy, 106(6): Gospel, H., & Pendleton, A. 2003. Finance, corporate gover- 1113–1155. nance and the management of labour: A conceptual and La Porta, R., Lopez-de-Silanes, F., & Shleifer, A. 1999. Corporate comparative analysis. British Journal of Industrial Relations, ownership around the world. Journal of Finance, 54(2): 41(3): 557–582. 471–517. Gospel, H., & Pendleton, A. (Eds) 2005. Corporate governance Longest, K. C., & Vaisey, S. 2008. Fuzzy: A program for and labour management: An international comparison. Oxford: performing qualitative comparative analyses (QCA) in stata. Oxford University Press. Stata Journal, 8(1): 79–104. Hall, B. H., Bessen, J., & Thoma, G. 2011. NBER patent data project. Mahoney, J., & Thelen, K. 2009. Explaining institutional change: https://sites.google.com/site/patentdataproject/Home. Ambiguity, agency, and power. Cambridge: Cambridge Univer- Hall, P. A., & Gingerich, D. W. 2009. Varieties of capitalism and sity Press. institutional complementarities in the political economy. Milgrom, P. R., & Roberts, J. 1994. Complementarities and British Journal of Political Science, 39(3): 449–482. systems: Understanding Japanese economic organization. Hall, B. H., Jaffe, A. B., & Trajtenberg, M. 2001. The NBER patent Estudios Economicos, 9(1): 3–42. citations data file: Lessons, insights and methodological tools. Milgrom, P. R., & Roberts, J. 1995. Complementarities and NBER working paper series no. 8498. fit: Strategy, structure, and organizational change in manu- Hall, P. A., & Soskice, D. 2001. An introduction to varieties of facturing. Journal of Accounting and Economics, 19(2/3): capitalism. In P. A. Hall, & D. Soskice (Eds), Varieties of 179–208. capitalism: The institutional foundations of comparative advan- Newbert, S. L. 2007. Empirical research on the resource-based tage: 1–68. Oxford: Oxford University Press. view of the firm: An assessment and suggestions for future Hall, P. A., & Thelen, K. 2009. Institutional change in varieties of research. Strategic Management Journal, 28(2): 121–140. capitalism. Socio-Economic Review, 7(1): 7–34. No¨lke, A., & Vliegenthart, A. 2009. Enlarging the varieties of Hancke, B., Rhodes, M., & Thatcher, M. (Eds) 2007. Beyond capitalism: The emergence of dependent market economies in varieties of capitalism: Contradictions, complementarities & east central Europe. World Politics, 61(4): 670–702. change. Oxford: Oxford University Press. OECD. 2003. Science, technology and industry scoreboard. Paris: Henisz, W. J. 2002. Politics and international investment: Mea- OECD. suring risks and protecting profits. London: Edward Elgar. OECD. 2013. OECD.Stat. www.oecd-ilibrary.org/content/data/ Herrmann, A. M., & Peine, A. 2011. When ‘national innovation data-00285-en.2013. system’ meet ‘varieties of capitalism’ arguments on labour Porter, M. E. 1990. The competitive advantage of nations. New qualifications: On the skill types and scientific knowledge York: Free Press. needed for radical and incremental product innovations. Ragin, C. C. 2000. Fuzzy-set social science. Chicago: University of Research Policy, 40(5): 687–701. Chicago Press. Ho¨pner, M. 2005. What connects industrial relations and Ragin, C. C. 2008. Redesigning social inquiry: Fuzzy sets and corporate governance? Explaining institutional complemen- beyond. Chicago: University of Chicago Press. tarity. Socio-Economic Review, 3(2): 331–358. Rajan, R. G., & Zingales, L. 2000. The governance of the new Jackson, G. 2003. Corporate governance in Germany and Japan: enterprises. In X. Vives (Ed), Corporate governance: Theoretical Liberalization pressures and responses. In K. Yamamura, & W. & empirical perspectives. Cambridge: Cambridge University Streeck (Eds), The end of diversity? Prospects for German and Press. Japanese capitalism: 261–305. Ithaca, NY: Cornell University Redding, G., & Witt, M. A. 2007. The future of Chinese capitalism: Press. Choices and chances. Oxford: Oxford University Press. Jackson, G. 2005. Employee representation in the board com- Redding, G., & Witt, M. 2009. China’s business system and its pared: A fuzzy sets analysis of corporate governance, unionism, future trajectory. Asia Pacific Journal of Management, 26(3): and political institutions. Industrielle Beziehungen, 12(3): 1–28. 381–399. Jackson, G., & Deeg, R. 2008. Comparing capitalisms: Under- Rodriguez, P., Uhlenbruck, L., & Eden, L. 2005. Government standing institutional diversity and its implications for inter- corruption and the entry strategies of multinationals. Academy national business. Journal of International Business Studies, of Management Review, 30(2): 383–396. 39(4): 540–561. Sako, M. 1992. Price, quality and trust: Inter-firm relations in Jackson, G., & Muellenborn, T. 2012. Understanding the role of britain and Japan. Cambridge: Cambridge University Press. institutions in industrial relations: Perspectives from classical Schneider, M. R., & Paunescu, M. 2012. Changing varieties of sociological theory. Industrial Relations: A Journal of Economy capitalism and revealed comparative advantages from 1990 to and Society, 51(S1): 472–500. 2005: A test of the Hall and Soskice claims. Socio-Economic Jackson, G., & Ni, N. 2013. Understanding complementarities as Review, 10(4): 731–753. organizational configurations using set theoretical methods. In Schneider, M. R., Schulze-Bentrop, C., & Paunescu, M. 2010. P. C. Fiss, B. Cambre, & A. Marx (Eds), Configurational theory Mapping the institutional capital of high-tech firms: A fuzzy- Journal of International Business Studies VOC and Comparative Advantage Michael A. Witt and Gregory Jackson set analysis of capitalism variety and export performance. Institute for Advanced Labour Studies, University of Journal of International Business Studies, 41(2): 246–266. Amsterdam. Schneider, C. Q., & Wagemann, C. 2012. Set-theoretic methods Vitols, S. 2002. Shareholder value, management culture and for the social sciences: A guide to qualitative comparative production regimes in the transformation of the German analysis. Cambridge: Cambridge University Press. chemical-pharmaceutical industry. Competition and Change, Singh, J. 2007. Asymmetry of knowledge spillovers between 6(3): 309–325. MNCs and host country firms. Journal of International Business Vitols, S. 2004. Negotiated shareholder value: The German Studies, 38(5): 764–786. version of an anglo-American practice. Competition and StataCorp. 2011. Stata statistical software: Release 12. College Change, 8(4): 1–18. Station, TX: StataCorp. LP. Whitley, R. 1999. Divergent capitalisms: The social structuring and StataCorp. 2013. Stata statistical software: Release 13. College change of business systems. Oxford: Oxford University Press. Station, TX: StataCorp. Witt, M. A. 2006. Changing Japanese capitalism: Societal coordi- Streeck, W. 1997. Beneficial constraints: On the economic limits nation and institutional adjustment. Cambridge: Cambridge of rational voluntarism. In J. R. Hollingsworth & R. Boyer (Eds), University Press. Contemporary capitalism: The embeddedness of institutions: Witt, M. A., & Lewin, A. Y. 2007. Outward foreign direct 197–219. Cambridge: Cambridge University Press. investment as escape response to home country institutional Streeck, W. 2008. Re-forming capitalism: Institutional change in constraints. Journal of International Business Studies, 38(4): the German political economy: Institutional change in the 579–594. German political economy. Oxford: Oxford University Press. Witt, M. A., & Redding, G. 2013. Asian business systems: Streeck, W. 2009. Re-forming capitalism: Institutional change in Institutional comparison, clusters and implications for varieties the German political economy. Oxford: Oxford University Press. of capitalism and business systems theory. Socio-Economic Streeck, W., & Thelen, K. (Eds) 2005. Beyond continuity: Review, 11(2): 265–300. Institutional change in advanced political economies. Oxford: Witt, M. A., Kabbach de Castro, L. R., Amaeshi, K., Mahroum, S., Oxford University Press. Bohle, D., & Saez, L. 2015. The business systems of the world’s Taylor, M. Z. 2004. Empirical evidence against varieties of leading 61 economies: Institutional comparison, clusters, and capitalism’s theory of technological innovation. International implications for varieties of capitalism and business systems Organization, 58(Summer): 601–631. research. INSEAD working paper series, INSEAD, Fon- Thelen, K. 2014. Varieties of liberalization and the new politics of tainebleau, France. social solidarity. Cambridge: Cambridge University Press. World Bank. 2013. World development indicators. http://data. Transparency International. 2013. Corruption perception index. worldbank.org/data-catalog/world-development-indicators. http://www.transparency.org.cpi2011. 2011. Trefler, D., & Zhu, S. C. 2000. Beyond the algebra of Yamamura, K., & Streeck, W. (Eds) 2003. The end of diversity? explanation: Hov for the technology age. American Economic Prospects for German and Japanese capitalism. Ithaca, NY: Review, 90(2): 145–149. Cornell University Press. Visser, J. 2011. Data base on institutional characteristics of trade unions, wage setting, state intervention and social pacts, 1960–2010 (ICTWSS),Version 3.0. Amsterdam: Amsterdam APPENDIX Data As a robustness test, we used regression analysis to Dependent variables were the contribution to the undertake an alternative test of H1. Regression trade balance plus the Balassa index, as discussed in analysis is not suitable for testing H2 and H3 the main text. because the configurational logic of these argu- The independent variable of interest was the ments would involve 6-way interaction terms. interaction term between radicality by number of However, the argument of coherence across all 5 citations received and the coordination index we spheres of the political economy inherent in H1 introduce below. To account for possible direct makes it possible to derive a single ‘‘coordination effects, we also included both the radicality mea- index’’ to use in a 2-way interaction. While fsQCA sure and the coordination measure in the still remains the method of choice for this kind of regression. investigation, not least because of its ability to accommodate asymmetric solutions (see discussion in the main text), we present this additional test as Coordination a robustness check. For each country and year, we calculated an index This Appendix reports the data, and methodol- of overall coordination. Hall and Gingerich (2009) used confirmatory factor analysis to derive a coor- ogy and main results. The results depend on the operationalization of the dependent variable: our dination index for 20 advanced industrialized preferred measure, contributions to the trade bal- economies, with index values closer to 0 denoting ance, does not support H1, while the Balassa index higher similarity with the prototypical LME and as an alternative offers inconsistent support of H1. closer to 1, with the prototypical CME. Their Journal of International Business Studies VOC and Comparative Advantage Michael A. Witt and Gregory Jackson coordination index is based on three indicators of Hall and Gingerich but were inspired by the 5 corporate governance and finance (an index of institutional domains described in the original for- shareholder protection, dispersion of control, and mulation of Varieties of Capitalism by Hall and the size of stock market) and three indicators of Soskice (2001). Taking z-transformed variables to employment relations (the level of wage bargain- create a uniform scale, we used the principal-factor ing, the degrees of coordination in wage bargain- method in Stata (StataCorp, 2011) to calculate a ing, and the proportion of employees with tenures single factor loading using squared multiple corre- of less than 1 year). This original index has a lations to estimate the commonality of an underly- number of limitations that made it unsuitable for ing factor. The single largest factor had a factor our study. First, it provides a point estimate for loading of 74 % and a large eigenvalue of 4.14, about the mid-1990s rather than a time-series over suggesting a strong correlation along this single the period of our study. Second, the original factor dimension of coordination. The second factor had scores calculated in Hall and Gingerich combined an eigenvalue of 1.14 and explained an additional indicators from the late 1980s to the mid-1990s. 20.7 % of the variation, which suggests that some When synchronizing the data as an annual time important variance is not fully captured in a single series for 1995–2003, we found that the indicators index. However, since we are interested in the extent did not load on a common underlying factor when to which countries vary on a single dimension, we applying confirmatory factor analysis as used by did not interpret any of the other factor scores. Hall and Gingerich. Figure A1 shows the factor scores for countries in We developed an alternative specification of the our sample over time. While there is some move- coordination index using principal component ment of CMEs toward the LME end of the spec- analysis for the period 1995 through 2003. We drew trum, supporting the notion that economies on the 12 indicators used in our main fsQCA model liberalized in the 1990s and early 2000s, we did as described in Table 2 of the main text. This set of not observe any major changes in the general indicators is more comprehensive that those used by ordering of positions. Table A1 lists the values. Au Au Au Au Au Au Au Au Au Au Au Au Au Aus s s s s s s s s s s s s st t t t t t t t t t t t t tr r r r r r r r r r r r r riiiiiiiiiiiiiia a a a a a a a a a a a a a Au Au Au Au Au Au Au Au Au Au Au Au Au Aus s s s s s s s s s s s s st t t t t t t t t t t t t tr r r r r r r r r r r r r riiiiiiiiiiiiiia a a a a a a a a a a a a a Au Au Au Au Au Au Au Au Au Au Au Au Au Aus s s s s s s s s s s s s st t t t t t t t t t t t t tr r r r r r r r r r r r r riiiiiiiiiiiiiia a a a a a a a a a a a a a Be Be Be Be Be Be Be Be Be Be Be Be Be Bellllllllllllllg g g g g g g g g g g g g giiiiiiiiiiiiiiu u u u u u u u u u u u u um m m m m m m m m m m m m m Au Au Au Au Au Au Au Au Au Au Au Au Au Aus s s s s s s s s s s s s st t t t t t t t t t t t t tr r r r r r r r r r r r r riiiiiiiiiiiiiia a a a a a a a a a a a a a Au Au Au Au Au Au Au Au Au Au Au Au Au Aus s s s s s s s s s s s s st t t t t t t t t t t t t tr r r r r r r r r r r r r riiiiiiiiiiiiiia a a a a a a a a a a a a a G G G G G G G G G G G G G Ge e e e e e e e e e e e e er r r r r r r r r r r r r rm m m m m m m m m m m m m many any any any any any any any any any any any any any Au Au Au Au Au Au Au Au Au Au Au Au Au Aus s s s s s s s s s s s s st t t t t t t t t t t t t tr r r r r r r r r r r r r riiiiiiiiiiiiiia a a a a a a a a a a a a a G G G G G G G G G G G G G Ge e e e e e e e e e e e e er r r r r r r r r r r r r rm m m m m m m m m m m m m many any any any any any any any any any any any any any Be Be Be Be Be Be Be Be Be Be Be Be Be Bellllllllllllllg g g g g g g g g g g g g giiiiiiiiiiiiiiu u u u u u u u u u u u u um m m m m m m m m m m m m m G G G G G G G G G G G G G Ge e e e e e e e e e e e e er r r r r r r r r r r r r rm m m m m m m m m m m m m many any any any any any any any any any any any any any Au Au Au Au Au Au Au Au Au Au Au Au Au Aus s s s s s s s s s s s s st t t t t t t t t t t t t tr r r r r r r r r r r r r riiiiiiiiiiiiiia a a a a a a a a a a a a a Be Be Be Be Be Be Be Be Be Be Be Be Be Bellllllllllllllg g g g g g g g g g g g g giiiiiiiiiiiiiiu u u u u u u u u u u u u um m m m m m m m m m m m m m Ge Ge Ge Ge Ge Ge Ge Ge Ge Ge Ge Ge Ge Ger r r r r r r r r r r r r rm m m m m m m m m m m m m ma a a a a a a a a a a a a an n n n n n n n n n n n n ny y y y y y y y y y y y y y Au Au Au Au Au Au Au Au Au Au Au Au Au Aus s s s s s s s s s s s s st t t t t t t t t t t t t tr r r r r r r r r r r r r riiiiiiiiiiiiiia a a a a a a a a a a a a a Au Au Au Au Au Au Au Au Au Au Au Au Au Aus s s s s s s s s s s s s st t t t t t t t t t t t t tr r r r r r r r r r r r r riiiiiiiiiiiiiia a a a a a a a a a a a a a Ge Ge Ge Ge Ge Ge Ge Ge Ge Ge Ge Ge Ge Ger r r r r r r r r r r r r rm m m m m m m m m m m m m ma a a a a a a a a a a a a an n n n n n n n n n n n n ny y y y y y y y y y y y y y Be Be Be Be Be Be Be Be Be Be Be Be Be Bellllllllllllllg g g g g g g g g g g g g giiiiiiiiiiiiiiu u u u u u u u u u u u u um m m m m m m m m m m m m m Ge Ge Ge Ge Ge Ge Ge Ge Ge Ge Ge Ge Ge Ger r r r r r r r r r r r r rm m m m m m m m m m m m m ma a a a a a a a a a a a a an n n n n n n n n n n n n ny y y y y y y y y y y y y y N N N N N N N N N N N N N Ne e e e e e e e e e e e e et t t t t t t t t t t t t ther her her her her her her her her her her her her herlllllllllllllla a a a a a a a a a a a a an n n n n n n n n n n n n nd d d d d d d d d d d d d ds s s s s s s s s s s s s s N N N N N N N N N N N N N Ne e e e e e e e e e e e e et t t t t t t t t t t t t ther her her her her her her her her her her her her herlllllllllllllla a a a a a a a a a a a a an n n n n n n n n n n n n nd d d d d d d d d d d d d ds s s s s s s s s s s s s s Ge Ge Ge Ge Ge Ge Ge Ge Ge Ge Ge Ge Ge Ger r r r r r r r r r r r r rm m m m m m m m m m m m m ma a a a a a a a a a a a a an n n n n n n n n n n n n ny y y y y y y y y y y y y y It It It It It It It It It It It It It Ita a a a a a a a a a a a a alllllllllllllly y y y y y y y y y y y y y N N N N N N N N N N N N N Ne e e e e e e e e e e e e et t t t t t t t t t t t t ther her her her her her her her her her her her her herlllllllllllllla a a a a a a a a a a a a an n n n n n n n n n n n n nd d d d d d d d d d d d d ds s s s s s s s s s s s s s Ge Ge Ge Ge Ge Ge Ge Ge Ge Ge Ge Ge Ge Ger r r r r r r r r r r r r rm m m m m m m m m m m m m ma a a a a a a a a a a a a an n n n n n n n n n n n n ny y y y y y y y y y y y y y Be Be Be Be Be Be Be Be Be Be Be Be Be Bellllllllllllllg g g g g g g g g g g g g giiiiiiiiiiiiiiu u u u u u u u u u u u u um m m m m m m m m m m m m m Ge Ge Ge Ge Ge Ge Ge Ge Ge Ge Ge Ge Ge Ger r r r r r r r r r r r r rm m m m m m m m m m m m m ma a a a a a a a a a a a a an n n n n n n n n n n n n ny y y y y y y y y y y y y y S S S S S S S S S S S S S Sw w w w w w w w w w w w w weden eden eden eden eden eden eden eden eden eden eden eden eden eden S S S S S S S S S S S S S Sw w w w w w w w w w w w w weden eden eden eden eden eden eden eden eden eden eden eden eden eden Be Be Be Be Be Be Be Be Be Be Be Be Be Bellllllllllllllg g g g g g g g g g g g g giiiiiiiiiiiiiiu u u u u u u u u u u u u um m m m m m m m m m m m m m It It It It It It It It It It It It It Ita a a a a a a a a a a a a alllllllllllllly y y y y y y y y y y y y y N N N N N N N N N N N N N Ne e e e e e e e e e e e e et t t t t t t t t t t t t ther her her her her her her her her her her her her herlllllllllllllla a a a a a a a a a a a a an n n n n n n n n n n n n nd d d d d d d d d d d d d ds s s s s s s s s s s s s s It It It It It It It It It It It It It Ita a a a a a a a a a a a a alllllllllllllly y y y y y y y y y y y y y No No No No No No No No No No No No No Nor r r r r r r r r r r r r rw w w w w w w w w w w w w wa a a a a a a a a a a a a ay y y y y y y y y y y y y y S S S S S S S S S S S S S Sw w w w w w w w w w w w w weden eden eden eden eden eden eden eden eden eden eden eden eden eden It It It It It It It It It It It It It Ita a a a a a a a a a a a a alllllllllllllly y y y y y y y y y y y y y Be Be Be Be Be Be Be Be Be Be Be Be Be Bellllllllllllllg g g g g g g g g g g g g giiiiiiiiiiiiiiu u u u u u u u u u u u u um m m m m m m m m m m m m m No No No No No No No No No No No No No Nor r r r r r r r r r r r r rw w w w w w w w w w w w w wa a a a a a a a a a a a a ay y y y y y y y y y y y y y N N N N N N N N N N N N N Ne e e e e e e e e e e e e et t t t t t t t t t t t t ther her her her her her her her her her her her her herlllllllllllllla a a a a a a a a a a a a an n n n n n n n n n n n n nd d d d d d d d d d d d d ds s s s s s s s s s s s s s Be Be Be Be Be Be Be Be Be Be Be Be Be Bellllllllllllllg g g g g g g g g g g g g giiiiiiiiiiiiiiu u u u u u u u u u u u u um m m m m m m m m m m m m m De De De De De De De De De De De De De Den n n n n n n n n n n n n nm m m m m m m m m m m m m ma a a a a a a a a a a a a ar r r r r r r r r r r r r rk k k k k k k k k k k k k k De De De De De De De De De De De De De Den n n n n n n n n n n n n nm m m m m m m m m m m m m ma a a a a a a a a a a a a ar r r r r r r r r r r r r rk k k k k k k k k k k k k k It It It It It It It It It It It It It Ita a a a a a a a a a a a a alllllllllllllly y y y y y y y y y y y y y De De De De De De De De De De De De De Den n n n n n n n n n n n n nm m m m m m m m m m m m m ma a a a a a a a a a a a a ar r r r r r r r r r r r r rk k k k k k k k k k k k k k S S S S S S S S S S S S S Sw w w w w w w w w w w w w weden eden eden eden eden eden eden eden eden eden eden eden eden eden Be Be Be Be Be Be Be Be Be Be Be Be Be Bellllllllllllllg g g g g g g g g g g g g giiiiiiiiiiiiiiu u u u u u u u u u u u u um m m m m m m m m m m m m m De De De De De De De De De De De De De Den n n n n n n n n n n n n nm m m m m m m m m m m m m ma a a a a a a a a a a a a ar r r r r r r r r r r r r rk k k k k k k k k k k k k k It It It It It It It It It It It It It Ita a a a a a a a a a a a a alllllllllllllly y y y y y y y y y y y y y S S S S S S S S S S S S S Sw w w w w w w w w w w w w wiiiiiiiiiiiiiit t t t t t t t t t t t t tzer zer zer zer zer zer zer zer zer zer zer zer zer zerlllllllllllllland and and and and and and and and and and and and and N N N N N N N N N N N N N Ne e e e e e e e e e e e e et t t t t t t t t t t t t ther her her her her her her her her her her her her herlllllllllllllla a a a a a a a a a a a a an n n n n n n n n n n n n nd d d d d d d d d d d d d ds s s s s s s s s s s s s s N N N N N N N N N N N N N Ne e e e e e e e e e e e e et t t t t t t t t t t t t ther her her her her her her her her her her her her herlllllllllllllla a a a a a a a a a a a a an n n n n n n n n n n n n nd d d d d d d d d d d d d ds s s s s s s s s s s s s s Gr Gr Gr Gr Gr Gr Gr Gr Gr Gr Gr Gr Gr Gre e e e e e e e e e e e e ee e e e e e e e e e e e e ec c c c c c c c c c c c c ce e e e e e e e e e e e e e N N N N N N N N N N N N N Ne e e e e e e e e e e e e et t t t t t t t t t t t t ther her her her her her her her her her her her her herlllllllllllllla a a a a a a a a a a a a an n n n n n n n n n n n n nd d d d d d d d d d d d d ds s s s s s s s s s s s s s Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Finl nl nl nl nl nl nl nl nl nl nl nl nl nland and and and and and and and and and and and and and Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Finl nl nl nl nl nl nl nl nl nl nl nl nl nland and and and and and and and and and and and and and Gr Gr Gr Gr Gr Gr Gr Gr Gr Gr Gr Gr Gr Gre e e e e e e e e e e e e ee e e e e e e e e e e e e ec c c c c c c c c c c c c ce e e e e e e e e e e e e e It It It It It It It It It It It It It Ita a a a a a a a a a a a a alllllllllllllly y y y y y y y y y y y y y S S S S S S S S S S S S S Sw w w w w w w w w w w w w weden eden eden eden eden eden eden eden eden eden eden eden eden eden N N N N N N N N N N N N N Ne e e e e e e e e e e e e et t t t t t t t t t t t t ther her her her her her her her her her her her her herlllllllllllllla a a a a a a a a a a a a an n n n n n n n n n n n n nd d d d d d d d d d d d d ds s s s s s s s s s s s s s No No No No No No No No No No No No No Nor r r r r r r r r r r r r rw w w w w w w w w w w w w wa a a a a a a a a a a a a ay y y y y y y y y y y y y y No No No No No No No No No No No No No Nor r r r r r r r r r r r r rw w w w w w w w w w w w w wa a a a a a a a a a a a a ay y y y y y y y y y y y y y G G G G G G G G G G G G G Gr r r r r r r r r r r r r reec eec eec eec eec eec eec eec eec eec eec eec eec eece e e e e e e e e e e e e e Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Finl nl nl nl nl nl nl nl nl nl nl nl nl nland and and and and and and and and and and and and and De De De De De De De De De De De De De Den n n n n n n n n n n n n nm m m m m m m m m m m m m ma a a a a a a a a a a a a ar r r r r r r r r r r r r rk k k k k k k k k k k k k k S S S S S S S S S S S S S Sw w w w w w w w w w w w w weden eden eden eden eden eden eden eden eden eden eden eden eden eden It It It It It It It It It It It It It Ita a a a a a a a a a a a a alllllllllllllly y y y y y y y y y y y y y Swi Swi Swi Swi Swi Swi Swi Swi Swi Swi Swi Swi Swi Swit t t t t t t t t t t t t tz z z z z z z z z z z z z ze e e e e e e e e e e e e er r r r r r r r r r r r r rlllllllllllllla a a a a a a a a a a a a an n n n n n n n n n n n n nd d d d d d d d d d d d d d No No No No No No No No No No No No No Nor r r r r r r r r r r r r rw w w w w w w w w w w w w wa a a a a a a a a a a a a ay y y y y y y y y y y y y y De De De De De De De De De De De De De Den n n n n n n n n n n n n nm m m m m m m m m m m m m ma a a a a a a a a a a a a ar r r r r r r r r r r r r rk k k k k k k k k k k k k k Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Franc anc anc anc anc anc anc anc anc anc anc anc anc ance e e e e e e e e e e e e e It It It It It It It It It It It It It Ita a a a a a a a a a a a a alllllllllllllly y y y y y y y y y y y y y Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Franc anc anc anc anc anc anc anc anc anc anc anc anc ance e e e e e e e e e e e e e Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Gr Gr Gr Gr Gr Gr Gr Gr Gr Gr Gr Gr Gr Gr nl nl nl nl nl nl nl nl nl nl nl nl nl nl e e e e e e e e e e e e e eand and and and and and and and and and and and and and e e e e e e e e e e e e e ec c c c c c c c c c c c c ce e e e e e e e e e e e e e No No No No No No No No No No No No No Nor r r r r r r r r r r r r rw w w w w w w w w w w w w wa a a a a a a a a a a a a ay y y y y y y y y y y y y y No No No No No No No No No No No No No Nor r r r r r r r r r r r r rw w w w w w w w w w w w w wa a a a a a a a a a a a a ay y y y y y y y y y y y y y No No No No No No No No No No No No No Nor r r r r r r r r r r r r rw w w w w w w w w w w w w wa a a a a a a a a a a a a ay y y y y y y y y y y y y y Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Franc anc anc anc anc anc anc anc anc anc anc anc anc ance e e e e e e e e e e e e e Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Finl nl nl nl nl nl nl nl nl nl nl nl nl nland and and and and and and and and and and and and and S S S S S S S S S S S S S Sw w w w w w w w w w w w w weden eden eden eden eden eden eden eden eden eden eden eden eden eden S S S S S S S S S S S S S Sw w w w w w w w w w w w w weden eden eden eden eden eden eden eden eden eden eden eden eden eden Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Franc anc anc anc anc anc anc anc anc anc anc anc anc ance e e e e e e e e e e e e e Swi Swi Swi Swi Swi Swi Swi Swi Swi Swi Swi Swi Swi Swit t t t t t t t t t t t t tz z z z z z z z z z z z z ze e e e e e e e e e e e e er r r r r r r r r r r r r rlllllllllllllla a a a a a a a a a a a a an n n n n n n n n n n n n nd d d d d d d d d d d d d d G G G G G G G G G G G G G Gr r r r r r r r r r r r r reec eec eec eec eec eec eec eec eec eec eec eec eec eece e e e e e e e e e e e e e G G G G G G G G G G G G G Gr r r r r r r r r r r r r reec eec eec eec eec eec eec eec eec eec eec eec eec eece e e e e e e e e e e e e e Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Franc anc anc anc anc anc anc anc anc anc anc anc anc ance e e e e e e e e e e e e e No No No No No No No No No No No No No Nor r r r r r r r r r r r r rw w w w w w w w w w w w w wa a a a a a a a a a a a a ay y y y y y y y y y y y y y G G G G G G G G G G G G G Gr r r r r r r r r r r r r reec eec eec eec eec eec eec eec eec eec eec eec eec eece e e e e e e e e e e e e e G G G G G G G G G G G G G Gr r r r r r r r r r r r r reec eec eec eec eec eec eec eec eec eec eec eec eec eece e e e e e e e e e e e e e Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Franc anc anc anc anc anc anc anc anc anc anc anc anc ance e e e e e e e e e e e e e S S S S S S S S S S S S S Sw w w w w w w w w w w w w weden eden eden eden eden eden eden eden eden eden eden eden eden eden P P P P P P P P P P P P P Po o o o o o o o o o o o o or r r r r r r r r r r r r rt t t t t t t t t t t t t tu u u u u u u u u u u u u ugal gal gal gal gal gal gal gal gal gal gal gal gal gal Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Franc anc anc anc anc anc anc anc anc anc anc anc anc ance e e e e e e e e e e e e e Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Franc anc anc anc anc anc anc anc anc anc anc anc anc ance e e e e e e e e e e e e e G G G G G G G G G G G G G Gr r r r r r r r r r r r r reec eec eec eec eec eec eec eec eec eec eec eec eec eece e e e e e e e e e e e e e S S S S S S S S S S S S S Sw w w w w w w w w w w w w wiiiiiiiiiiiiiit t t t t t t t t t t t t tzer zer zer zer zer zer zer zer zer zer zer zer zer zerlllllllllllllland and and and and and and and and and and and and and Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Franc anc anc anc anc anc anc anc anc anc anc anc anc ance e e e e e e e e e e e e e P P P P P P P P P P P P P Po o o o o o o o o o o o o or r r r r r r r r r r r r rt t t t t t t t t t t t t tu u u u u u u u u u u u u ugal gal gal gal gal gal gal gal gal gal gal gal gal gal P P P P P P P P P P P P P Po o o o o o o o o o o o o or r r r r r r r r r r r r rt t t t t t t t t t t t t tu u u u u u u u u u u u u ugal gal gal gal gal gal gal gal gal gal gal gal gal gal De De De De De De De De De De De De De Den n n n n n n n n n n n n nm m m m m m m m m m m m m ma a a a a a a a a a a a a ar r r r r r r r r r r r r rk k k k k k k k k k k k k k De De De De De De De De De De De De De Den n n n n n n n n n n n n nm m m m m m m m m m m m m ma a a a a a a a a a a a a ar r r r r r r r r r r r r rk k k k k k k k k k k k k k Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Kor r r r r r r r r r r r r re e e e e e e e e e e e e ea a a a a a a a a a a a a a Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Finl nl nl nl nl nl nl nl nl nl nl nl nl nland and and and and and and and and and and and and and S S S S S S S S S S S S S Sw w w w w w w w w w w w w wiiiiiiiiiiiiiit t t t t t t t t t t t t tzer zer zer zer zer zer zer zer zer zer zer zer zer zerlllllllllllllland and and and and and and and and and and and and and S S S S S S S S S S S S S Sw w w w w w w w w w w w w wiiiiiiiiiiiiiit t t t t t t t t t t t t tzer zer zer zer zer zer zer zer zer zer zer zer zer zerlllllllllllllla a a a a a a a a a a a a and nd nd nd nd nd nd nd nd nd nd nd nd nd S S S S S S S S S S S S S Sw w w w w w w w w w w w w wiiiiiiiiiiiiiit t t t t t t t t t t t t tzer zer zer zer zer zer zer zer zer zer zer zer zer zerlllllllllllllland and and and and and and and and and and and and and De De De De De De De De De De De De De Den n n n n n n n n n n n n nm m m m m m m m m m m m m ma a a a a a a a a a a a a ar r r r r r r r r r r r r rk k k k k k k k k k k k k k S S S S S S S S S S S S S Sw w w w w w w w w w w w w wiiiiiiiiiiiiiit t t t t t t t t t t t t tzer zer zer zer zer zer zer zer zer zer zer zer zer zerlllllllllllllla a a a a a a a a a a a a and nd nd nd nd nd nd nd nd nd nd nd nd nd Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Finl nl nl nl nl nl nl nl nl nl nl nl nl nland and and and and and and and and and and and and and P P P P P P P P P P P P P Po o o o o o o o o o o o o or r r r r r r r r r r r r rt t t t t t t t t t t t t tu u u u u u u u u u u u u ugal gal gal gal gal gal gal gal gal gal gal gal gal gal S S S S S S S S S S S S S Sw w w w w w w w w w w w w wiiiiiiiiiiiiiit t t t t t t t t t t t t tzer zer zer zer zer zer zer zer zer zer zer zer zer zerlllllllllllllland and and and and and a a a a a a a and nd nd nd nd nd nd nd Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Kor r r r r r r r r r r r r re e e e e e e e e e e e e ea a a a a a a a a a a a a a Po Po Po Po Po Po Po Po Po Po Po Po Po Por r r r r r r r r r r r r rt t t t t t t t t t t t t tu u u u u u u u u u u u u ug g g g g g g g g g g g g ga a a a a a a a a a a a a allllllllllllll Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Finl nl nl nl nl nl nl nl nl nl nl nl nl nland and and and and and and and and and and and and and Po Po Po Po Po Po Po Po Po Po Po Po Po Por r r r r r r r r r r r r rt t t t t t t t t t t t t tu u u u u u u u u u u u u ug g g g g g g g g g g g g ga a a a a a a a a a a a a allllllllllllll Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Finl nl nl nl nl nl nl nl nl nl nl nl nl nland and and and and and and and and and and and and and Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Kor r r r r r r r r r r r r re e e e e e e e e e e e e ea a a a a a a a a a a a a a Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Kor r r r r r r r r r r r r re e e e e e e e e e e e e ea a a a a a a a a a a a a a Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Po Po Po Po Po Po Po Po Po Po Po Po Po Por r r r r r r r r r r r r rt t t t t t t t t t t t t tu u u u u u u u u u u u u ug g g g g g g g g g g g g ga a a a a a a a a a a a a allllllllllllll Po Po Po Po Po Po Po Po Po Po Po Po Po Por r r r r r r r r r r r r rt t t t t t t t t t t t t tu u u u u u u u u u u u u ug g g g g g g g g g g g g ga a a a a a a a a a a a a allllllllllllll S S S S S S S S S S S S S Spai pai pai pai pai pai pai pai pai pai pai pai pai pain n n n n n n n n n n n n n Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Kor r r r r r r r r r r r r re e e e e e e e e e e e e ea a a a a a a a a a a a a a Po Po Po Po Po Po Po Po Po Po Po Po Po Por r r r r r r r r r r r r rt t t t t t t t t t t t t tu u u u u u u u u u u u u ug g g g g g g g g g g g g ga a a a a a a a a a a a a allllllllllllll Sp Sp Sp Sp Sp Sp Sp Sp Sp Sp Sp Sp Sp Spa a a a a a a a a a a a a aiiiiiiiiiiiiiin n n n n n n n n n n n n n S S S S S S S S S S S S S Spai pai pai pai pai pai pai pai pai pai pai pai pai pain n n n n n n n n n n n n n S S S S S S S S S S S S S Spai pai pai pai pai pai pai pai pai pai pai pai pai pain n n n n n n n n n n n n n Sp Sp Sp Sp Sp Sp Sp Sp Sp Sp Sp Sp Sp Sp Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan a a a a a a a a a a a a a aiiiiiiiiiiiiiin n n n n n n n n n n n n n Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko S S S S S S S S S S S S S Spai pai pai pai pai pai pai pai pai pai pai pai pai pai r r r r r r r r r r r r r re e e e e e e e e e e e e en n n n n n n n n n n n n n a a a a a a a a a a a a a a S S S S S S S S S S S S S Spai pai pai pai pai pai pai pai pai pai pai pai pai pain n n n n n n n n n n n n n Ir Ir Ir Ir Ir Ir Ir Ir Ir Ir Ir Ir Ir Ire e e e e e e e e e e e e elllllllllllllla a a a a a a a a a a a a an n n n n n n n n n n n n nd d d d d d d d d d d d d d S S S S S S S S S S S S S Spai pai pai pai pai pai pai pai pai pai pai pai pai pain n n n n n n n n n n n n n S S S S S S S S S S S S S Spai pai pai pai pai pai pai pai pai pai pai pai pai pain n n n n n n n n n n n n n Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Ir Ir Ir Ir Ir Ir Ir Ir Ir Ir Ir Ir Ir Ire e e e e e e e e e e e e elllllllllllllla a a a a a a a a a a a a an n n n n n n n n n n n n nd d d d d d d d d d d d d d Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Kor r r r r r r r r r r r r re e e e e e e e e e e e e ea a a a a a a a a a a a a a Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne New Z w Z w Z w Z w Z w Z w Z w Z w Z w Z w Z w Z w Z w Ze e e e e e e e e e e e e ea a a a a a a a a a a a a alllllllllllllla a a a a a a a a a a a a an n n n n n n n n n n n n nd d d d d d d d d d d d d d Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan I I I I I I I I I I I I I Ir r r r r r r r r r r r r rel el el el el el el el el el el el el elan an an an an an an an an an an an an and d d d d d d d d d d d d d Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Kor r r r r r r r r r r r r re e e e e e e e e e e e e ea a a a a a a a a a a a a a Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Ir Ir Ir Ir Ir Ir Ir Ir Ir Ir Ir Ir Ir Ire e e e e e e e e e e e e elllllllllllllla a a a a a a a a a a a a an n n n n n n n n n n n n nd d d d d d d d d d d d d d I I I I I I I I I I I I I Ir r r r r r r r r r r r r rel el el el el el el el el el el el el elan an an an an an an an an an an an an and d d d d d d d d d d d d d Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Kor r r r r r r r r r r r r re e e e e e e e e e e e e ea a a a a a a a a a a a a a I I I I I I I I I I I I I Ir r r r r r r r r r r r r rel el el el el el el el el el el el el elan an an an an an an an an an an an an and d d d d d d d d d d d d d N N N N N N N N N N N N N Ne e e e e e e e e e e e e ew w w w w w w w w w w w w w Z Z Z Z Z Z Z Z Z Z Z Z Z Zeal eal eal eal eal eal eal eal eal eal eal eal eal ealan an an an an an an an an an an an an and d d d d d d d d d d d d d I I I I I I I I I I I I I Ir r r r r r r r r r r r r rel el el el el el el el el el el el el elan an an an an an an an an an an an an and d d d d d d d d d d d d d Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne New Z w Z w Z w Z w Z w Z w Z w Z w Z w Z w Z w Z w Z w Ze e e e e e e e e e e e e ea a a a a a a a a a a a a alllllllllllllla a a a a a a a a a a a a an n n n n n n n n n n n n nd d d d d d d d d d d d d d I I I I I I I I I I I I I Ir r r r r r r r r r r r r rel el el el el el el el el el el el el elan an an an an an an an an an an an an and d d d d d d d d d d d d d I I I I I I I I I I I I I Ir r r r r r r r r r r r r rel el el el el el el el el el el el el elan an an an an an an an an an an an an and d d d d d d d d d d d d d Au Au Au Au Au Au Au Au Au Au Au Au Au Aus s s s s s s s s s s s s st t t t t t t t t t t t t tr r r r r r r r r r r r r ra a a a a a a a a a a a a alllllllllllllliiiiiiiiiiiiiia a a a a a a a a a a a a a Au Au Au Au Au Au Au Au Au Au Au Au Au Aus s s s s s s s s s s s s st t t t t t t t t t t t t tr r r r r r r r r r r r r ra a a a a a a a a a a a a alllllllllllllliiiiiiiiiiiiiia a a a a a a a a a a a a a Au Au Au Au Au Au Au Au Au Au Au Au Au Aus s s s s s s s s s s s s st t t t t t t t t t t t t tr r r r r r r r r r r r r ra a a a a a a a a a a a a alllllllllllllliiiiiiiiiiiiiia a a a a a a a a a a a a a Au Au Au Au Au Au Au Au Au Au Au Au Au Aus s s s s s s s s s s s s st t t t t t t t t t t t t tr r r r r r r r r r r r r ra a a a a a a a a a a a a alllllllllllllliiiiiiiiiiiiiia a a a a a a a a a a a a a Au Au Au Au Au Au Au Au Au Au Au Au Au Aus s s s s s s s s s s s s st t t t t t t t t t t t t tr r r r r r r r r r r r r ra a a a a a a a a a a a a alllllllllllllliiiiiiiiiiiiiia a a a a a a a a a a a a a Au Au Au Au Au Au Au Au Au Au Au Au Au Aus s s s s s s s s s s s s st t t t t t t t t t t t t tr r r r r r r r r r r r r ra a a a a a a a a a a a a alllllllllllllliiiiiiiiiiiiiia a a a a a a a a a a a a a Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne New Z w Z w Z w Z w Z w Z w Z w Z w Z w Z w Z w Z w Z w Ze e e e e e e e e e e e e ea a a a a a a a a a a a a alllllllllllllla a a a a a a a a a a a a an n n n n n n n n n n n n nd d d d d d d d d d d d d d US US US US US US US US US US US US US US C C C C C C C C C C C C C Ca a a a a a a a a a a a a an n n n n n n n n n n n n nada ada ada ada ada ada ada ada ada ada ada ada ada ada C C C C C C C C C C C C C Ca a a a a a a a a a a a a an n n n n n n n n n n n n nada ada ada ada ada ada ada ada ada ada ada ada ada ada C C C C C C C C C C C C C Ca a a a a a a a a a a a a an n n n n n n n n n n n n nada ada ada ada ada ada ada ada ada ada ada ada ada ada Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne New Z w Z w Z w Z w Z w Z w Z w Z w Z w Z w Z w Z w Z w Ze e e e e e e e e e e e e ea a a a a a a a a a a a a alllllllllllllla a a a a a a a a a a a a an n n n n n n n n n n n n nd d d d d d d d d d d d d d Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne New Z w Z w Z w Z w Z w Z w Z w Z w Z w Z w Z w Z w Z w Ze e e e e e e e e e e e e ea a a a a a a a a a a a a alllllllllllllla a a a a a a a a a a a a an n n n n n n n n n n n n nd d d d d d d d d d d d d d US US US US US US US US US US US US US US C C C C C C C C C C C C C Ca a a a a a a a a a a a a an n n n n n n n n n n n n nada ada ada ada ada ada ada ada ada ada ada ada ada ada C C C C C C C C C C C C C Ca a a a a a a a a a a a a an n n n n n n n n n n n n nada ada ada ada ada ada ada ada ada ada ada ada ada ada Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne New Z w Z w Z w Z w Z w Z w Z w Z w Z w Z w Z w Z w Z w Ze e e e e e e e e e e e e ea a a a a a a a a a a a a alllllllllllllla a a a a a a a a a a a a an n n n n n n n n n n n n nd d d d d d d d d d d d d d C C C C C C C C C C C C C Ca a a a a a a a a a a a a an n n n n n n n n n n n n nada ada ada ada ada ada ada ada ada ada ada ada ada ada Au Au Au Au Au Au Au Au Au Au Au Au Au Aus s s s s s s s s s s s s st t t t t t t t t t t t t tr r r r r r r r r r r r r ra a a a a a a a a a a a a alllllllllllllliiiiiiiiiiiiiia a a a a a a a a a a a a a Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne New Z w Z w Z w Z w Z w Z w Z w Z w Z w Z w Z w Z w Z w Ze e e e e e e e e e e e e ea a a a a a a a a a a a a alllllllllllllla a a a a a a a a a a a a an n n n n n n n n n n n n nd d d d d d d d d d d d d d Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne New Z w Z w Z w Z w Z w Z w Z w Z w Z w Z w Z w Z w Z w Ze e e e e e e e e e e e e ea a a a a a a a a a a a a alllllllllllllla a a a a a a a a a a a a an n n n n n n n n n n n n nd d d d d d d d d d d d d d C C C C C C C C C C C C C Ca a a a a a a a a a a a a an n n n n n n n n n n n n nada ada ada ada ada ada ada ada ada ada ada ada ada ada C C C C C C C C C C C C C Ca a a a a a a a a a a a a an n n n n n n n n n n n n nada ada ada ada ada ada ada ada ada ada ada ada ada ada US US US US US US US US US US US US US US Au Au Au Au Au Au Au Au Au Au Au Au Au Aus s s s s s s s s s s s s st t t t t t t t t t t t t tr r r r r r r r r r r r r ra a a a a a a a a a a a a alllllllllllllliiiiiiiiiiiiiia a a a a a a a a a a a a a C C C C C C C C C C C C C Ca a a a a a a a a a a a a an n n n n n n n n n n n n nada ada ada ada ada ada ada ada ada ada ada ada ada ada US US US US US US US US US US US US US US UK UK UK UK UK UK UK UK UK UK UK UK UK UK Au Au Au Au Au Au Au Au Au Au Au Au Au Aus s s s s s s s s s s s s st t t t t t t t t t t t t tr r r r r r r r r r r r r ra a a a a a a a a a a a a alllllllllllllliiiiiiiiiiiiiia a a a a a a a a a a a a a US US US US US US US US US US US US US US UK UK UK UK UK UK UK UK UK UK UK UK UK UK US US US US US US US US US US US US US US US US US US US US US US US US US US US US UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK US US US US US US US US US US US US US US UK UK UK UK UK UK UK UK UK UK UK UK UK UK US US US US US US US US US US US US US US UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK 1994 1996 1998 2000 2002 2004 Year Figure A1 Coordination index, 1995–2003. Journal of International Business Studies Index of Coordination 0 .2 .4 .6 .8 1 VOC and Comparative Advantage Michael A. Witt and Gregory Jackson Table A1 Coordination index for 22 OECD countries, 1995–2003 Country 1995 1996 1997 1998 1999 2000 2001 2002 2003 Australia 0.24 0.23 0.23 0.23 0.23 0.20 0.13 0.11 0.09 Austria 0.96 0.96 0.94 0.91 0.91 0.89 0.86 0.85 0.82 Belgium 0.92 0.88 0.86 0.80 0.75 0.72 0.69 0.69 0.67 Canada 0.17 0.17 0.17 0.15 0.12 0.11 0.12 0.14 0.13 Denmark 0.68 0.67 0.67 0.68 0.61 0.61 0.54 0.52 0.51 Finland 0.64 0.64 0.62 0.59 0.58 0.52 0.50 0.48 0.47 France 0.60 0.60 0.58 0.57 0.55 0.55 0.55 0.56 0.57 Germany 0.90 0.88 0.86 0.84 0.82 0.80 0.77 0.75 0.74 Greece 0.65 0.64 0.62 0.59 0.57 0.57 0.56 0.55 0.56 Ireland 0.37 0.34 0.31 0.28 0.25 0.25 0.26 0.27 0.26 Italy 0.76 0.72 0.71 0.70 0.68 0.66 0.64 0.62 0.60 Japan 0.46 0.43 0.43 0.37 0.36 0.34 0.33 0.32 0.30 Netherlands 0.79 0.77 0.76 0.71 0.69 0.65 0.63 0.65 0.65 New Zealand 0.33 0.26 0.25 0.19 0.17 0.16 0.13 0.13 0.13 Norway 0.71 0.69 0.62 0.62 0.61 0.56 0.59 0.59 0.58 Portugal 0.54 0.55 0.54 0.50 0.48 0.47 0.43 0.43 0.41 South Korea 0.52 0.49 0.46 0.45 0.41 0.37 0.33 0.31 0.27 Spain 0.40 0.41 0.39 0.37 0.39 0.37 0.36 0.37 0.36 Sweden 0.73 0.72 0.70 0.67 0.63 0.61 0.58 0.58 0.56 Switzerland 0.65 0.61 0.57 0.55 0.52 0.51 0.50 0.51 0.52 UK 0.10 0.08 0.06 0.03 0.01 0.02 0.01 0.04 0.06 USA 0.10 0.08 0.07 0.04 0.03 0.06 0.11 0.15 0.18 Note: Higher values indicate higher levels of CME-style coordination. We used this variable in its interaction with EU, NAFTA These country-level control variables radicality (which then provided the independent are dummy variables coded 1 if the respective variable of interest for testing H1) and its direct effect country was a member of the EU or NAFTA and 0 in the regression to control for the possibility that otherwise. It controlled for the possibility that coordination in itself affects revealed comparative membership in these free trade areas may have an advantages without interacting with radicality. impact on the structure of trade of a country and Control variables included the following: thus on the structure of contributions to the trade Number of patents This industry-level control balance. variable is the natural logarithm of the number of GDP per capita This country-level control variable patents in the given country, industry, and year is the natural logarithm of per capita GDP at (OECD, 2013). It controlled for the possibility that purchasing power parity and constant year 2000 greater contributions to the trade balance in a given US dollars (OECD, 2013). It controlled for possible industry may be the result of higher levels of structural effects in contributions to the trade innovation in general, as opposed to radical balance based on higher levels of economic innovation. development. For instance, it is possible that Corruption levels and regulatory quality These coun- higher GDP levels may be associated with greater try-level control variables are the corruption per- comparative advantage in higher technology ception index, obtained from Transparency industries. International (2013), and the regulatory quality GERD per GDP This country-level control variable measures from the World Bank Development is the ratio of Gross Expenses on R&D over GDP Indicators database (2013). They controlled for (OECD, 2013). It controlled for possible effects that the possibility that institutional shortcomings may higher levels of spending on R&D in an economy prevent patent owners from making full use of may have on contributions to the trade balance. their patents, which could reduce contributions to We further included dummies for all years in the the trade balance an economy can derive from sample. Table A2 shows descriptive statistics for all radical innovations. variables. Journal of International Business Studies VOC and Comparative Advantage Michael A. Witt and Gregory Jackson Model Specification We used hierarchical linear modeling (HLM), as implemented in the ‘‘mixed’’ command in Stata 13.1 (StataCorp, 2013). This choice was condi- tioned by the presence of multiple violations of the assumption that data are independent and identically distributed. Specifically, we expected clustering effects at two levels, country and indus- try. In line with the Varieties of Capitalism approach, we assumed that country-level charac- teristics supersede industry-level characteristics and consequently specified a three-level HLM model that nested observations first in industries and then in countries. We further specified robust standard errors with correction for AR(1) autoregressive effects. Robustness tests with AR(2) and AR(3) specification yielded qualitatively identical results. Results Models (1) through (3) in Table A3 show the build- up to our main model (3), which is a three-level HLM specification regressing the contribution to the trade balance variable on the interaction of coordination and radicality expressed in citations received. The coefficient estimate for the interaction effect is negative, as expected, and statistically significant. However, further analysis revealed that the result in model (3) is dependent on the presence of two specific sets of data points: the chemical and pharmaceutical industries in Ireland, both of which have received unusually large levels of inward foreign direct investment for production for export. Dropping these two industries for Ireland resulted in insignificant coefficient estimates, as model (4) indicates. No other countries or industries showed similar impact. The problem persisted when using OLS with clustered standard errors, either by industry or country, or when changing the lag to 2 or 3 years, as well as when using the alternative measures of radicality discussed in the main text. We conclude that the results of model (3) are contingent on 18 influential data points represent- ing said two industries in Ireland and thus not generally supported. Model 5 shows the results for the Balassa index as the dependent variable. The coefficient estimate for the interaction effect is negative, as expected, and statistically significant. Unlike in model 3, this result is not contingent on the presence of Irish outlier data points, and it is robust to inverting the Journal of International Business Studies Table A2 Descriptive statistics and correlations Mean SD Min Max 1234567 89 10 1 Contribution to the trade balance -0.119 1.717 -7.328 10.893 2 Balassa index 0.920 0.725 0.000 6.317 0.590 3 Coordination index 1.482 0.244 1.010 1.960 0.020 0.080 4 Radicality 0.029 0.038 0.000 0.417 0.110 0.050 -0.220 5 ln(no. of patents) 3.645 2.253 0.000 9.816 0.260 0.230 -0.180 0.270 6 Corruption 7.602 1.681 2.990 10.000 -0.050 -0.030 -0.150 0.160 0.040 7 Regulatory quality 1.374 0.408 0.258 2.065 -0.060 -0.040 -0.170 0.130 -0.050 0.790 8 EU membership 0.636 0.481 0.000 1.000 -0.010 0.000 0.500 -0.200 -0.290 -0.130 0.110 9 NAFTA membership 0.091 0.288 0.000 1.000 0.020 0.020 -0.470 0.320 0.390 0.140 0.140 -0.420 10 GERD/GDP 1.913 0.810 0.434 4.130 0.160 0.130 0.090 0.170 0.610 0.380 0.120 -0.170 0.130 11 ln(GDP/cap) 10.232 0.214 9.627 10.719 0.110 0.110 -0.020 0.290 0.360 0.500 0.450 -0.210 0.300 0.340 Correlations of abs(0.0403) or greater are statistically significant at the 0.05 level, and of abs(0.0501) or greater, at the 0.01 level. VOC and Comparative Advantage Michael A. Witt and Gregory Jackson Table A3 Regression results (1) (2) (3) (4) (5) Contribution to Contribution to Contribution to Contribution to trade Balassa trade balance trade balance trade balance balance, no outliers Controls ln(number of 0.0807*** 0.0804*** 0.0800*** 0.0775*** 0.000702 patents) (0.0184) (0.0184) (0.0184) (0.0170) (0.00614) Corruption levels -0.00944 -0.0100 -0.00840 -0.00844 0.000640 (0.0189) (0.0189) (0.0189) (0.0174) (0.00647) Regulatory quality -0.0438 -0.0436 -0.0330 -0.0118 0.00733 (0.0748) (0.0752) (0.0753) (0.0690) (0.0263) EU membership 0.103 0.103 0.0954 0.00765 -0.00803 (0.216) (0.223) (0.222) (0.215) (0.0923) NAFTA membership -0.275 -0.283 -0.327 -0.169 0.0623 (0.366) (0.373) (0.373) (0.361) (0.155) GERD/GDP 0.0296 0.0298 0.0308 0.0587 0.0217 (0.0593) (0.0593) (0.0593) (0.0549) (0.0211) ln(GDP/cap) 0.926** 0.904** 0.871** 0.451 0.213 (0.284) (0.286) (0.286) (0.268) (0.113) Year dummies Yes Yes Yes Yes Yes Coordination index 0.00524 0.109 0.101 0.242* (0.281) (0.284) (0.265) (0.107) Radicality 0.399 5.482** 0.587 1.603* (0.289) (2.073) (1.930) (0.699) Coordination 9 radicality -3.432* -0.482 -1.076* (1.386) (1.285) (0.468) Constant -9.760*** -9.552*** -9.394** -5.164 -1.675 (2.859) (2.865) (2.862) (2.692) (1.136) N 2772 2772 2772 2754 2772 AIC 3108.5 3110.6 3106.5 2627.6 -2821.2 BIC 3227.0 3241.0 3242.8 2763.8 -2684.9 Log likelihood -1534.2 -1533.3 -1530.2 -1290.8 1433.6 v 40.98 42.90 49.13 34.53 17.48 Note: Standard errors in parentheses. * p \ 0.05, ** p \ 0.01, ***p \ 0.001. nesting of the HLM specification. Graphing the these results even if the Balassa index represented a interaction showed that the higher Balassa scores of meaningful measure of revealed comparative CME countries decline with increasing radicality, advantage, which we doubt. whereas the lower Balassa scores of LME countries increase with radicality. However, the estimates of Balassa at high levels of radicality have wide confi- ABOUT THE AUTHORS dence intervals, so that we cannot say with confi- Michael A. Witt is Affiliate Professor of Strategy dence that LMEs outperform CMEs at high levels – a and International Business at INSEAD. He earned conservative statement would be to say the disad- his PhD at Harvard University and explores vantages of LMEs are diminished. Looking at other national institutions and their impact on manage- model specifications, the interaction term became rial and organizational outcomes. His papers have statistically insignificant if OLS with clustered stan- appeared in journals such as JIBS, SMJ, JMS, BEQ, dard errors by industry or by country was used. JBE, MOR, and SER. He has published four scholarly Using originality as a measure of radicality resulted books, including The Oxford Handbook of Asian in a statistically significant result with the wrong Business Systems (with Gordon Redding). (positive) sign, while the use of generality produced no statistically significant result. Taking all this Gregory Jackson is Professor of Management at the together, we see at best tentative support of H1 in Freie Universita ¨t Berlin and Einstein Research Journal of International Business Studies VOC and Comparative Advantage Michael A. Witt and Gregory Jackson Fellow at the Berlin Social Science Center. He other third party material in this article are earned his PhD in Sociology at Columbia Univer- included in the article’s Creative Commons license, sity, and his research interests include corporate unless indicated otherwise in the credit line; if the governance, corporate social responsibility and material is not included under the Creative Com- irresponsibility and comparative management. mons license, users will need to obtain permission from the license holder to reproduce the material. To view a copy of this license, visit http:// This work is licensed under a Creative creativecommons.org/licenses/by-nc-nd/3.0/ Commons Attribution-NonCommer- cial-NoDerivs 3.0 Unported License. The images or Michael A. Witt and Gregory Jackson have contributed equally to this article. The authors have a policy of alternating author order of joint papers. Accepted by Jaeyong Song, Area Editor, 18 April 2016. This article has been with the authors for three revisions. Journal of International Business Studies http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of International Business Studies Springer Journals

Varieties of Capitalism and institutional comparative advantage: A test and reinterpretation

Loading next page...
 
/lp/springer-journals/varieties-of-capitalism-and-institutional-comparative-advantage-a-test-A3lHwUv7XZ

References (108)

Publisher
Springer Journals
Copyright
Copyright © 2016 by The Authors
Subject
Business and Management; Business and Management, general; International Business; Management; Organization; Business Strategy/Leadership
ISSN
0047-2506
eISSN
1478-6990
DOI
10.1057/s41267-016-0001-8
Publisher site
See Article on Publisher Site

Abstract

Michael A. Witt and How do national-level institutions relate to national comparative advantages? Gregory Jackson We seek to shed light on this question by exploring two different sets of hypotheses based on the Varieties of Capitalism and other branches of INSEAD, 1 Ayer Rajah Avenue, comparative capitalisms literature. Applying fuzzy-set qualitative comparative Singapore 138676, Singapore; Free University analysis to data from 14 industries in 22 countries across 9 years, we find that of Berlin, Berlin, Germany comparative advantages in industries with radical innovation emerge in specific configurations mixing coordinated and liberal institutional features. Correspondence: Michael A. Witt, INSEAD, 1 Ayer Rajah Institutional comparative advantage in industries with radical innovation may Avenue, Singapore 138676, Singapore. thus be based on the ‘‘beneficial constraints’’ of opposing institutional logics Tel: +65-6799-5388; rather than on the self-reinforcing institutional coherence envisioned in much Fax: +65-6799-5399; of the Varieties of Capitalism literature. By contrast, we find that coordinated e-mail: [email protected] market economies may have comparative advantages in industries with incremental innovation, as envisioned in the Varieties of Capitalism literature. Our article contributes to our understanding of the ‘‘so what?’’ related to capitalist diversity and its implications for location decisions of multinational enterprises. We further present a coordination index going beyond Hall and Gingerich (Br J Polit Sci 39:449–482, 2009) with annual values for 22 OECD countries from 1995 through 2003. Journal of International Business Studies (2016) 47, 778–806. doi:10.1057/s41267-016-0001-8 Keywords: Varieties of Capitalism; economic sociology; comparative advantage; insti- tutional context; fuzzy-set methods The online version of this article is available Open Access INTRODUCTION What explains national comparative advantages? Almost 200 years after the publication of David Ricardo’s On the Principles of Political Economy and Taxation in 1817, the question has not conclusively been answered. Traditional economics explanations, including Ricardo’s and the later Heckscher–Ohlin model of trade, have emphasized the importance of inherited natural endowments with production factors such as labor and land. Empirical performance of these models, however, is weak (Trefler & Zhu, 2000). Received: 7 November 2014 More recent alternative explanations are based on the diverse Revised: 18 April 2016 institutional characteristics of national economies. Comparative Accepted: 18 May 2016 studies have argued that different types of institutions constrain Online publication date: 22 July 2016 VOC and Comparative Advantage Michael A. Witt and Gregory Jackson and enable different forms of economic activity of action, rather than coherence. We hypothesize (Jackson & Deeg, 2008). Most prominent among that certain combinations of liberal market and these has been the Varieties of Capitalism frame- coordinated logics across two or more institutional work (Hall & Soskice, 2001), which contrasted domains may enable institutional comparative ‘‘liberal market economies’’ and ‘‘coordinated mar- advantage by compensating for institutional weak- ket economies.’’ Meanwhile, other theories of nesses inherent in ‘‘pure’’ configurations. For exam- comparative capitalism categorize countries in dif- ple, liberal corporate governance may provide an ferent ways based on diverse governance modes important external monitoring of strongly coordi- (Crouch, 2005; Crouch & Streeck, 1997) or the nated and otherwise insider-oriented governance concepts of National Business Systems (Whitley, institutions (Aoki, 2010). This view posits the 1999). The common objective of this comparative potential for ‘‘beneficial constraints’’ based on capitalisms literature has been to shed light on how institutional arrangements with conflicting logics. the institutional diversity of advanced capitalist The article explores these ideas empirically by economies shapes economic and business comparing the trade patterns of countries during outcomes. the period of 1995–2003. In particular, we use A core argument of the literature is that institu- fuzzy-set qualitative comparative analysis (fsQCA) tions may generate distinct profiles of institutional to test whether different configurations of institu- comparative advantage in production, which mani- tions are sufficient for high performance in differ- fest themselves in nationally distinct patterns of ent sectors, characterized by either radical or economic performance and specialization across incremental forms of innovation. We find that different industrial sectors. For example, Schneider, pure CMEs have comparative advantages in indus- Schulze-Bentrop, & Paunescu (2010) explained rel- tries with incremental innovation. However, we do ative export performance in high-tech and med- not find that LMEs have comparative advantages in ium-tech industries based on different institutional industries with radical innovation. Rather, the characteristics of countries. Rather than a single results suggest that institutional comparative ‘‘best’’ set of institutional arrangements, different advantage involves very specific combinations of types of institutions give rise to distinct forms of both liberal and coordinated types of institutions, comparative advantage. Strong evidence in support which is in line with our own hypotheses. of institutional comparative advantage would have We conclude with a discussion of implications important implications for government policy as and limitations of this study. In particular, our well as the location choices of multinational findings suggest the need to distinguish conceptu- enterprises (MNEs) seeking to avoid home country ally between complementarity and coherence of disadvantages, project competitive advantages institutions and pay greater attention to how related to their home country, or exploit comple- tensions resulting from opposing institutional log- mentary resources and knowledge related to host ics can result in beneficial outcomes. By exploring country institutional environments (cf. Jackson & the linkage between institutional configurations Deeg, 2008; Singh, 2007; Witt & Lewin, 2007). and economic outcomes, our article contributes to In this article we explore the question of institu- our understanding of the ‘‘so what?’’ of variety in tional comparative advantage in relation to two capitalisms. Finally, our analysis shows that a arguments stemming from different branches of number of countries have, over time, diverged the comparative capitalisms literature. First, we test from the commonly held notions about their the well-known but widely contested hypothesis by institutional make-up. Germany, for instance, has Hall and Soskice (2001) that liberal market econo- evolved away from the pure-type coordinated mar- mies (LMEs) have an institutional comparative ket economy that it is commonly believed to advantage in industries featuring radical innova- represent. tion, whereas coordinated market economies (CMEs) have an institutional comparative advan- tage in industries with incremental innovation. LITERATURE REVIEW AND HYPOTHESES This argument posits strong complementarities among relatively coherent sets of institutions that Comparative Capitalisms follow similar logics across all institutional Renewed interest in institutions has led to a large and complex literature on comparative capitalisms domains of the economy. Second, we develop an focused on institutional diversity across national alternative view that is based on conflictual logics Journal of International Business Studies VOC and Comparative Advantage Michael A. Witt and Gregory Jackson contexts (Jackson & Deeg, 2008). Considerable 1999), no index is available. Some efforts have agreement exists that national diversity exists sought to overcome these shortcomings by drawing across several core institutional domains such as directly on the underlying institutional dimensions education and skills formation, employment rela- (e.g., Judge, Fainshmidt, & Brown III, 2014; Sch- tions, financial system, interfirm networks, internal neider & Paunescu, 2012). However, obtaining dynamics of the firm, ownership and corporate institutional measures with construct validity has governance, and the institutions of the state itself proved difficult, especially for countries outside the (Witt & Redding, 2013). Many studies explore how OECD (Witt & Redding, 2013). Together, these such differences cluster into distinct types of insti- obstacles have limited the wider application of tutional configurations. Several competing typolo- comparative institutional analysis in International gies exist. For instance, Hall and Soskice (2001) Business. A substantial research agenda remains to distinguished CMEs and LMEs as two distinct forms understand how institutions shape the comparative of capitalism. Amable (2003) identified five types: institutional advantage of different countries in the market-based, Asian, Continental European, social- world economy. democratic, and Mediterranean. Whitley (1999) As a step to overcoming this challenge, this paper describes six major types of business systems: examines the key arguments from the Varieties of fragmented, coordinated industrial district, com- Capitalism framework that link institutions to partmentalized, state-organized, collaborative, and comparative institutional advantage. Despite much highly coordinated. Looking beyond the advanced criticism (Allen, 2004; Blyth, 2003) and alternative industrialized countries, further types have been theoretical frameworks for comparing institutions proposed (cf. Witt, Kabbach de Castro, Amaeshi, (Amable, 2003; Crouch & Streeck, 1997; Whitley, Mahroum, Bohle, & Saez, 2015). 1999), the Varieties of Capitalism framework con- The challenge of creating theoretically rich and tinues to have a uniquely powerful hold on the valid typologies is complicated by institutional field. Critics have made various important claims change. While most comparative capitalism schol- about the categorization of countries and how this ars reject the possibility of institutional conver- changes over time: the parsimony of Varieties of gence (Hall & Soskice, 2001; Whitley 1999; Capitalism may overlook substantial institutional Yamamura & Streeck, 2003), many types of capi- variety within the LME and CME categories (Yama- talism have undergone substantial liberalization. mura & Streeck, 2003), emerging and transition Thus a growing literature now focuses on the economies cannot be understood using this mechanisms of institutional change (Mahoney & approach (No ¨ lke & Vliegenthart, 2009; Redding & Thelen, 2009; Streeck & Thelen, 2005) and the Witt, 2007), and its emphasis on complementari- diverging trajectories in different types of capital- ties and path dependence systematically overstates ism (Hall & Thelen, 2009; Streeck, 2008; Witt & institutional stability and underestimates institu- Lewin, 2007). tional change (Crouch, 2005). These criticisms Meanwhile, research linking institutional differ- notwithstanding, recent evidence supports the ences to specific economic outcomes has remained notion that advanced industrialized nations do surprisingly underdeveloped. First, a lack of con- indeed cluster very broadly into LMEs and CMEs sensus continues about relevant typologies. Conse- (Witt & Redding, 2013). More importantly, how- quently, scholars have continued to put energy ever, most critiques of Varieties of Capitalism do into criticizing specific typologies (Allen, 2004; not center on its core claims about institutional Blyth, 2003), validating typologies with empirical comparative advantage. data (Brewster, Wood, & Brookes, 2006) or extend- Next, we turn to the main hypothesis of the ing these to new geographic contexts (Bohle & Varieties of Capitalism approach and review empir- Greskovits, 2009; Carney, Gedajlovic, & Yang, ical literature that tests these claims. After this, we 2009). Second, suitable measures of institutional will explore and develop an alternative hypothesis diversity have limited availability. For instance, drawing on some alternative strands of the com- Varieties of Capitalism scholars have only produced parative capitalism literature. an index of institutional diversity for 22 OECD The Varieties of Capitalism Hypothesis: countries for a single time point, the mid-1990s Comparative Advantage Through Coherence (Hall & Gingerich, 2009). No time series is avail- In developing the Varieties of Capitalism approach, able, nor does the index cover countries outside the Hall and Soskice (2001) proposed that the advanced OECD. For the business systems approach (Whitley, Journal of International Business Studies VOC and Comparative Advantage Michael A. Witt and Gregory Jackson industrialized societies fall into two main types: manifest in comparative strengths and weaknesses LMEs and CMEs. Firms in LMEs tend to rely more in different types of industries. By contrast, coun- on liberal market mechanisms, while firms in CMEs tries with lower levels of coherence – ‘‘mixed types’’ tend to coordinate business transactions through – are argued to lack the institutional complemen- non-market relationships. The prototypical LME, tarities needed for institutional comparative such as the US, features a market-driven financial advantage. system, flexible use of external labor markets, Based on this logic, Hall and Soskice (2001) generalist education and training systems, low proposed that LMEs and CMEs would show distinct levels of networks and alliances among firms, and patterns of institutional comparative advantage for management-driven, top-down decision-making radical or incremental innovations. Hall and Sos- structures inside firms. The classic CME, such as kice (2001: 38f.) defined radical innovation as Germany around 1995, has a bank-led financial ‘‘entail[ing] substantial shifts in product lines, the system providing patient capital, stronger internal development of entirely new goods, or major labor markets based on employment protection, changes to the production process,’’ while incre- skills formation systems conducive to the develop- mental innovation is ‘‘marked by continuous but ment of specialized skills, high levels of networks small-scale improvements to existing product lines and alliances among firms, and consensual deci- and production processes.’’ They argued that the sion-making inside firms bringing together man- combination of patient capital, long-term employ- agement and labor. These institutional differences ment, and firm-specific skills in CMEs would enable encourage firms to invest in and utilize transferable more efficient production in industries with incre- assets to a greater extent in LMEs, where institu- mental patterns of innovation because the relative tions do not bind economic actors to long-term immobility of labor and capital in CMEs enabled commitments but support their ‘‘keeping options and constrained firms to focus their efforts on open’’ to using the external market. By contrast, the improving existing lines of products. Meanwhile, use of relational assets in CMEs implies investments fluid capital markets with short-term employment whose value is specific to the continuation of long- and general skills in LMEs would enable more term relationships among company stakeholders. efficient production in industries with radical pat- Such investments require different sorts of institu- terns of innovation, as these conditions support tional support – such as protection of investments firms using external markets to mobilize risky in firm-specific human capital and contracting equity finance and workers with different skill sets, arrangements, or mechanisms to govern collective and thereby take advantage of new technological action problems (e.g., free rider problems) and breakthroughs. Since comparative advantage is support wider patterns of cooperation across a defined as the ability to make one product more network or industry. efficiently than another, this suggests the presence The Varieties of Capitalism framework posits that of comparative advantage in industries drawing on LMEs and CMEs exhibit strong complementarities incremental innovation in CMEs and radical inno- by being organized around a coherent institutional vation in LMEs. By the Law of Comparative logic (Deeg, 2007). Complementarities exist where Advantage (Dearsdorff, 1980), this advantage features of institutional structures reinforce each should be visible in international trade patterns. other by mutually generating increasing returns. This leads to the following hypothesis: Meanwhile, coherence is present where institutions H1 LMEs show trade patterns consistent with follow the same market or coordinated logic across institutional comparative advantage in industries domains of the economy, such as corporate gover- featuring radical innovation, whereas CMEs show nance and employment relations. Since coherent trade patterns consistent with institutional com- institutions are hypothesized to uniformly support parative advantages in industries with incre- either transferrable or relational assets, the Varieties mental innovation. of Capitalism approach claims that institutions in different domains will complement one another by Prior Tests mutually reinforcing such investments. Based on Despite its centrality, some fifteen years after the their capacity to enable these different kinds of publication of Hall and Soskice’s piece, H1 remains investments, Varieties of Capitalism argues that open territory. Attempts to test it have faced several complementary sets of institutions favor certain interrelated conceptual and methodological patterns of economic activity over others that Journal of International Business Studies VOC and Comparative Advantage Michael A. Witt and Gregory Jackson challenges. First, mapping institutional diversity Paunescu (2012) used trade outcomes to claim onto the two LME and CME categories is less qualified support of the Hall and Soskice hypoth- straightforward than commonly assumed (see more esis. Both papers relied on exports as a measure of generally Jackson & Deeg, 2008). Different studies comparative advantage. However, as the case of use different sets of indicators, or benchmark electronics in China illustrates (e.g., Redding & countries against assumed ideal–typical countries Witt, 2009), high levels of exports in an industry like Germany for CMEs (Allen & Aldred, 2009; may occur on the back of high levels of imports of Schneider et al., 2010). Second, the concept of high value added inputs in the same industry, in radical and incremental innovation has proven which case only a small proportion of the value of slippery. Operationalizing radicality requires a valid the exports in these industries are produced in the way of classifying industries according to different respective country. These imports need taking into types of innovation. Most existing studies have account to establish whether a comparative advan- relied on categorizations supplied by the OECD to tage exists. The two papers also relied on OECD describe high-tech or medium high-tech industries. classifications of research intensity to express rad- This classification is based on R&D intensity, which icality. However, radicality is an output of research is a measure of investment or inputs into the activity, while research intensity is an input mea- innovation process rather the output of actual sure based on R&D spending relative to industry innovation. Third, empirical work has focused size. In sum, prior works have either sought to either on the influence of institutions on innova- explain innovation patterns rather than compara- tion patterns or the influence of institutions on tive advantage, or suffered from issues with con- export success, but missed what we see as the core struct validity of key variables. As will be explained of the Varieties of Capitalism hypothesis – namely, in the methodology section, this article will offer a the idea that the influence of institutions on trade more comprehensive test of H1 than the previous outcomes is mediated by or contingent upon types literature. of innovation. Cross-country differences exist in Alternative Hypothesis: Complementarities the patterns of radical or incremental innovation Through Beneficial Constraints (Furman, Porter, & Stern, 2002), and this capacity The Varieties of Capitalism approach has assumed does relate to institutional features of the economy, that complementarities require coherence of insti- such as employment relations (Bassanini & Ernst, tutions (e.g., pure LME types, where all institutions 2002). However, we see this as a separate debate. In are liberal). The underlying logic stems from the our view, the core of the Varieties of Capitalism concept of complementarities as developed in hypothesis actually revolves around comparative economics (Milgrom & Roberts, 1994, 1995), advantage in products as the dependent variable, whereby investments in transferable assets in one rather than R&D spending or patents. The Varieties institutional domain are complemented by trans- of Capitalism argument stresses the ability of firms ferable assets in other domains. For example, to draw on these innovations in the productive portable skills are expected to increase in value for process, not their ability to make radical innova- workers if firms also have corporate governance tions. Firm-level studies using both approaches arrangements whereby capital investment is shifted have found these to be potentially complementary, rapidly from one firm to another. Conversely, long- but in fact very distinct, phenomena (Herrmann & term relational employment relationships are pre- Peine, 2011). sumed to benefit from ‘‘patient capital’’ of coordi- To our knowledge, only four major published nated blockholders, whose long-term strategic studies have attempted to explicitly test the Vari- investment in the firm will protect the firm-specific eties of Capitalism hypothesis. Taylor (2004) and skill investments of workers. Akkermans, Castaldi, & Los (2009) examined inno- Based on this reasoning, Varieties of Capitalism vation patterns as visible in patent data. While implies a very restrictive view of how institutions Taylor found no support of the hypothesis, Akker- combine. Comparative advantage relates to the mans and colleagues claimed qualified support. coherence of incentives provided by the set of However, both papers tested only the ability of institutions within a country, and thus comple- economies to produce radically innovative patents mentarities only form when institutions cluster and not actual institutional comparative advan- into only two ideal–typical patterns. We believe tage, which is the outcome of interest in this paper. that these assumptions about complementarities Schneider et al. (2010) as well as Schneider and Journal of International Business Studies VOC and Comparative Advantage Michael A. Witt and Gregory Jackson are problematic and at odds with empirical work To elaborate this idea, we focus on the two most linking comparative advantage to more complex or salient stakeholders of the firm: shareholders and hybrid combinations of coordinated and liberal employees. Here we see a potential counterbalanc- institutions (Boyer, 2004). We thus draw on a ing role to be played by each of these groups. stream of the comparative capitalisms literature Shareholders may benefit from liberal corporate grounded in economic sociology to develop an governance and establish this as a counterweight to alternative set of hypotheses regarding comple- more coordinated institutions. Likewise, employ- mentarities based on opposing institutional logics. ees may favor more coordinated employment Our intuition here is to see complementarities as relations and benefit from coordination as a coun- essentially grounded within political projects. terweight to otherwise liberal institutions. These Crouch (2005) argued that institutions may have mixed configurations reflect a compromise or positive effects by compensating for the weaknesses balancing between opposing logics, whereby the of other institutions. For example, political systems interests of one key stakeholder are addressed often utilize competing institutional logics that within an overall context that is more hostile to serve as checks and balances on one another – such their interests. as when the rule of law in the judiciary acts to In developing these hypotheses, we turn to the balance the logic of majority rule in the legislature. concept of ‘‘beneficial constraints’’ in economic Along these lines, sociological theories of compar- sociology (Streeck, 1997). Streeck argues that ative capitalism have observed that opposing logics socially institutionalized constraints on the or principles of social organizations may be useful rational voluntarism associated with markets may complements (Crouch, Streeck, Boyer, Amable, be economically beneficial. Normative constraints Hall, & Jackson, 2005;Ho ¨ pner, 2005). Here the and social obligation are needed to correct market political economy of capitalism involves not simply failures, and thus also enable economic actors to a choice between coordination or liberal markets, develop and protect collective goods or commit- but reflects political coalitions and compromises ments to stakeholders that would otherwise fall that seek a balance between them. The interests victim to hyper-rational economizing. Rather than and relative powers of different actors groups, allowing economic actors to pursue their given particularly corporate stakeholders, come into play. preferences, institutions may transform the identi- Liberal markets produce winners and losers, also ties and interests of actors by constraining their generating counter-movements seeking to contain pursuit of short-term aims and leading them to the influence of markets through efforts of coordi- search for new and alternative strategies that are nation and social protection (Block & Somers, legitimate in their institutional context. In doing 2014). In terms of the CME and LME typology, this so, institutional constraints can turn institutions perspective implies that mixed cases or hybrids may into beneficial resources by simultaneously have beneficial characteristics. For example, Boyer enabling new forms of coordination and coopera- (2004) showed that Nordic economies achieved tion, thus potentially attaining higher levels of economic growth in emerging ITC sectors based on economic performance than might be possible in a specific combination of liberal markets and their absence. By constraining certain market coordinated institutions. However, whether or not behaviors, institutions may help create or protect mixed cases work well together depends on the critical resources based on coordinated investments extent to which the specific combination of market by stakeholders. As Streeck argued: ‘‘Beneficial and coordinated logics across two or more domains constraint…is a dialectical concept, suggesting a may help compensate for one another. Unfortu- relationship of both mutually subversive and mutu- nately, the comparative capitalism literature offers ally supportive conflict between the economic only scant theorizing about specific hybrid and the social, ruling out any lasting harmony combinations. between the two’’ (Streeck, 1997: 207). Importantly, Based on these considerations, we develop two this view of institutions goes beyond the notion hypotheses whereby coordinated institutions help of property rights or embeddedness in social net- to prevent or remedy market failures in liberal works that facilitate market exchange. Rather, settings or, following the inverse logic, liberal institutions act as constraints on markets, on market institutions help to remedy problems of the one hand, but serve to stabilize them, on the rigidity in highly coordinated settings. Here oppos- other – thus suggesting their Durkheimian charac- ing institutional logics complement one another. ter (Jackson & Muellenborn, 2012). Journal of International Business Studies VOC and Comparative Advantage Michael A. Witt and Gregory Jackson Viewed from this perspective, the coherent set of institutions that regulate pay at sectoral or even market-oriented institutions characterizing LMEs national level. Likewise, Japanese firms utilize may have inherently self-destructive tendencies. As strong norms of lifetime employment for their core markets across different domains of the economy workforce while limiting external labor markets to reinforce one another by creating incentives for a very narrow range of specialists or to temporary greater market-orientation, these markets may employees. While economists usually see these become overheated or hyper-rational. Time restrictions as politically motivated, by the logic horizons may become too short and market failures of beneficial constraints, they may be economically may occur in relation to long-term financial invest- beneficial by providing an institutional basis on ments, development of cooperative supplier which companies can develop and retain necessary relationships, or fostering of firm-specific skills in resources. productive capabilities (Campbell, 2011; Jackson & Rather than complementarities based on coher- Petraki, 2010). Positively, firms in LMEs may ence, we expect that firms facing radical innovation respond quickly to radical innovations by shifting in countries with liberal market institutions in their investments in markets for capital and labor. most domains may benefit from complementary However, left unconstrained, firms may also be institutions that support coordination in employ- likely to focus on predatory responses, such as ment relations. A combination of liberal and coor- using mergers and acquisitions to obtain new dinated institutions may be beneficial for dealing technologies or protect market shares or aggres- with radical innovation by providing for both sively restructure corporate hierarchies by shedding flexibility in restructuring of economic organiza- employees and boosting shareholder returns. As a tion as well as trust and coordination to solve result, firms may be unable to capitalize on radical problems of asymmetric information or hold-up. innovations by developing related firm-specific Applying the same logic as Hall and Soskice (2001) advantages that lead to sustained comparative linking these benefits to comparative advantage, we advantages, as suggested by the resource-based thus posit the following: view and other theories (Newbert, 2007). H2 Countries with coordinated institutions in These deficits of purely liberal markets are likely employment relations and liberal institutions in to be most acute in the area of employment other domains will show trade patterns consis- relations. Employee skills and organizational capa- tent with comparative advantage in industries bilities based upon them are essential for compar- with radical innovation. ative advantage during times of discontinuous change (Aoki & Jackson, 2008; Rajan & Zingales, Beneficial constraints also extend to a second 2000). In particular, firms entering and expanding argument. Similar to the inherent problems with new lines of business as a result of radical innova- unconstrained market-oriented transactions, high tion will need skilled employees with sufficient levels of coordination may lead to an overextension understanding of the workings of the firm (i.e., of relational logics. To the extent that coordination firm-specific skills) to capitalize on the opportuni- becomes strongly self-reinforcing, firms may lack ties offered by new technology. While active exter- incentives to take risks and enter into new lines of nal labor markets may help firms gain access to new business – employment is long-term, firm-specific skills, left fully to the logic of the market, firms are skills are well developed, workers participate in firm also likely to suffer from substantial problems to decision-making, inter-firm relations rely on stable partners, and owners take a long-term strate- retain key employees through periods of dynamic change. Economic sociology has thus suggested gic view. These patterns may be beneficial in that markets need to be embedded within social stable industry environments characterized by low relations, not only for moral or political reasons, levels of innovation or even incremental patterns, but for economic ones. Institutional constraints as hypothesized by Varieties of Capitalism. How- ever, it seems unlikely to these institutional imposed by coordination may be beneficial for markets precisely because they curtail the oppor- arrangements can support adjustments to more tunism and hyper-rationality that lock actors into radical and discontinuous forms of innovation potential market failures due to collective action (Witt, 2006). problems. Many countries in Europe thus restrict The potential deficits of high coordination may create particular problems for shareholders, who competition between firms based on wages through Journal of International Business Studies VOC and Comparative Advantage Michael A. Witt and Gregory Jackson are disadvantaged by low competitiveness. Conse- METHODOLOGY, DATA AND CALIBRATION quently, we see more liberal forms of corporate Methodology: fsQCA governance as having potential to exert a counter- We used fsQCA in Stata (StataCorp, 2013)as balancing effect on high coordination and thus to provided by Longest and Vaisey (2008). Qualitative represent an economically beneficial compromise comparative analysis draws on Boolean algebra to between competing logics. Here firms benefit from determine which configurations of causal condi- high coordination across different domains, includ- tions are related to an observed outcome (Ragin, ing employment relations, and develop strong 2000). The method is increasingly used in the capabilities based on their human assets and rela- business and management literature (e.g., Bell, tional linkages to other firms. However, a literature Filatotchev, & Aguilera, 2013; Crilly, 2011; Fiss, on hybrid forms of corporate governance suggests 2011; Judge et al., 2015; Schneider et al., 2010). that increasing pressures from capital markets may We adopted fsQCA because of its distinct advan- lead stakeholder-oriented firms to take bolder and tages for studying institutional complementarities faster adjustments to changes in technology and (Jackson & Ni, 2013). First, fsQCA is well suited to product markets (Vitols, 2002, 2004). For example, capture conjunctural types of causation that under- Aoki (2010) argues that external monitoring of lie theories of complementarity. Complementari- internal linkages of corporate organization may be ties are based on specific configurations where beneficial in giving corporate insiders important multiple causal factors combine to produce an signals about their overall business strategy and outcome. Traditional statistical methods model help benchmark their efforts against competitors. ‘‘net effects’’ of single variables, while holding other Unlike monitoring by ‘‘patient’’ investors, a more factors constant. While interaction effects can liberal market-oriented process may be a beneficial capture specific conjunctions, such as radical inno- resource for investors seeking to constrain the vation with institutional coordination, these meth- tendencies toward stagnation among corporate ods are not well suited to more complex insiders, such as employees and managers. While conjunctions. Consequently, most work on com- commitments to corporate stakeholders remains parative capitalism has used factor analysis to strong, some pressure from capital markets and simplify institutional configurations into highly influence from owners may help to catalyze more simplified one-dimensional constructs (Hall & Gin- dynamic, albeit ‘‘negotiated’’ responses to radical gerich, 2009) or group countries into a limited innovations through more discontinuous manage- number of types (Amable, 2003), at the risk of ment strategies or changes in production technolo- losing important distinctions in the process. fsQCA gies. Indeed, a growing literature on the relation does not require such simplifying steps. Second, between corporate governance and employment fsQCA can identify how multiple different combi- patterns suggest more complex relationships nations of independent variables can produce the between these variables than found in the Varieties same outcome (equifinality). Third, fsQCA evalu- of Capitalism literature (Gospel & Pendleton, ates set theoretical relationships, rather than corre- 2003, 2005;Ho ¨ pner, 2005). Despite their potential lations among different factors. This under- to destabilize commitments to stakeholders, liberal appreciated facet allows for the possibility of market institutions in corporate governance may asymmetrical relationships, whereby high and low also potentially help counterbalance risk averse values of the outcomes are driven by different tendencies of corporate insiders, thus paradoxically causal conditions. For example, if low coordination helping to catalyze their dynamic potential in and radicality are jointly sufficient for high perfor- terms of functional flexibility and ability to foster mance outcomes, it does not follow that coordina- trust, but here in the service of adjustments to tion is sufficient for outcomes of low performance. radical innovation. Applying the same logic about Finally, the method is not sensitive to outliers the linkage to comparative advantage and trade, we because it does not assume an underlying proba- thus posit: bility distribution (Fiss, 2011). H3 Countries with liberal institutions in corpo- rate governance and coordinated institutions in Sample and Data Structure other domains will show trade patterns consis- Our dataset consists of 2772 observations, made up tent with comparative advantage in industries of 14 industries at the 2-digit ISIC-3 level in 22 with radical innovation. OECD countries, with annual observations from Journal of International Business Studies VOC and Comparative Advantage Michael A. Witt and Gregory Jackson 1995 through 2003. Each case is a particular to other industries. This implies a need to compare industry for a given country and year. The choice within countries, as our measure does, rather than of countries, industries, and years was affected by across countries, as the Balassa index does. For data availability as explained below. example, assume that country A has two industries, X and Y, and is better at producing goods in these industries than any other country in the world. Outcome Measure: Contribution to the Trade This would give it a competitive (absolute) advan- Balance Our outcome measure was the contribution to the tage relative to other countries, which the Balassa trade balance for manufacturing industries. It is index would pick up as an advantage in both industries. However, if country A is better at based on the ‘‘Law of Comparative Advantage’’ in producing X than at Y, we can say that it has a economics (Dearsdorff, 1980), which states that comparative advantage at X over Y. The Balassa comparative advantages in manufacturing should be revealed in trade patterns. Much of the eco- index would miss this, while our preferred measure nomics literature has measured revealed compara- would pick this up. For robustness testing, we will later check our results against those of the Balassa tive advantage using the Balassa (1965) index, index. which compares a country’s export market share In choosing our industries, we omitted low in a given industry with the average export market share of all countries. We followed an alternative technology industries, such as food and beverages, calculation used by the OECD (2003) that we textiles, and wood and cork. Low technology industries are defined by the lowest research activ- believe to have higher construct validity, as we will ity, which implies a weak link between innovation explain below. and resultant comparative advantage. Our analy- For each country, the contribution to the trade balance (CTB in the formula) in the OECD concep- sis thus focused on low-medium technology, tualization and as used in this article is computed as medium–high technology, and high technology sectors. Since the contribution to the trade balance follows: is likely to manifest itself with a delay after an ðÞ X þ M i i invention occurred, our analysis used a lead-time of CTB ¼ðÞ X  MðÞ X  M ; ð1Þ i i ðÞ X þ M one year relative to the causal conditions (i.e., the contribution to the trade balance in year 2004 will where (X - M ) is the observed trade balance in i i be analyzed with respect to institutional factors for industry i in the given country and the remainder 2003). for the formula is the theoretical trade balance for this same industry i in the same country (OECD, Causal Conditions: Indicators of Radicality 2003: 150). To make the results comparable across and Institutional Coordination countries, they are expressed a percentage of total trade for the respective country. A positive value Indicators of Radicality denotes the presence of a comparative advantage, a Following Akkermans et al. (2009), we calculated negative value, of a comparative disadvantage for three distinct, patent-based measures to gauge the that country in a particular industry (OECD, 2013). extent of radical innovation in a given society. In We obtained these data from the OECD (2013). line with the prior literature, incremental innova- This measure improves over the Balassa index in tion is operationalized as the inverse of radical two important ways. First, it takes into account the innovation – innovations that are not radical are level of imports in a given industry. While the assumed to be incremental. While we will later Balassa index focuses entirely on exports and thus question this assumption, we retained it since no cannot distinguish whether a country has a high established measure of incremental innovation export market share because it produces much of exists in the literature. the value-added itself (a sign of comparative advan- In our first step, we obtained all entries from tage) or because it imports and re-exports (as in the 1990 through 2005 of the NBER Patent-Citations case of China already discussed), our measure can Data File (Hall, Bessen, & Thoma, 2011). This distinguish between these two cases. Second, the database contains the patent number and three Balassa index is a measure of competitive advantage indicators of radicality for each US patent issued rather than comparative advantage. Comparative during this period: number of citations received, advantage exists within each country with respect generality, and originality. The number of citations Journal of International Business Studies VOC and Comparative Advantage Michael A. Witt and Gregory Jackson received operationalizes the concept of radicality contents of the NBER database with the PATSIC- by measuring the impact on future technological CONAME database using patent numbers, which development, as expressed in subsequent citations are contained in both databases. by other patents. Generality refers to the propensity We calculated the relative proportions of radi- of basic innovation to diffuse to many different cally innovative patents by industry and country industries and areas of technology, that is, their for each of the three measures of radicality. Similar versatility. Originality expresses the idea that truly to Akkermans et al. (2009), we define a patent to be radical innovations are likely to cite patents from radically innovative if it scores higher than the different areas of technology. Further explanations 95th percentile for a given measure. For instance, if of these measures are contained in the works of the 95th percentile for citations received in the Hall, Jaffe, & Trajtenberg (2001) and Akkermans pharmaceuticals industry was 15, we would count et al. (2009). as radical all patents with more than 15 citations For all three measures, the innovation literature received. We consequently calculated the threshold has interpreted higher numbers to indicate higher separately for each year because the year of publi- levels of radicality (Akkermans et al., 2009). We cation of a patent affects how many citations it has believe that of the three measures, the number of received, with younger patents on average receiv- citations received is likely to be closest to a valid ing fewer citations than older ones. indicator of radicality. Similar to citations to In the final step, we calculated the proportion of scholarly articles, citations to patents indicate their patents above the threshold in each country, impact. Just as highly cited articles are relatively industry and year. To generate a more stable picture, more likely to be in some way seminal than less we assigned to each observation the three-year cited pieces, patents with higher citations counts average of these proportions from t - 2 years to t. are likely to have a greater impact on the course of We discarded data from 2004 and 2005 because the technology development and thus represent radical database ends in 2005, which means that there was steps in technological development. We are less insufficient time for citations from these two years convinced of the construct validity of the other two to receive sufficient numbers of citations for dis- measures. Breakthrough ideas in one area do not criminating between radically and incrementally necessarily have to be of interest to other areas, nor innovative patents. Given the strong interest in do they need to assemble insights from many innovation patterns in the Varieties of Capitalism different preexisting areas. Despite our doubts, we debate, Table 1 shows the cross-tabulation of rad- retained both generality and originality for robust- icality, using the citations received indicator, by ness tests. country and industry. The raw data required two further steps for our Institutional Indicators purposes. First, the generality and originality mea- The starting point in understanding institutions sures are defined as Herfindahl indexes and there- was past work categorizing countries into either fore not given for patents that received no citations LME or CME types of capitalism. However, different or contained no citations. In these cases, we studies based classifications on different institu- assigned a value of 0 for each measure. Second, tional domains and have made different assump- both measures are biased downward for patents tions about how institutions combine into these involving small numbers of citations (Hall, Jaffe, & broader types. For example, Hall and Gingerich Trajtenberg, 2001). We corrected for this bias as (2009) used confirmatory factor analysis to derive a proposed in Hall, Jaffe, & Trajtenberg (2001). coordination index for 20 advanced industrialized Since our outcome measure is categorized by economies based on three corporate governance industries following the ISIC 3 classification and three employment relations variables. How- scheme, we next matched each patent with one ever, this approach covered only two of the five or several ISIC codes. The United States Patent and institutional domains discussed by Varieties of Trademark Office (USPTO) assigns each patent to Capitalism (Hall & Soskice, 2001). In addition, the one or several industries to which the patent is construction of a single index assumes that com- likely to be relevant. While the original assignment plementarities are achieved through cohesion (all scheme is incompatible with ISIC classifications, domains being either fully coordinated or fully the USPTO identifies the SIC codes for each patent liberal), an assumption this paper seeks to relax and in its PATSIC-CONAME database. We obtained the test (H2, H3). year 2008 version of the database and matched the Journal of International Business Studies VOC and Comparative Advantage Michael A. Witt and Gregory Jackson Table 1 Proportion of radical patents by citations received, by country and industry, 1995–2003 Industry AT AU BE CA DK FI FR DE GR IE IT Aircraft, spacecraft 0.012 0.051 0.056 0.064 0.127 0.000 0.007 0.043 0.000 0.000 0.031 Basic metals 0.000 0.094 0.019 0.027 0.000 0.000 0.034 0.036 0.000 0.056 0.058 Chemicals other than pharmaceuticals 0.020 0.037 0.020 0.046 0.027 0.032 0.028 0.030 0.000 0.105 0.018 Electrical machinery 0.007 0.046 0.034 0.095 0.008 0.034 0.026 0.020 0.000 0.016 0.019 Machinery 0.017 0.036 0.028 0.053 0.015 0.023 0.017 0.022 0.000 0.042 0.012 Medical, precision, optical equipment 0.007 0.042 0.009 0.052 0.043 0.015 0.019 0.018 0.000 0.099 0.008 Metal products 0.039 0.038 0.038 0.066 0.025 0.016 0.030 0.029 0.000 0.000 0.024 Motor vehicles 0.041 0.018 0.028 0.084 0.028 0.019 0.024 0.037 0.000 0.053 0.020 Non-metallic products 0.006 0.079 0.016 0.067 0.023 0.026 0.030 0.023 0.000 0.000 0.018 Office, accounting, computing equipment 0.000 0.051 0.026 0.065 0.000 0.026 0.011 0.010 0.000 0.018 0.006 Other transport equipment 0.015 0.037 0.000 0.041 0.000 0.000 0.024 0.060 0.000 0.000 0.048 Pharmaceuticals 0.018 0.022 0.018 0.043 0.023 0.037 0.025 0.021 0.000 0.100 0.013 Radio, TV, communication equipment 0.008 0.048 0.023 0.072 0.002 0.066 0.019 0.015 0.000 0.044 0.004 Rubber, plastic 0.036 0.038 0.024 0.059 0.031 0.026 0.033 0.027 0.000 0.035 0.027 Average 0.016 0.046 0.024 0.060 0.025 0.023 0.023 0.028 0.000 0.041 0.022 Industry JP NL NO NZ PO KR ES SE CH UK US Aircraft, spacecraft 0.073 0.026 0.000 0.000 0.000 0.008 0.000 0.021 0.002 0.015 0.065 Basic metals 0.038 0.073 0.000 0.000 0.000 0.010 0.111 0.000 0.037 0.055 0.081 Chemicals other than pharmaceuticals 0.031 0.027 0.032 0.033 0.037 0.025 0.035 0.052 0.041 0.033 0.072 Electrical machinery 0.038 0.020 0.021 0.082 0.000 0.021 0.022 0.021 0.015 0.035 0.076 Machinery 0.057 0.023 0.015 0.009 0.000 0.025 0.018 0.018 0.014 0.029 0.077 Medical, precision, optical equipment 0.017 0.033 0.024 0.005 0.000 0.010 0.000 0.022 0.029 0.025 0.074 Metal products 0.054 0.038 0.006 0.000 0.000 0.034 0.024 0.031 0.022 0.035 0.077 Motor vehicles 0.049 0.014 0.065 0.000 0.000 0.010 0.016 0.032 0.060 0.022 0.078 Non-metallic products 0.042 0.052 0.000 0.000 0.000 0.033 0.009 0.063 0.024 0.026 0.080 Office, accounting, computing equipment 0.011 0.021 0.139 0.037 0.000 0.008 0.037 0.038 0.005 0.027 0.077 Other transport equipment 0.076 0.000 0.000 0.000 0.000 0.004 0.000 0.007 0.000 0.031 0.071 Pharmaceuticals 0.011 0.055 0.093 0.028 0.000 0.019 0.025 0.057 0.052 0.045 0.074 Radio, TV, communication equipment 0.026 0.015 0.058 0.019 0.000 0.021 0.022 0.058 0.014 0.035 0.074 Rubber, plastic 0.039 0.036 0.005 0.000 0.000 0.033 0.034 0.063 0.034 0.030 0.080 Average 0.040 0.031 0.033 0.015 0.003 0.019 0.025 0.034 0.025 0.032 0.075 We thus considered all five domains: corporate original indicator derived by LLSV (La Porta, Lopez- governance, inter-firm relations, hierarchies within de-Silanes, & Shleifer, 1999) and an updated mea- firms, employment relations, and education. For sure for 2003 (Botero, Djankov, La Porta, Lopez-de- each institutional domain, we conceptualized of Silanes, & Shleifer, 2004), interpolating annual institutions as varying along a single dimension values based on the movement between these two spanning between liberal institutions characterized periods. Shareholder protection indicates the by a high use of market relationships or transferable degree to which business firms are likely to have a assets and coordinated institutions characterized by liberal orientation toward shareholder control. high use of long-term strategic relationships or Second, we obtained stock market valuation as a relationship-specific assets. We adopted a separate percentage of GDP from OECD statistics to capture indicator for each and later will explore the differ- the salience of the stock market for company ent ways in which they combine into a number of financing. Third, we measured the dispersion of complex configurations. share ownership by the percentage of large firms in In terms of corporate governance, we used three each country that have a blockholder with an indicators. First, an index of legal rules for share- ownership stake of 10 % or greater. We used the holder protection measures whether laws in a LLSV indicator of ownership dispersion based on country adopt particular protections related to the percentage of large firms without an owner disclosure, voting rights, and so forth (La Porta, holding a 10 % stake or more, and calculated our Lopez-de-Silanes, Schleifer, & Vishney, 1998). Since own data for 2003 using ownership data from the no alternative suitable measures exist, we took the ownership module of Thomson Banker One. We Journal of International Business Studies VOC and Comparative Advantage Michael A. Witt and Gregory Jackson interpolated annual data based on the year-to-year Thus we examined only the proportion of employ- changes between these two dates. High blockhold- ees with short job tenure of less than one year. ing is associated with coordination. While all countries have some new hires, a high For inter-firm relationships, we used two indica- proportion of new hires indicate a more liberal tors based on merger and acquisition activity from market-oriented to employment practices, whereas the Capital IQ database. First, the number of a low proportion indicates that employment may merger and acquisition deals by acquiring firms in be more long-term and coordination. Second, we each country indicates the use of the market in examined the strictness of employment protection governing relationships between firms. While we for regular employees based on well-known OECD lack internationally comparable measures of coor- indices. Greater difficulty in firing workers indicates dination such as that achieved through long-term again a higher degree of coordination in employ- relational contracts in buyer–supplier relations, ment relations. Finally, we measured the degree of prior literature (Schneider & Paunescu, 2012) sug- coordination in wage bargaining across firms, gested that mergers and acquisitions indicate that industries and national levels. This indicator shows firms seek to obtain assets from target firms the degree to which firms engage in individualized through a market-driven acquisition of ownership market transactions in setting wages, or whether rights. Second, the proportion of merger and these are coordinated with other employers in the acquisition deals that take the form of a full-scale same industry or even nationally (Visser, 2011). merger indicates the use of strong market relation- Finally, we used two indicators to examine the ships. Mergers and acquisitions may involve arms’ characteristics of education and training institu- length transactions, where control is purchased and tions. We drew on OECD statistics on the propor- leads to the dissolution of a target firm, or acqui- tion of graduates from different types of sitions of minority ownership stakes. In the latter educational institutions within the wider popula- case, the target firms remain legally independent tion. First, we took the share of graduates from but enter into a coordinated long-term relationship upper secondary or post-secondary non-tertiary as with the acquiring firm. a broad indicator for occupationally-based voca- For firm hierarchies, we examined the degree of tional training (Schneider et al., 2010). Second, we employee participation at the level of corporate used university training, as measured by the boards and through works councils and similar number of graduates from university as a share labor-management consultation bodies. Board- of the population in the typical graduation age level employee representation has been categorized range, as an indicator of general training (Schnei- based the percentage of employees on the board der et al., 2010). Taken together, these indicators and on whether codetermination requirements show the importance of two pathways toward apply broadly to all large private companies, are skilled workforces based either on more occupa- restricted to public corporations, or do not exist at tionally based and thus relationship-specific train- all (Jackson, 2005). Strong legal rights to employee ing in particular occupations, which reflect higher representation on boards indicate long-term rela- degrees of coordination, or on university-based tional coordination of firm operations between education that offers more general set of skills that management and employees. In addition, many tend to be portable across firms or very broad countries anchor employee participation at the range of jobs according to individual career level of the establishment or even workplace in trajectories. the form of works councils. We measured work Table 2 presents an overview of the descriptive council rights by the rights to information, consul- statistics for all indicators used in the study. tation or even codetermination of managerial Calibration decisions, as well as whether these rights pertain fsQCA requires calibration of all data into set to social or also economic matters of the enterprise membership scores that fall between 0 (full absence (Visser, 2011). of the given causal condition) to 1 (full presence) In terms of employment relations, we used three (Ragin, 2008). Calibration combines information indicators. First, we considered the duration of about qualitative differences in kind (e.g., countries tenure with a particular employer. Unfortunately, with or without employee participation in firm existing data from the OECD do not permit detailed hierarchies) with information about differences in cross-national comparisons of average job tenure or degree (e.g., different degrees of participation). the proportion of employees with very high tenure. Journal of International Business Studies VOC and Comparative Advantage Michael A. Witt and Gregory Jackson Table 2 Descriptive statistics Indicator N Mean SD Min Max 1 Shareholder protection 198 3.24 1.1 0 5 2 Dispersion of control 198 0.33 0.24 0 0.9 3 Size of stock market 198 78.5 54.19 8.92 317.03 4 Mergers and acquisitions 198 0.24 0.27 0 1.1 5 Mergers 198 0.38 0.19 0 1 6 Wage coordination degree 198 0.18 0.06 0.08 0.41 7 Short-term employment 198 3.07 1.25 1 5 8 Employment protection 198 2.06 0.86 0.17 4.33 9 Board-level codetermination 198 0.43 0.39 0 1 10 Works council rights 198 1.19 1.1 0 3 11 Occupational training 198 43.89 19.25 2 82 12 University training 198 29.5 9.7 9 53.3 13 Radicality, citations received 198 0.03 0.04 0 0.42 16 Radicality, generality 198 0.04 0.04 0 0.48 17 Radicality, originality 198 0.07 0.06 0 0.67 14 Contribution to the trade balance 198 -0.12 1.72 -7.33 10.9 15 Balassa index 198 0.92 0.73 0 6.32 Ragin (2008) stresses using external benchmarks set of countries with coordinated corporate gover- rather than sample variance in defining thresholds nance was based on three indicators: the absence of for set membership. Rather than assuming that dispersed ownership and either weak stock markets being ‘‘above average’’ or ‘‘below average’’ on a or weak shareholder rights. Before combining with particular indicator is sufficient for having high or the ownership condition, we first took the maxi- low membership in a particular set, the extent of mum score for membership in the set of countries membership depends on theoretical criteria and with an absence of stock market activity or weak substantive knowledge of the cases at hand. While shareholder rights. We treat these as functionally we adopt this qualitative approach to data calibra- equivalents for coordination, since the absence of tion, the results of our calibration scoring correlate either condition means that managers are unlikely at levels of 95 % or above with scores derived from to feel strong pressures from shareholders. Turning more automated coding procedures such as using a to employment relations, high coordination is pre- standardized rank ordering of cases. sent in countries with coordinated wage bargaining Table 3 presents a summary of all data sources, and either long employee tenures or strong the calibrations of each institutional indicator and employment protection. Again, the outcomes of construction of higher-order constructs used in the high tenure or legal protection of regular workers empirical analysis. are treated as functionally equivalent indicators of We combined the indicators into five higher- long-term employment relations. order constructs, one for each domain, taking the Table 4 presents a summary overview of our minimum score among the selected indicators for calibrated measures for each country, showing each country (on set operators logical AND and OR, whether they fall into more coordinated or more see Ragin, 2008), whereby higher membership were liberal ‘‘types’’ for each institutional domain based coded to indicate greater coordination. Firm hier- on the average scores across the period 1995–2003. archy is the membership in the set of countries with In our analysis, we did not use these averages but both high coordination for board codetermination yearly indicators, since many countries actually and works councils. Inter-firm relations are highly changed configurations during this time period. coordinated in countries with both low mergers To calibrate set membership for radicality, we and acquisitions per capita and where full mergers considered membership in the set of industries are not the predominant form of mergers and with high radicality based on the proportion of acquisitions. Education is highly coordinated in patents that are highly cited. Similar to the contri- countries that have high occupational training bution to the trade balance measure, we calibrated and low university education. Membership in the radicality in terms of the concentration of radical Journal of International Business Studies VOC and Comparative Advantage Michael A. Witt and Gregory Jackson Journal of International Business Studies Table 3 Data sources and calibration of measures for institutional coordination Coordination in Indicator Explanation of the measure Anchor points for Years Sources institutional calibration of fuzzy-set available domain membership Corporate Membership in set of countries with low dispersion and either low shareholder protection or low stock market size (all items reverse coded) governance Shareholder Legal rights of shareholders (0–6 scale) Fully-out: 1.5 1995, 2003 Botero et al. (2004) and La Porta et al. protection Cross-over: 2.5 (1999) Fully-in: 5 Dispersion of Proportion of largest 20 firms by market Fully-out: 0 1995, 2003 La Porta et al. (1999) and own control capitalization without a blockholder (10 % stake Cross-over: 0.4 calculations from Thomson Banker One or greater) Fully-in: 0.8 database Size of stock Stock market capitalization as a percentage of Fully-out: 10 Continuous OECD statistics market GDP Cross-over: 60 Fully-in: 110 Inter-firm Membership in set of countries with low M&A and low mergers (all items reverse coded sets) relations Mergers and Number of mergers and acquisitions by domestic Fully-out: 0 Continuous S&P Capital IQ acquisitions acquiring firms divided by GDP (billion USD) Cross-over: 0.25 (M&A) Fully-in: 0.5 Mergers Proportion of M&A transactions by domestic Fully-out: 0 Continuous S&P Capital IQ acquiring firms involving full-scale merger of Cross-over: 0.4 target firm Fully-in: 0.8 Employment Membership in set of countries with high wage coordination and either low short-term employment (reverse coded) or high employment protection relations Wage 5 = economy-wide Fully-out: 1 Continuous Jelle Visser, http://www.uva-aias.net/ coordination 4 = industry and economy-wide Cross-over: 3.5 208 degree 3 = industry Fully-in: 5 2 = industry and firm level 1 = mostly at company level Short-term Proportion of employment with less than 1 year Fully-out: 0.1 Continuous OECD Statistics employment of tenure Cross-over: 0.15 Fully-in: 0.25 Employment Strictness of legal rules concerning dismissal (0–6 Fully-out: 0 Continuous OECD Statistics protection scale) Cross-over: 1.5 Fully-in: 3 Firm hierarchy Membership in set of countries with high codetermination and high works council rights Board-level Legal rights for employee representation on Fully-out: 0 2003 Jackson (2005) codetermination corporate board Cross-over: 0.5 1 = parity representation Fully-in: 1 0.8 = 1/3 representation 0.6 = information rights 0.4 = representation in public companies only 0.2 = constitutional protection 0 = no representation VOC and Comparative Advantage Michael A. Witt and Gregory Jackson innovations in particular industries relative to the mean level of radical innovations for each country. We considered citations in the bottom 25 % and top 75 % to be fully out or fully in the set of highly cited patents, whereas zero difference with the country mean would indicate a crossover point between comparative disadvantage or advantage for that industry relative to overall country performance. While this indicators looks at relative advantages of countries across indus- tries, the results of our fsQCA analysis proved identical to an alternative calibration that directly compared the proportion of radical patents in each industry between different countries. In order to further check the robustness of this radicality measure, we applied the same procedure to mea- sures of radicality based on generality or original- ity dimensions, and then constructed a higher- order set membership scores for radical innovation based on citations received, originality or generality. Finally, our outcome condition of contribution to the trade balance was calibrated based on -2as being fully out, 2 as being fully in, and zero as the cross-over point. Substantively, the contribution to the trade balance score of zero offers an intuitive crossover point, since this score repre- sents no comparative advantage or disadvantage. The anchors for no and full membership are slightly greater than two standard deviations around the crossover point. This calibration cor- relates highly with the rank ordering of cases, and our results proved robust to alternative anchor points. Data Analysis Our analysis used fsQCA to evaluate which of the 2 possible combinations of causal conditions (where 6 represents 5 institutional conditions plus indus- try-level radicality) are sufficient conditions for the outcome of high contribution to the trade balance. In evaluating the data, our solutions needed to meet several thresholds (e.g., Fiss, 2011; Judge et al., 2015; Ragin, 2008). First, we accepted only solutions with a consistency score (a measure of reliability) that showed a statistically significant level above the common benchmark of 0.80. Second, we excluded any solutions that were simultaneously consistent with the negation of the outcome. This problem of simultaneous subsets is possible where set membership in causal condi- tions is very low (Schneider & Wagemann, 2012). Third, valid solutions required a threshold of 40 Journal of International Business Studies Table 3 (Continued) Coordination in Indicator Explanation of the measure Anchor points for calibration of Years Sources institutional domain fuzzy-set membership available Works council 3 = economic and social rights, including codetermination on Fully-out: 0 Continuous Jelle Visser, rights some issues Cross-over: 1.5 http://www. 2 = economic and social rights, consultation (advice) only Fully-in: 3 uva-aias.net/ 1 = social rights 208 0 = no rights (sanctions), only information. Education and Membership in set of countries with high occupational training and low university training training Occupational Graduates from upper secondary or post-secondary non-tertiary Fully-out: 20 Continuous OECD statistics training training as percentage of graduation age population Cross-over: 40 Fully-in: 60 University Graduates from tertiary education as percentage of graduation age Fully-out: 10 Continuous OECD statistics training population Cross-over: 25 Fully-in: 40 VOC and Comparative Advantage Michael A. Witt and Gregory Jackson Table 4 Set membership of countries in highly coordinated institutions, 1995–2003 average Note: Set membership scores 0.5–1 indicate medium to full membership in the category of high coordinated, whereas scores 0 to 0.5 indicate zero to medium membership (shaded cells) in the set of coordinated countries and thereby displaying liberal institutional features. cases. Standard applications of fsQCA are usually assumptions and interpret the results of the fsQCA small-N, with a common threshold of two cases complex solutions only. (e.g., Fiss, 2011). Since we are using fsQCA in a large N setting, we chose a stricter threshold on the consideration that any valid solution would have to RESULTS contain observations from about two years or two Table 5 shows the results of our analysis. It suggests industries per country. Solutions with fewer cases three pathways to high contribution to the trade than 40 were treated as being inconsistent. We balance, with a coverage of 0.482 (i.e., they jointly confirmed robustness of this choice by increasing explain 48.2 % of membership in our dataset). and decreasing the threshold. Finally, our analysis Consistency levels (similar to statistical signifi- does not contain any strictly directional hypothe- cance) are all high throughout at levels of 0.914 ses linking institutional factors to comparative and higher, which is considerably better than the advantage outcomes, since our hypotheses suggest commonly used minimum threshold of 0.800. that both liberal and coordinated institutions may Solution 1 combines low radicality with high lead to comparative advantage under certain con- coordination across all institutional spheres. This ditions. Thus while we identify the ‘‘core condi- corresponds to incremental innovation producing tions’’ as provided by the parsimonious solutions a comparative advantage in CMEs and thus offers (cf. Fiss, 2011), we do not postulate any ‘‘easy partial support of H1. Austria follows this CME counterfactuals’’ to be used as simplifying pattern, achieving comparative advantage across a Journal of International Business Studies VOC and Comparative Advantage Michael A. Witt and Gregory Jackson Table 5 Sufficient conditions for high revealed comparative advantage Solution 1 Solution 2 Solution 3 Radicality Corporate governance coordination Employment relations coordination Education and skills coordination Firm hierarchy coordination Interfirm coordination Raw coverage 0.170 0.376 0.177 Unique coverage 0.057 0.248 0.026 Consistency 0.939 0.928 0.949 Total coverage, all solutions 0.482 Solution consistency, all solutions 0.914 Note: = presence of condition, = absence of condition, blank = absence or presence does not matter; large symbols show core conditions (parsimonious solution). range of industries such as office equipment, comparative advantage in motor vehicles, chemicals, aircraft, and metals. Notably, Austrian industries pharmaceuticals, machinery and metal products. Dur- experienced only incremental forms of innovation, ing the 1990s, Germany underwent substantial liber- while radical innovation was essentially absent. alization of its corporate governance arrangements, However, this result only gives partial support for despite relative continuity of its coordinated employ- H1, as the second pattern predicted by H1, high ment relations and strong occupational training insti- radicality with low coordination across all institu- tutions (Jackson, 2003). During this period, changes in tional spheres, is absent from the set of solutions. corporate governance exerted substantial pressure on Solution 2 combines radicality with low coordina- organized labor to cooperate in firm restructuring and tion in all institutional spheres except employment support changes in business models. This, we argue, relations. Interestingly, inter-firm relations are irrele- ultimately served here as a type of ‘‘beneficial con- vant to this solution and may be either liberal or straint’’ on coordinated patterns of business organiza- coordinated. This result is consistent with H2 and tion in the sense of balancing the logic of coordination describes Japan (e.g., motor vehicles, aircraft), Switzer- with a stronger market-orientation. The importance of land during the late 1990s (e.g., chemicals), and this ‘‘opposing’’ institutional logic is clear if we com- Ireland (e.g., chemicals, medical equipment sectors) pare Germany to its neighboring Austria, which as key cases. While Japan is often seen as an exemplar represents a pure CME-type of economy and excelled of CMEs (e.g., Witt, 2006; Yamamura & Streeck, 2003), only in incrementally innovative industries with a albeit with unusual characteristics, our measures notable absence of success with radical innovation. present a different picture. Japan is characterized by We conducted a number of robustness tests. First, a fairly liberal set of institutions – education is focused we used a more stringent calibration of the dependent more on university graduates with general skills, variable measuring contribution to the trade balance employee voice in firm hierarchies is not strongly by changing the anchor points for low and high rooted in law, and after substantial changes to corpo- contribution to the trade balance from -2to -4and rate governance during the 1990s, shareholder rights ?2to ?4, respectively. This produced precisely the and patterns of corporate ownership became fairly same three solutions as our main model. Coverage liberal. Nonetheless, Japanese firms sustained ‘‘micro- dropped minimally from 0.485 to 0.482, while con- corporatist’’ arrangements within the domain of sistency improved from 0.914 to 0.933. Second, we employment relations based on strong employment used the Balassa index instead of contribution to the protection and low labor turnover. Unlike more pure trade balance as a dependent variable. This produced a LMEeconomies,stableemploymentis coreto the single solution that is identical with Solution 3 in our success of Japanese manufacturing, such as Toyota. main model(describing thecaseofGermany). Cover- Solution 3 combines high radicality with high age was 0.194, consistency, 0.925. This lends support coordination in all institutional spheres except for to H3 only. At the same time, given the issues with corporate governance, which is consistent with H3. construct validity of the Balassa index as noted earlier, Germany falls into this category and achieved high we are not very confident that this support, or the lack Journal of International Business Studies VOC and Comparative Advantage Michael A. Witt and Gregory Jackson Table 6 Sufficient conditions for low revealed comparative advantage Solution 1 Solution 2 Solution 3 Solution 4 Radicality Corporate governance coordination Employment relations coordination Education and skills coordination Firm hierarchy coordination Interfirm coordination Raw coverage 0.345 0.199 0.283 0.408 Unique coverage 0.008 0.048 0.120 0.042 Consistency 0.880 0.921 0.914 0.854 Total coverage, all solutions 0.595 Solution consistency, all solutions 0.844 Note: = presence of condition, = absence of condition, blank = absence or presence does not matter; large symbols show core conditions (parsimonious solution). of support of the other hypotheses, is particularly relations and education. This configuration is consis- meaningful. Third, we used a more comprehensive tently associated with low contribution to trade measure of radicality. The three measures of radicality balance in industries if radical innovation is absent. discussed earlier – number of patents received, origi- Unlike pure CMEs, coordination is absent in certain nality, and generality – represent functional substi- domains such as employment relations, being rather tutes in that all three have been described as potential focused on a more liberal flexicurity employment expressions of radicality in innovation. Following model. At the same time, coordinated corporate Ragin’s (2008) approach for functional substitutes, we governance does not act as a beneficial constraint vis- combined all three variables using an OR operator – a-vis market-oriented employment patterns, since that is, the highest value of any of these three variables concentrated share ownership is not sufficient in itself became the radicality measure for each given data to induce investments in firm-specific skills and point. This model produced two solutions, which are capacities needed for incremental innovation. In equivalent to Solutions 2 and 3 in our main model. addition, Solution 3 showsaconfiguration forlow Joint coverage was 0.362, and all consistency scores contribution to the trade balance describing Greece were above 0.908. This provides additional support of and Portugal, as well as some years for New Zealand H2 and H3, but no support of H1. and Spain. This configuration shows that radicality is Overall, these results offer strong support of H3 associated with poor performance where coordinated and, given our doubts about the Balassa index, H2. institutions for corporate governance and inter-firm Partial support of H1 is confined to our main model relations are mixed with liberal patterns for firm and thus seems more tentative. hierarchies and education. Despite the mix of coordi- Unlike in correlational methods, solutions consis- nated and liberal elements, the configuration lacks tent with high and low contribution to the trade complementarities based on beneficial constraints. balance are not necessarily symmetrical in fsQCA Again, coordinated corporate governance, such as (Schneider & Wagemann, 2012). The additional results highly concentrated ownership patterns, or weak in Table 6 are helpful in interpreting our findings for M&A activity in inter-firm relations do not act as a the main model. Solutions 1 and 4 show that low constraining counterbalance to liberal market-ori- contribution to the trade balance is consistent with ented patterns of education and employment. As countries with liberal institutions across 3 or 4 insti- such, these particular ‘‘mixed market economies’’ fail tutional domains and the absence of radical innova- to obtain comparative advantage in radical industries tion. This finding is essentially a mirror of Solution 1 in (Hancke ´, Rhodes, & Thatcher, 2007). Table 5, but shows that countries approximating LME- type configurations do not achieve comparative advantage in industries with low radicality. Solution DISCUSSION In this article we used fsQCA to examine how 2 describes several Nordic cases, such as Norway, country-level institutional configurations relate to Denmark and Finland that have coordinated corporate governance and firm hierarchies with liberal inter-firm institutional comparative advantage across Journal of International Business Studies VOC and Comparative Advantage Michael A. Witt and Gregory Jackson different industrial sectors characterized by either theorize how complementarities may alternatively more radical or incremental patterns of innovation. be based on the combination of conflicting insti- Our main finding is that radical innovation con- tutional logics. Here complementarities may result sistently leads to high comparative advantage in from opposing logics, in which liberal and coordi- countries with institutions that combine specific nated institutions compensate for one another. liberal and coordinated elements: Radical innova- Specifically, coordinated employment relations tion leads to comparative advantage in economies institutions may constrain the influence of liberal with predominantly liberal institutions but coordi- markets, thereby enabling firms to make and nated employment relations and, conversely, in protect investments in core human resources. economies with predominantly coordinated insti- Meanwhile, in highly coordinated economies, lib- tutions but liberal corporate governance. We inter- eral corporate governance may act as an external pret these ‘‘mixed’’ or hybrid combinations of monitor that constrains the behavior of company liberal and coordinated institutions as having stakeholders, but in ways that also enable more complementarities derived from beneficial con- rapid and successful adaptation to radical innova- straints, whereby an institutional logic of either tion. Nonetheless, our analysis shows clearly that market or coordinated exchange is counter-bal- not all hybrid combinations of markets and coor- anced by an opposing institutional logic. dination achieve these sorts of beneficial con- One implication of our results relates to the straints. In mixed market economies, such as validity of the Varieties of Capitalism hypothesis Greece of Portugal, neither does liberal corporate (H1). While we find some evidence that CMEs have governance act as an external monitor for highly an institutional comparative advantage in indus- coordinated patterns of organization, nor does tries with incremental innovation, the data show coordinated employment relations serve as a con- no indication that LMEs have an institutional straint of liberal market organization. Of course, comparative advantage in industries with radical our analysis leaves a number of cases unexplained, innovation. Additional regression analysis, which suggests a research agenda to better map and reported in the Appendix, also shows no clear test hybrid institutional logics based on beneficial support of H1. This lack of strong and consistent constraints. support based on state-of-the-art measures and Complementarities based on opposing principles methodology suggests that the Varieties of Capital- of social organization have a further and poten- ism framework needs substantial reinterpretation tially wide-ranging implication: that complemen- and further theoretical development. tarities should be understood as political projects, At the same time, our results are broadly consis- not equilibrium outcomes. For example, as Ger- tent with earlier studies linking performance in many faced economic turmoil in the early 1990s, certain sectors with institutional configurations the political process of liberalization triggered a that go beyond the LME and CME dichotomy. For process where more coordinated employment rela- example, Boyer (2004) explored what institutional tions were defended and adapted to new con- configurations are consistent with a technology-led straints of more liberal corporate governance. growth regime based on information and commu- While this institutional dynamism may have ener- nications technology. He identified three different gized the economy, one consequence has been configurations, none consistent with Hall and increasing inequality and the political temptation Soskice’s (2001) notion of complementarities to liberalize institutions further, which may undo through coherence. Similarly, Schneider et al. the success (Streeck, 2009). Thus new hybrid con- (2010) linked hybrid economies with mixed insti- figurations of institutions are no more stable equi- tutional logics to strong export performance in libria than more coherent configurations; rather, high technology industries. Our article lends addi- institutions change dynamically all the time. The tional support to the argument that certain types of Celtic Tiger or Danish flexicurity are other good hybrids may be more successful than the pure types examples. These ‘‘success models’’ of one era even- envisioned by Hall and Soskice (2001). tually run their course, often because the winners The key contribution here is to go beyond these and losers of changing institutional arrangements previous studies by offering an ex ante prediction of push for further processes of institutional change the kinds of results we expect to see, drawing on the (Thelen, 2014). As countries travel past certain concept of beneficial constraints. This approach, windows, they may gain comparative advantages, which builds on Streeck (1997), allowed us to but these combinations of institutional Journal of International Business Studies VOC and Comparative Advantage Michael A. Witt and Gregory Jackson arrangements may eventually exhaust themselves the potential benefits of using fsQCA to elucidate as other countries and innovation patterns change the conjunctural effects of different combinations in unison (Boyer, 2004). Here a key finding of our of institutional indicators (Ragin, 2008). While study is that the competitive advantage of Ger- previous studies have applied fsQCA methods to many was based not simply on its presumed the Varieties of Capitalism literature, our results are character as an ideal–typical CME country; rather, broadly consistent with past results that have it did better than pure CMEs like Austria by linked more complex institutional configurations incorporating hybrid institutional logics. Similarly, that go beyond the simple LME and CME types we can go beyond other country stereotypes in the with stronger economic growth or export perfor- literature to compare other unexpected pairs, such mance (Kogut & Ragin, 2006; Schneider et al., as Ireland and Japan. Rather than being opposed as 2010). The multidimensionality evident in our LME and CME types, we can further investigate the analysis calls for international business scholars to parallel ways in which these countries combine pay greater attention to differences in kinds of liberal institutions with high coordination in institutions and their combination, rather than employment relations. conceptualizing of institutions as one-dimensional Our findings also add to the still relatively sparse differences in degrees, as done widely in literatures literature linking the comparative capitalism liter- on institutional distance, institutional develop- ature to economic outcomes. There is little doubt ment or even Varieties of Capitalism (Jackson & that there are institutional differences, but the ‘‘so Deeg, 2008). It also suggests that benefits usually what?’’ – whether and where these variations associated with social structures may be contingent matter – has been much less explored. Our study on the presence of other, supportive institutional finds that institutional comparative advantage may structures. For instance, while our analysis gener- indeed exist, though possibly not in the manner ally supports the widely accepted notion that previously envisaged. Importantly, this suggests interorganizational networks may produce benefi- adopting research designs that allow for causal cial outcomes (e.g., Dyer & Singh, 1998; Sako, complexity and equifinality, whereby different 1992), Solution 3 in Table 6 suggests that in some institutional structures can lead to similar out- configurations, they may be associated with weak comes. The original Varieties of Capitalism performance. approach (Hall & Soskice, 2001) suggested equifi- Our findings may further have important impli- nality between CMEs and LMEs with respect to cations for understanding the location choices of overall economic wealth, but posited a more simple multinational enterprises (MNEs) seeking to avoid relationship between LMEs and the ability to home country disadvantages, project competitive leverage radical innovation. By contrast, our results advantages related to their home country, or suggest that at least two different institutional exploit complementary resources and knowledge configurations are associated with comparative related to host country institutional environments advantage in the presence of radical innovation. (cf. Jackson & Deeg, 2008; Singh, 2007; Witt & Furthermore, these configurations are dramatically Lewin, 2007). The literature has previously identi- different: either all elements but corporate gover- fied host country institutions as sources of loca- nance are coordinated, or all but employment tional advantages (Dunning and Lundan 2008), relations are liberal. which suggests the possibility of institutional arbi- Our conceptualization of complementarities trage (Jackson & Deeg, 2008; Witt & Lewin, 2007): based on opposing logics and the empirical results Firms may move their operations, in part or in on equifinality challenge some existing conceptu- whole, to institutional contexts that better support alizations of institutional differences in Interna- these operations. The Varieties of Capitalism liter- tional Business. For example, recent works have ature would have suggested that firms should locate already noted that the impact of distances may not activities involving radical innovations in LMEs be symmetric or even indicative of difficulty oper- such as the United States. Our results suggest a ating abroad (Cuervo-Cazurra & Genc, 2012). In different picture: Production involving radical this context, we hope that our approach can help innovation seems to thrive in economies pairing provide a basis for future empirical work on the either liberal corporate governance with coordina- economic impact of Varieties of Capitalism in tion in the other spheres of the political economy, particular and of institutional differences in Inter- or coordinated employment relations with formally national Business more generally. Our study shows liberal institutions in other areas. These patterns are Journal of International Business Studies VOC and Comparative Advantage Michael A. Witt and Gregory Jackson far more complex than connections between insti- In terms of radicality, we faced the same problem tutions and locational choices of MNEs previously that has plagued all prior attempts at exploring the explored in the literature, such as political risk or hypothesis: how to measure it. Our operationaliza- corruption (Henisz, 2002; Rodriguez, Uhlenbruck, tion follows procedures used and accepted by the & Eden, 2005). At the same time, it is possible that specialist literature on R&D. However, patent data experience and evolutionary processes have shaped and even patent citations have well-known limita- MNE investment patterns accordingly. To the tions. Not all innovations can be patented (e.g., extent this is the case, one would expect that MNEs new software), and firms do not necessarily patent relying more on radical innovations locate rela- their innovations for cost or strategic reasons tively more of their activities in these economies. (Archibugi & Planta, 1996). Moreover, radical One would also expect that the performance of innovation may also be associated with process MNEs relying more on radical innovations should innovations driven by changes in organization for correspond to the extent to which they locate their which new technology is a necessary but not a operations in economies with suitable institutional sufficient condition. Our results analyze the impact configurations. Exploring these hypotheses would of radicality across a wide range of industries, but seem to be a fruitful avenue for future IB research. different dynamics may occur and different insti- tutional features may be salient within specific Limitations industries. As such, our article does not exhaust this Though our empirical findings are robust to a large agenda. number of alternative specifications, they have The notion of radicality can also be questioned at important methodological limitations. a deeper level. It remains unclear whether radical First, our results may be limited to the specific time innovation is properly understood at the level of frame and country sample. Our study is based on a patents, meaning proprietary inventions, or 9-year window from 1995 through 2003, for reasons whether it entails shifts in broader technological of data availability as explained earlier. This period paradigms independent of particular inventions or covers the time period for which the Varieties of applications. For example, major shifts in informa- Capitalism approach claimed validity (cf. Hall & tion technology are not centered on single radical Gingerich, 2009) and constitutes a strong test of the inventions, but may imply a radical shift in busi- original hypothesis. Nonetheless, as newer data ness models or the locus of competitive advantage. become available, future studies might examine These broader phenomena, like the information whether the hypothesized comparative advantages technology revolution, may give firms in LMEs new manifest themselves post-2003. For example, the full opportunities to leverage institutions for fast entry effects of the Internet revolution, shifts in biotech- and exit from lines of business, or to use external nology or expansion of global value chains may be labor markets to retool their human capital and clearer over a longer time span. Moreover, while our organizational capabilities. But we cannot discount sample covers all countries discussed by Hall and the idea that firms in CMEs will leverage their Soskice, and while Varieties of Capitalism theory institutions to make incremental innovations focuses on advanced industrialized economies, it is around this same context, whereby information possible that a larger range of countries might lead to technologies are integrated successfully into estab- more comprehensive insights. lished product lines in ways leading to quality or A second limitation concerns the construct valid- process improvements. Put differently, comparative ity of the coordination and radicality measures. advantage may be reconceived as being less cen- Concerning the former, Hall and Soskice discussed tered on advantages related to a fixed boundary of the notion of transferable versus relational assets an industry or technology. Rather, innovation may across the five institutional domains of corporate be refracted through the lens of institutions in ways governance and finance, employment relations, that are leveraged for more radical or incremental education, managerial hierarchies within firms, shifts in business strategy and organization. Here and inter-firm relations. However, little consensus multinational firms with capabilities for institu- exists about which institutional indicators should tional bricolage may have distinctive advantages. be used to best capture ‘‘coordination’’ in each Indeed, research on Central and Eastern Europe has domain. While we believe that our variables max- started to find competitive advantage in high-tech imized construct validity to the extent possible, industries based on strong FDI with complex insti- better measures may become available. tutional hybrids that involve combinations of Journal of International Business Studies VOC and Comparative Advantage Michael A. Witt and Gregory Jackson liberal elements with either strong employee par- NOTES ticipation or longer-term employment relations Porter’s (1990) work on the competiveness of (Allen & Aldred, 2011). As such, we think the nations has been an important influence in the agenda of understanding how coordinated rela- development of these new approaches. We exclude tions act as beneficial constraints on markets has a it from the discussion because Porter’s focus is on long future ahead. competitiveness and thus on absolute advantage rather than on comparative advantage. On the notion of institutions as resources more generally, see also Hall and Thelen (2009) and Jackson ACKNOWLEDGMENTS and Deeg (2008). We thank Jaeyong Song for his constructive editorial This original index has also important empirical guidance and the three anonymous reviewers for their limitations, please refer to the Appendix. insightful feedback. We are further grateful for com- Generally, raising thresholds for full membership ments and input received from Robert Boyer, Daniel produced very similar results, whereas lowering W. Gingerich, Bronwyn H. Hall, Sebastian Koos, and thresholds to just one standard deviation produced Gerald A. McDermott. Rashid Ansari at INSEAD and fewer consistent results, as the resultant calibration no So¨ren Salzwedel at FU Berlin provided valuable longer distinguishes between cases of high and very research assistance, and we thank the INSEAD libraries high contributions to the trade balance. for their support. Michael Witt also thanks the Hum- boldt Foundation for funding two 3-month research stays in Berlin to undertake this project. REFERENCES Akkermans, D., Castaldi, C., & Los, B. 2009. Do ‘liberal market Blyth, M. 2003. Same as it never was: Temporality and typology economies’ really innovate more radically than ‘coordinated in the varieties of capitalism. Comparative European Politics, market economies’? Hall and Soskice reconsidered. Research 1(2): 215–225. Policy, 38(1): 181–191. Bohle, D., & Greskovits, B. 2009. Varieties of capitalism and Allen, M. 2004. The varieties of capitalism paradigm: Not capitalism ‘‘tout court’’. European Journal of Sociology, 50(3): enough variety? Socio-Economic Review, 2(1): 87–108. 355–386. Allen, M. M. C., & Aldred, M. L. 2009. Varieties of capitalism, Botero, J. C., Djankov, S., La Porta, R., Lopez-de-Silanes, F., & varieties of innovation? A comparison of old and new EU Shleifer, A. 2004. The regulation of labor. Quarterly Journal of member states. Journal of Contemporary European Research, Economics, 19(4): 1339–1382. 5(4): 581–596. Boyer, R. 2004. New growth regimes, but still institutional Allen, M. M. C., & Aldred, M. L. 2011. Varieties of capitalism, diversity. Socio-Economic Review, 2(1): 1–32. governance, and high-tech export performancea fuzzy-set Brewster, C., Wood, G., & Brookes, M. 2006. Varieties of analysis of the new EU member states. Employee Relations, capitalism and varieties of firm. In P. James, & G. Wood (Eds), 33(4): 334–355. Institutions, production, and working life. Oxford: Oxford Amable, B. 2003. The diversity of modern capitalism. Oxford: University Press. Oxford University Press. Campbell, J. L. 2011. The US financial crisis: Lessons for theories of Aoki, M. 2010. Corporations in evolving diversity: Cognition, institutional complementarity. Socio-Economic Review, 9(2): governance, and institutions. Oxford: Oxford University Press. 211–234. Aoki, M., & Jackson, G. 2008. Understanding an emergent Carney, M., Gedajlovic, E., & Yang, X. 2009. Varieties of Asian diversity of corporate governance and organizational archi- capitalism: Toward an institutional theory of Asian enterprise. tecture: An essentiality-based analysis. Industrial and Corporate Asia Pacific Journal of Management, 26(3): 361–380. Change, 17(1): 1–27. Crilly, D. 2011. Predicting stakeholder orientation in the Archibugi, D., & Planta, M. 1996. Measuring technological multinational enterprise: A mid-range theory. Journal of change through patents and innovation surveys. Technova- International Business Studies, 42(5): 694–717. tion, 16(9): 451–519. Crouch, C. 2005. Capitalist diversity and change. Recombinant Balassa, B. 1965. Trade liberalisation and ‘‘revealed’’ compara- governance and institutional entrepreneurs. Oxford: Oxford tive advantage. The Manchester School, 33(2): 99–123. University Press. Bassanini, A., & Ernst, E. 2002. Labour market regulation, Crouch, C., & Streeck, W. 1997. Political economy of modern industrial relations and technological regimes: A tale of capitalism: Mapping convergence and diversity. London: Sage. comparative advantage. Industrial and Corporate Change, Crouch, C., Streeck, W., Boyer, R., Amable, B., Hall, P. A., & 11(3): 391–426. Jackson, G. 2005. Dialogue on ‘institutional complementar- Bell, G., Filatotchev, I., & Aguilera, R. 2013. Corporate gover- ity and political economy. Socio-Economic Review, 2(4): nance and investors’ perceptions of foreign ipo value: An 359–382. institutional perspective. Academy of Management Journal, Cuervo-Cazurra, A., & Genc, M. E. 2012. Categories of distance 57(1): 301–320. and international business. In G. Wood, & M. Demirbag (Eds), Block, F. L., & Somers, M. R. 2014. The power of market Handbook of institutional approaches to international business: fundamentalism: Karl Polanyi’s critique. Cambridge, MA: Har- 219–235. Cheltenham: Edward Elgar. vard University Press. Journal of International Business Studies VOC and Comparative Advantage Michael A. Witt and Gregory Jackson Dearsdorff, A. V. 1980. The general validity of the law of and methods in organizational research: 129–158. Bingley: comparative advantage. Journal of Political Economy, 88(5): Emerald Group Publishing. 941–957. Jackson, G., & Petraki, A. 2010. Understanding short-termism: Deeg, R. 2007. Institutional complementarity and change in The role of corporate governance. Stockholm: Glasshouse capitalist systems. Journal of European Public Policy, 14(4): Forum. 612–623. Judge, W. Q., Fainshmidt, S., & Brown, J. L, I. I. I. 2014. Which Dunning, J. H., & Lundan, S. M. 2008. Institutions and the oli model of capitalism best delivers both wealth and equality? paradigm of the multinational enterprise. Asia Pacific Journal of Journal of International Business Studies, 45(4): 363–386. Management, 25(4): 573–593. Judge, W. Q., et al. 2015. Configurations of capacity for change Dyer, J. H., & Singh, H. 1998. The relational view: Cooperative in entrepreneurial threshold firms: Imprinting and strategic strategy and sources of interorganizational competitive advan- choice perspectives. Journal of Management Studies, 52(4): tage. Academy of Management Review, 23(4): 660–679. 506–530. Fiss, P. C. 2011. Building better causal theories: A fuzzy set Kogut, B., & Ragin, C. C. 2006. Exploring complexity when approach to typologies in organization research. Academy of diversity is limited: Institutional complementarity in theories of Management Journal, 54(2): 393–420. rule of law and national systems revisited. European Manage- Furman, J. L., Porter, M. E., & Stern, S. 2002. The determinants ment Review, 3(1): 44–59. of national innovative capacity. Research Policy, 31(6): La Porta, R., Lopez-de-Silanes, F., Schleifer, A., & Vishney, R. W. 899–933. 1998. Law and finance. Journal of Political Economy, 106(6): Gospel, H., & Pendleton, A. 2003. Finance, corporate gover- 1113–1155. nance and the management of labour: A conceptual and La Porta, R., Lopez-de-Silanes, F., & Shleifer, A. 1999. Corporate comparative analysis. British Journal of Industrial Relations, ownership around the world. Journal of Finance, 54(2): 41(3): 557–582. 471–517. Gospel, H., & Pendleton, A. (Eds) 2005. Corporate governance Longest, K. C., & Vaisey, S. 2008. Fuzzy: A program for and labour management: An international comparison. Oxford: performing qualitative comparative analyses (QCA) in stata. Oxford University Press. Stata Journal, 8(1): 79–104. Hall, B. H., Bessen, J., & Thoma, G. 2011. NBER patent data project. Mahoney, J., & Thelen, K. 2009. Explaining institutional change: https://sites.google.com/site/patentdataproject/Home. Ambiguity, agency, and power. Cambridge: Cambridge Univer- Hall, P. A., & Gingerich, D. W. 2009. Varieties of capitalism and sity Press. institutional complementarities in the political economy. Milgrom, P. R., & Roberts, J. 1994. Complementarities and British Journal of Political Science, 39(3): 449–482. systems: Understanding Japanese economic organization. Hall, B. H., Jaffe, A. B., & Trajtenberg, M. 2001. The NBER patent Estudios Economicos, 9(1): 3–42. citations data file: Lessons, insights and methodological tools. Milgrom, P. R., & Roberts, J. 1995. Complementarities and NBER working paper series no. 8498. fit: Strategy, structure, and organizational change in manu- Hall, P. A., & Soskice, D. 2001. An introduction to varieties of facturing. Journal of Accounting and Economics, 19(2/3): capitalism. In P. A. Hall, & D. Soskice (Eds), Varieties of 179–208. capitalism: The institutional foundations of comparative advan- Newbert, S. L. 2007. Empirical research on the resource-based tage: 1–68. Oxford: Oxford University Press. view of the firm: An assessment and suggestions for future Hall, P. A., & Thelen, K. 2009. Institutional change in varieties of research. Strategic Management Journal, 28(2): 121–140. capitalism. Socio-Economic Review, 7(1): 7–34. No¨lke, A., & Vliegenthart, A. 2009. Enlarging the varieties of Hancke, B., Rhodes, M., & Thatcher, M. (Eds) 2007. Beyond capitalism: The emergence of dependent market economies in varieties of capitalism: Contradictions, complementarities & east central Europe. World Politics, 61(4): 670–702. change. Oxford: Oxford University Press. OECD. 2003. Science, technology and industry scoreboard. Paris: Henisz, W. J. 2002. Politics and international investment: Mea- OECD. suring risks and protecting profits. London: Edward Elgar. OECD. 2013. OECD.Stat. www.oecd-ilibrary.org/content/data/ Herrmann, A. M., & Peine, A. 2011. When ‘national innovation data-00285-en.2013. system’ meet ‘varieties of capitalism’ arguments on labour Porter, M. E. 1990. The competitive advantage of nations. New qualifications: On the skill types and scientific knowledge York: Free Press. needed for radical and incremental product innovations. Ragin, C. C. 2000. Fuzzy-set social science. Chicago: University of Research Policy, 40(5): 687–701. Chicago Press. Ho¨pner, M. 2005. What connects industrial relations and Ragin, C. C. 2008. Redesigning social inquiry: Fuzzy sets and corporate governance? Explaining institutional complemen- beyond. Chicago: University of Chicago Press. tarity. Socio-Economic Review, 3(2): 331–358. Rajan, R. G., & Zingales, L. 2000. The governance of the new Jackson, G. 2003. Corporate governance in Germany and Japan: enterprises. In X. Vives (Ed), Corporate governance: Theoretical Liberalization pressures and responses. In K. Yamamura, & W. & empirical perspectives. Cambridge: Cambridge University Streeck (Eds), The end of diversity? Prospects for German and Press. Japanese capitalism: 261–305. Ithaca, NY: Cornell University Redding, G., & Witt, M. A. 2007. The future of Chinese capitalism: Press. Choices and chances. Oxford: Oxford University Press. Jackson, G. 2005. Employee representation in the board com- Redding, G., & Witt, M. 2009. China’s business system and its pared: A fuzzy sets analysis of corporate governance, unionism, future trajectory. Asia Pacific Journal of Management, 26(3): and political institutions. Industrielle Beziehungen, 12(3): 1–28. 381–399. Jackson, G., & Deeg, R. 2008. Comparing capitalisms: Under- Rodriguez, P., Uhlenbruck, L., & Eden, L. 2005. Government standing institutional diversity and its implications for inter- corruption and the entry strategies of multinationals. Academy national business. Journal of International Business Studies, of Management Review, 30(2): 383–396. 39(4): 540–561. Sako, M. 1992. Price, quality and trust: Inter-firm relations in Jackson, G., & Muellenborn, T. 2012. Understanding the role of britain and Japan. Cambridge: Cambridge University Press. institutions in industrial relations: Perspectives from classical Schneider, M. R., & Paunescu, M. 2012. Changing varieties of sociological theory. Industrial Relations: A Journal of Economy capitalism and revealed comparative advantages from 1990 to and Society, 51(S1): 472–500. 2005: A test of the Hall and Soskice claims. Socio-Economic Jackson, G., & Ni, N. 2013. Understanding complementarities as Review, 10(4): 731–753. organizational configurations using set theoretical methods. In Schneider, M. R., Schulze-Bentrop, C., & Paunescu, M. 2010. P. C. Fiss, B. Cambre, & A. Marx (Eds), Configurational theory Mapping the institutional capital of high-tech firms: A fuzzy- Journal of International Business Studies VOC and Comparative Advantage Michael A. Witt and Gregory Jackson set analysis of capitalism variety and export performance. Institute for Advanced Labour Studies, University of Journal of International Business Studies, 41(2): 246–266. Amsterdam. Schneider, C. Q., & Wagemann, C. 2012. Set-theoretic methods Vitols, S. 2002. Shareholder value, management culture and for the social sciences: A guide to qualitative comparative production regimes in the transformation of the German analysis. Cambridge: Cambridge University Press. chemical-pharmaceutical industry. Competition and Change, Singh, J. 2007. Asymmetry of knowledge spillovers between 6(3): 309–325. MNCs and host country firms. Journal of International Business Vitols, S. 2004. Negotiated shareholder value: The German Studies, 38(5): 764–786. version of an anglo-American practice. Competition and StataCorp. 2011. Stata statistical software: Release 12. College Change, 8(4): 1–18. Station, TX: StataCorp. LP. Whitley, R. 1999. Divergent capitalisms: The social structuring and StataCorp. 2013. Stata statistical software: Release 13. College change of business systems. Oxford: Oxford University Press. Station, TX: StataCorp. Witt, M. A. 2006. Changing Japanese capitalism: Societal coordi- Streeck, W. 1997. Beneficial constraints: On the economic limits nation and institutional adjustment. Cambridge: Cambridge of rational voluntarism. In J. R. Hollingsworth & R. Boyer (Eds), University Press. Contemporary capitalism: The embeddedness of institutions: Witt, M. A., & Lewin, A. Y. 2007. Outward foreign direct 197–219. Cambridge: Cambridge University Press. investment as escape response to home country institutional Streeck, W. 2008. Re-forming capitalism: Institutional change in constraints. Journal of International Business Studies, 38(4): the German political economy: Institutional change in the 579–594. German political economy. Oxford: Oxford University Press. Witt, M. A., & Redding, G. 2013. Asian business systems: Streeck, W. 2009. Re-forming capitalism: Institutional change in Institutional comparison, clusters and implications for varieties the German political economy. Oxford: Oxford University Press. of capitalism and business systems theory. Socio-Economic Streeck, W., & Thelen, K. (Eds) 2005. Beyond continuity: Review, 11(2): 265–300. Institutional change in advanced political economies. Oxford: Witt, M. A., Kabbach de Castro, L. R., Amaeshi, K., Mahroum, S., Oxford University Press. Bohle, D., & Saez, L. 2015. The business systems of the world’s Taylor, M. Z. 2004. Empirical evidence against varieties of leading 61 economies: Institutional comparison, clusters, and capitalism’s theory of technological innovation. International implications for varieties of capitalism and business systems Organization, 58(Summer): 601–631. research. INSEAD working paper series, INSEAD, Fon- Thelen, K. 2014. Varieties of liberalization and the new politics of tainebleau, France. social solidarity. Cambridge: Cambridge University Press. World Bank. 2013. World development indicators. http://data. Transparency International. 2013. Corruption perception index. worldbank.org/data-catalog/world-development-indicators. http://www.transparency.org.cpi2011. 2011. Trefler, D., & Zhu, S. C. 2000. Beyond the algebra of Yamamura, K., & Streeck, W. (Eds) 2003. The end of diversity? explanation: Hov for the technology age. American Economic Prospects for German and Japanese capitalism. Ithaca, NY: Review, 90(2): 145–149. Cornell University Press. Visser, J. 2011. Data base on institutional characteristics of trade unions, wage setting, state intervention and social pacts, 1960–2010 (ICTWSS),Version 3.0. Amsterdam: Amsterdam APPENDIX Data As a robustness test, we used regression analysis to Dependent variables were the contribution to the undertake an alternative test of H1. Regression trade balance plus the Balassa index, as discussed in analysis is not suitable for testing H2 and H3 the main text. because the configurational logic of these argu- The independent variable of interest was the ments would involve 6-way interaction terms. interaction term between radicality by number of However, the argument of coherence across all 5 citations received and the coordination index we spheres of the political economy inherent in H1 introduce below. To account for possible direct makes it possible to derive a single ‘‘coordination effects, we also included both the radicality mea- index’’ to use in a 2-way interaction. While fsQCA sure and the coordination measure in the still remains the method of choice for this kind of regression. investigation, not least because of its ability to accommodate asymmetric solutions (see discussion in the main text), we present this additional test as Coordination a robustness check. For each country and year, we calculated an index This Appendix reports the data, and methodol- of overall coordination. Hall and Gingerich (2009) used confirmatory factor analysis to derive a coor- ogy and main results. The results depend on the operationalization of the dependent variable: our dination index for 20 advanced industrialized preferred measure, contributions to the trade bal- economies, with index values closer to 0 denoting ance, does not support H1, while the Balassa index higher similarity with the prototypical LME and as an alternative offers inconsistent support of H1. closer to 1, with the prototypical CME. Their Journal of International Business Studies VOC and Comparative Advantage Michael A. Witt and Gregory Jackson coordination index is based on three indicators of Hall and Gingerich but were inspired by the 5 corporate governance and finance (an index of institutional domains described in the original for- shareholder protection, dispersion of control, and mulation of Varieties of Capitalism by Hall and the size of stock market) and three indicators of Soskice (2001). Taking z-transformed variables to employment relations (the level of wage bargain- create a uniform scale, we used the principal-factor ing, the degrees of coordination in wage bargain- method in Stata (StataCorp, 2011) to calculate a ing, and the proportion of employees with tenures single factor loading using squared multiple corre- of less than 1 year). This original index has a lations to estimate the commonality of an underly- number of limitations that made it unsuitable for ing factor. The single largest factor had a factor our study. First, it provides a point estimate for loading of 74 % and a large eigenvalue of 4.14, about the mid-1990s rather than a time-series over suggesting a strong correlation along this single the period of our study. Second, the original factor dimension of coordination. The second factor had scores calculated in Hall and Gingerich combined an eigenvalue of 1.14 and explained an additional indicators from the late 1980s to the mid-1990s. 20.7 % of the variation, which suggests that some When synchronizing the data as an annual time important variance is not fully captured in a single series for 1995–2003, we found that the indicators index. However, since we are interested in the extent did not load on a common underlying factor when to which countries vary on a single dimension, we applying confirmatory factor analysis as used by did not interpret any of the other factor scores. Hall and Gingerich. Figure A1 shows the factor scores for countries in We developed an alternative specification of the our sample over time. While there is some move- coordination index using principal component ment of CMEs toward the LME end of the spec- analysis for the period 1995 through 2003. We drew trum, supporting the notion that economies on the 12 indicators used in our main fsQCA model liberalized in the 1990s and early 2000s, we did as described in Table 2 of the main text. This set of not observe any major changes in the general indicators is more comprehensive that those used by ordering of positions. Table A1 lists the values. Au Au Au Au Au Au Au Au Au Au Au Au Au Aus s s s s s s s s s s s s st t t t t t t t t t t t t tr r r r r r r r r r r r r riiiiiiiiiiiiiia a a a a a a a a a a a a a Au Au Au Au Au Au Au Au Au Au Au Au Au Aus s s s s s s s s s s s s st t t t t t t t t t t t t tr r r r r r r r r r r r r riiiiiiiiiiiiiia a a a a a a a a a a a a a Au Au Au Au Au Au Au Au Au Au Au Au Au Aus s s s s s s s s s s s s st t t t t t t t t t t t t tr r r r r r r r r r r r r riiiiiiiiiiiiiia a a a a a a a a a a a a a Be Be Be Be Be Be Be Be Be Be Be Be Be Bellllllllllllllg g g g g g g g g g g g g giiiiiiiiiiiiiiu u u u u u u u u u u u u um m m m m m m m m m m m m m Au Au Au Au Au Au Au Au Au Au Au Au Au Aus s s s s s s s s s s s s st t t t t t t t t t t t t tr r r r r r r r r r r r r riiiiiiiiiiiiiia a a a a a a a a a a a a a Au Au Au Au Au Au Au Au Au Au Au Au Au Aus s s s s s s s s s s s s st t t t t t t t t t t t t tr r r r r r r r r r r r r riiiiiiiiiiiiiia a a a a a a a a a a a a a G G G G G G G G G G G G G Ge e e e e e e e e e e e e er r r r r r r r r r r r r rm m m m m m m m m m m m m many any any any any any any any any any any any any any Au Au Au Au Au Au Au Au Au Au Au Au Au Aus s s s s s s s s s s s s st t t t t t t t t t t t t tr r r r r r r r r r r r r riiiiiiiiiiiiiia a a a a a a a a a a a a a G G G G G G G G G G G G G Ge e e e e e e e e e e e e er r r r r r r r r r r r r rm m m m m m m m m m m m m many any any any any any any any any any any any any any Be Be Be Be Be Be Be Be Be Be Be Be Be Bellllllllllllllg g g g g g g g g g g g g giiiiiiiiiiiiiiu u u u u u u u u u u u u um m m m m m m m m m m m m m G G G G G G G G G G G G G Ge e e e e e e e e e e e e er r r r r r r r r r r r r rm m m m m m m m m m m m m many any any any any any any any any any any any any any Au Au Au Au Au Au Au Au Au Au Au Au Au Aus s s s s s s s s s s s s st t t t t t t t t t t t t tr r r r r r r r r r r r r riiiiiiiiiiiiiia a a a a a a a a a a a a a Be Be Be Be Be Be Be Be Be Be Be Be Be Bellllllllllllllg g g g g g g g g g g g g giiiiiiiiiiiiiiu u u u u u u u u u u u u um m m m m m m m m m m m m m Ge Ge Ge Ge Ge Ge Ge Ge Ge Ge Ge Ge Ge Ger r r r r r r r r r r r r rm m m m m m m m m m m m m ma a a a a a a a a a a a a an n n n n n n n n n n n n ny y y y y y y y y y y y y y Au Au Au Au Au Au Au Au Au Au Au Au Au Aus s s s s s s s s s s s s st t t t t t t t t t t t t tr r r r r r r r r r r r r riiiiiiiiiiiiiia a a a a a a a a a a a a a Au Au Au Au Au Au Au Au Au Au Au Au Au Aus s s s s s s s s s s s s st t t t t t t t t t t t t tr r r r r r r r r r r r r riiiiiiiiiiiiiia a a a a a a a a a a a a a Ge Ge Ge Ge Ge Ge Ge Ge Ge Ge Ge Ge Ge Ger r r r r r r r r r r r r rm m m m m m m m m m m m m ma a a a a a a a a a a a a an n n n n n n n n n n n n ny y y y y y y y y y y y y y Be Be Be Be Be Be Be Be Be Be Be Be Be Bellllllllllllllg g g g g g g g g g g g g giiiiiiiiiiiiiiu u u u u u u u u u u u u um m m m m m m m m m m m m m Ge Ge Ge Ge Ge Ge Ge Ge Ge Ge Ge Ge Ge Ger r r r r r r r r r r r r rm m m m m m m m m m m m m ma a a a a a a a a a a a a an n n n n n n n n n n n n ny y y y y y y y y y y y y y N N N N N N N N N N N N N Ne e e e e e e e e e e e e et t t t t t t t t t t t t ther her her her her her her her her her her her her herlllllllllllllla a a a a a a a a a a a a an n n n n n n n n n n n n nd d d d d d d d d d d d d ds s s s s s s s s s s s s s N N N N N N N N N N N N N Ne e e e e e e e e e e e e et t t t t t t t t t t t t ther her her her her her her her her her her her her herlllllllllllllla a a a a a a a a a a a a an n n n n n n n n n n n n nd d d d d d d d d d d d d ds s s s s s s s s s s s s s Ge Ge Ge Ge Ge Ge Ge Ge Ge Ge Ge Ge Ge Ger r r r r r r r r r r r r rm m m m m m m m m m m m m ma a a a a a a a a a a a a an n n n n n n n n n n n n ny y y y y y y y y y y y y y It It It It It It It It It It It It It Ita a a a a a a a a a a a a alllllllllllllly y y y y y y y y y y y y y N N N N N N N N N N N N N Ne e e e e e e e e e e e e et t t t t t t t t t t t t ther her her her her her her her her her her her her herlllllllllllllla a a a a a a a a a a a a an n n n n n n n n n n n n nd d d d d d d d d d d d d ds s s s s s s s s s s s s s Ge Ge Ge Ge Ge Ge Ge Ge Ge Ge Ge Ge Ge Ger r r r r r r r r r r r r rm m m m m m m m m m m m m ma a a a a a a a a a a a a an n n n n n n n n n n n n ny y y y y y y y y y y y y y Be Be Be Be Be Be Be Be Be Be Be Be Be Bellllllllllllllg g g g g g g g g g g g g giiiiiiiiiiiiiiu u u u u u u u u u u u u um m m m m m m m m m m m m m Ge Ge Ge Ge Ge Ge Ge Ge Ge Ge Ge Ge Ge Ger r r r r r r r r r r r r rm m m m m m m m m m m m m ma a a a a a a a a a a a a an n n n n n n n n n n n n ny y y y y y y y y y y y y y S S S S S S S S S S S S S Sw w w w w w w w w w w w w weden eden eden eden eden eden eden eden eden eden eden eden eden eden S S S S S S S S S S S S S Sw w w w w w w w w w w w w weden eden eden eden eden eden eden eden eden eden eden eden eden eden Be Be Be Be Be Be Be Be Be Be Be Be Be Bellllllllllllllg g g g g g g g g g g g g giiiiiiiiiiiiiiu u u u u u u u u u u u u um m m m m m m m m m m m m m It It It It It It It It It It It It It Ita a a a a a a a a a a a a alllllllllllllly y y y y y y y y y y y y y N N N N N N N N N N N N N Ne e e e e e e e e e e e e et t t t t t t t t t t t t ther her her her her her her her her her her her her herlllllllllllllla a a a a a a a a a a a a an n n n n n n n n n n n n nd d d d d d d d d d d d d ds s s s s s s s s s s s s s It It It It It It It It It It It It It Ita a a a a a a a a a a a a alllllllllllllly y y y y y y y y y y y y y No No No No No No No No No No No No No Nor r r r r r r r r r r r r rw w w w w w w w w w w w w wa a a a a a a a a a a a a ay y y y y y y y y y y y y y S S S S S S S S S S S S S Sw w w w w w w w w w w w w weden eden eden eden eden eden eden eden eden eden eden eden eden eden It It It It It It It It It It It It It Ita a a a a a a a a a a a a alllllllllllllly y y y y y y y y y y y y y Be Be Be Be Be Be Be Be Be Be Be Be Be Bellllllllllllllg g g g g g g g g g g g g giiiiiiiiiiiiiiu u u u u u u u u u u u u um m m m m m m m m m m m m m No No No No No No No No No No No No No Nor r r r r r r r r r r r r rw w w w w w w w w w w w w wa a a a a a a a a a a a a ay y y y y y y y y y y y y y N N N N N N N N N N N N N Ne e e e e e e e e e e e e et t t t t t t t t t t t t ther her her her her her her her her her her her her herlllllllllllllla a a a a a a a a a a a a an n n n n n n n n n n n n nd d d d d d d d d d d d d ds s s s s s s s s s s s s s Be Be Be Be Be Be Be Be Be Be Be Be Be Bellllllllllllllg g g g g g g g g g g g g giiiiiiiiiiiiiiu u u u u u u u u u u u u um m m m m m m m m m m m m m De De De De De De De De De De De De De Den n n n n n n n n n n n n nm m m m m m m m m m m m m ma a a a a a a a a a a a a ar r r r r r r r r r r r r rk k k k k k k k k k k k k k De De De De De De De De De De De De De Den n n n n n n n n n n n n nm m m m m m m m m m m m m ma a a a a a a a a a a a a ar r r r r r r r r r r r r rk k k k k k k k k k k k k k It It It It It It It It It It It It It Ita a a a a a a a a a a a a alllllllllllllly y y y y y y y y y y y y y De De De De De De De De De De De De De Den n n n n n n n n n n n n nm m m m m m m m m m m m m ma a a a a a a a a a a a a ar r r r r r r r r r r r r rk k k k k k k k k k k k k k S S S S S S S S S S S S S Sw w w w w w w w w w w w w weden eden eden eden eden eden eden eden eden eden eden eden eden eden Be Be Be Be Be Be Be Be Be Be Be Be Be Bellllllllllllllg g g g g g g g g g g g g giiiiiiiiiiiiiiu u u u u u u u u u u u u um m m m m m m m m m m m m m De De De De De De De De De De De De De Den n n n n n n n n n n n n nm m m m m m m m m m m m m ma a a a a a a a a a a a a ar r r r r r r r r r r r r rk k k k k k k k k k k k k k It It It It It It It It It It It It It Ita a a a a a a a a a a a a alllllllllllllly y y y y y y y y y y y y y S S S S S S S S S S S S S Sw w w w w w w w w w w w w wiiiiiiiiiiiiiit t t t t t t t t t t t t tzer zer zer zer zer zer zer zer zer zer zer zer zer zerlllllllllllllland and and and and and and and and and and and and and N N N N N N N N N N N N N Ne e e e e e e e e e e e e et t t t t t t t t t t t t ther her her her her her her her her her her her her herlllllllllllllla a a a a a a a a a a a a an n n n n n n n n n n n n nd d d d d d d d d d d d d ds s s s s s s s s s s s s s N N N N N N N N N N N N N Ne e e e e e e e e e e e e et t t t t t t t t t t t t ther her her her her her her her her her her her her herlllllllllllllla a a a a a a a a a a a a an n n n n n n n n n n n n nd d d d d d d d d d d d d ds s s s s s s s s s s s s s Gr Gr Gr Gr Gr Gr Gr Gr Gr Gr Gr Gr Gr Gre e e e e e e e e e e e e ee e e e e e e e e e e e e ec c c c c c c c c c c c c ce e e e e e e e e e e e e e N N N N N N N N N N N N N Ne e e e e e e e e e e e e et t t t t t t t t t t t t ther her her her her her her her her her her her her herlllllllllllllla a a a a a a a a a a a a an n n n n n n n n n n n n nd d d d d d d d d d d d d ds s s s s s s s s s s s s s Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Finl nl nl nl nl nl nl nl nl nl nl nl nl nland and and and and and and and and and and and and and Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Finl nl nl nl nl nl nl nl nl nl nl nl nl nland and and and and and and and and and and and and and Gr Gr Gr Gr Gr Gr Gr Gr Gr Gr Gr Gr Gr Gre e e e e e e e e e e e e ee e e e e e e e e e e e e ec c c c c c c c c c c c c ce e e e e e e e e e e e e e It It It It It It It It It It It It It Ita a a a a a a a a a a a a alllllllllllllly y y y y y y y y y y y y y S S S S S S S S S S S S S Sw w w w w w w w w w w w w weden eden eden eden eden eden eden eden eden eden eden eden eden eden N N N N N N N N N N N N N Ne e e e e e e e e e e e e et t t t t t t t t t t t t ther her her her her her her her her her her her her herlllllllllllllla a a a a a a a a a a a a an n n n n n n n n n n n n nd d d d d d d d d d d d d ds s s s s s s s s s s s s s No No No No No No No No No No No No No Nor r r r r r r r r r r r r rw w w w w w w w w w w w w wa a a a a a a a a a a a a ay y y y y y y y y y y y y y No No No No No No No No No No No No No Nor r r r r r r r r r r r r rw w w w w w w w w w w w w wa a a a a a a a a a a a a ay y y y y y y y y y y y y y G G G G G G G G G G G G G Gr r r r r r r r r r r r r reec eec eec eec eec eec eec eec eec eec eec eec eec eece e e e e e e e e e e e e e Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Finl nl nl nl nl nl nl nl nl nl nl nl nl nland and and and and and and and and and and and and and De De De De De De De De De De De De De Den n n n n n n n n n n n n nm m m m m m m m m m m m m ma a a a a a a a a a a a a ar r r r r r r r r r r r r rk k k k k k k k k k k k k k S S S S S S S S S S S S S Sw w w w w w w w w w w w w weden eden eden eden eden eden eden eden eden eden eden eden eden eden It It It It It It It It It It It It It Ita a a a a a a a a a a a a alllllllllllllly y y y y y y y y y y y y y Swi Swi Swi Swi Swi Swi Swi Swi Swi Swi Swi Swi Swi Swit t t t t t t t t t t t t tz z z z z z z z z z z z z ze e e e e e e e e e e e e er r r r r r r r r r r r r rlllllllllllllla a a a a a a a a a a a a an n n n n n n n n n n n n nd d d d d d d d d d d d d d No No No No No No No No No No No No No Nor r r r r r r r r r r r r rw w w w w w w w w w w w w wa a a a a a a a a a a a a ay y y y y y y y y y y y y y De De De De De De De De De De De De De Den n n n n n n n n n n n n nm m m m m m m m m m m m m ma a a a a a a a a a a a a ar r r r r r r r r r r r r rk k k k k k k k k k k k k k Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Franc anc anc anc anc anc anc anc anc anc anc anc anc ance e e e e e e e e e e e e e It It It It It It It It It It It It It Ita a a a a a a a a a a a a alllllllllllllly y y y y y y y y y y y y y Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Franc anc anc anc anc anc anc anc anc anc anc anc anc ance e e e e e e e e e e e e e Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Gr Gr Gr Gr Gr Gr Gr Gr Gr Gr Gr Gr Gr Gr nl nl nl nl nl nl nl nl nl nl nl nl nl nl e e e e e e e e e e e e e eand and and and and and and and and and and and and and e e e e e e e e e e e e e ec c c c c c c c c c c c c ce e e e e e e e e e e e e e No No No No No No No No No No No No No Nor r r r r r r r r r r r r rw w w w w w w w w w w w w wa a a a a a a a a a a a a ay y y y y y y y y y y y y y No No No No No No No No No No No No No Nor r r r r r r r r r r r r rw w w w w w w w w w w w w wa a a a a a a a a a a a a ay y y y y y y y y y y y y y No No No No No No No No No No No No No Nor r r r r r r r r r r r r rw w w w w w w w w w w w w wa a a a a a a a a a a a a ay y y y y y y y y y y y y y Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Franc anc anc anc anc anc anc anc anc anc anc anc anc ance e e e e e e e e e e e e e Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Finl nl nl nl nl nl nl nl nl nl nl nl nl nland and and and and and and and and and and and and and S S S S S S S S S S S S S Sw w w w w w w w w w w w w weden eden eden eden eden eden eden eden eden eden eden eden eden eden S S S S S S S S S S S S S Sw w w w w w w w w w w w w weden eden eden eden eden eden eden eden eden eden eden eden eden eden Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Franc anc anc anc anc anc anc anc anc anc anc anc anc ance e e e e e e e e e e e e e Swi Swi Swi Swi Swi Swi Swi Swi Swi Swi Swi Swi Swi Swit t t t t t t t t t t t t tz z z z z z z z z z z z z ze e e e e e e e e e e e e er r r r r r r r r r r r r rlllllllllllllla a a a a a a a a a a a a an n n n n n n n n n n n n nd d d d d d d d d d d d d d G G G G G G G G G G G G G Gr r r r r r r r r r r r r reec eec eec eec eec eec eec eec eec eec eec eec eec eece e e e e e e e e e e e e e G G G G G G G G G G G G G Gr r r r r r r r r r r r r reec eec eec eec eec eec eec eec eec eec eec eec eec eece e e e e e e e e e e e e e Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Franc anc anc anc anc anc anc anc anc anc anc anc anc ance e e e e e e e e e e e e e No No No No No No No No No No No No No Nor r r r r r r r r r r r r rw w w w w w w w w w w w w wa a a a a a a a a a a a a ay y y y y y y y y y y y y y G G G G G G G G G G G G G Gr r r r r r r r r r r r r reec eec eec eec eec eec eec eec eec eec eec eec eec eece e e e e e e e e e e e e e G G G G G G G G G G G G G Gr r r r r r r r r r r r r reec eec eec eec eec eec eec eec eec eec eec eec eec eece e e e e e e e e e e e e e Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Franc anc anc anc anc anc anc anc anc anc anc anc anc ance e e e e e e e e e e e e e S S S S S S S S S S S S S Sw w w w w w w w w w w w w weden eden eden eden eden eden eden eden eden eden eden eden eden eden P P P P P P P P P P P P P Po o o o o o o o o o o o o or r r r r r r r r r r r r rt t t t t t t t t t t t t tu u u u u u u u u u u u u ugal gal gal gal gal gal gal gal gal gal gal gal gal gal Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Franc anc anc anc anc anc anc anc anc anc anc anc anc ance e e e e e e e e e e e e e Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Franc anc anc anc anc anc anc anc anc anc anc anc anc ance e e e e e e e e e e e e e G G G G G G G G G G G G G Gr r r r r r r r r r r r r reec eec eec eec eec eec eec eec eec eec eec eec eec eece e e e e e e e e e e e e e S S S S S S S S S S S S S Sw w w w w w w w w w w w w wiiiiiiiiiiiiiit t t t t t t t t t t t t tzer zer zer zer zer zer zer zer zer zer zer zer zer zerlllllllllllllland and and and and and and and and and and and and and Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Fr Franc anc anc anc anc anc anc anc anc anc anc anc anc ance e e e e e e e e e e e e e P P P P P P P P P P P P P Po o o o o o o o o o o o o or r r r r r r r r r r r r rt t t t t t t t t t t t t tu u u u u u u u u u u u u ugal gal gal gal gal gal gal gal gal gal gal gal gal gal P P P P P P P P P P P P P Po o o o o o o o o o o o o or r r r r r r r r r r r r rt t t t t t t t t t t t t tu u u u u u u u u u u u u ugal gal gal gal gal gal gal gal gal gal gal gal gal gal De De De De De De De De De De De De De Den n n n n n n n n n n n n nm m m m m m m m m m m m m ma a a a a a a a a a a a a ar r r r r r r r r r r r r rk k k k k k k k k k k k k k De De De De De De De De De De De De De Den n n n n n n n n n n n n nm m m m m m m m m m m m m ma a a a a a a a a a a a a ar r r r r r r r r r r r r rk k k k k k k k k k k k k k Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Kor r r r r r r r r r r r r re e e e e e e e e e e e e ea a a a a a a a a a a a a a Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Finl nl nl nl nl nl nl nl nl nl nl nl nl nland and and and and and and and and and and and and and S S S S S S S S S S S S S Sw w w w w w w w w w w w w wiiiiiiiiiiiiiit t t t t t t t t t t t t tzer zer zer zer zer zer zer zer zer zer zer zer zer zerlllllllllllllland and and and and and and and and and and and and and S S S S S S S S S S S S S Sw w w w w w w w w w w w w wiiiiiiiiiiiiiit t t t t t t t t t t t t tzer zer zer zer zer zer zer zer zer zer zer zer zer zerlllllllllllllla a a a a a a a a a a a a and nd nd nd nd nd nd nd nd nd nd nd nd nd S S S S S S S S S S S S S Sw w w w w w w w w w w w w wiiiiiiiiiiiiiit t t t t t t t t t t t t tzer zer zer zer zer zer zer zer zer zer zer zer zer zerlllllllllllllland and and and and and and and and and and and and and De De De De De De De De De De De De De Den n n n n n n n n n n n n nm m m m m m m m m m m m m ma a a a a a a a a a a a a ar r r r r r r r r r r r r rk k k k k k k k k k k k k k S S S S S S S S S S S S S Sw w w w w w w w w w w w w wiiiiiiiiiiiiiit t t t t t t t t t t t t tzer zer zer zer zer zer zer zer zer zer zer zer zer zerlllllllllllllla a a a a a a a a a a a a and nd nd nd nd nd nd nd nd nd nd nd nd nd Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Finl nl nl nl nl nl nl nl nl nl nl nl nl nland and and and and and and and and and and and and and P P P P P P P P P P P P P Po o o o o o o o o o o o o or r r r r r r r r r r r r rt t t t t t t t t t t t t tu u u u u u u u u u u u u ugal gal gal gal gal gal gal gal gal gal gal gal gal gal S S S S S S S S S S S S S Sw w w w w w w w w w w w w wiiiiiiiiiiiiiit t t t t t t t t t t t t tzer zer zer zer zer zer zer zer zer zer zer zer zer zerlllllllllllllland and and and and and a a a a a a a and nd nd nd nd nd nd nd Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Kor r r r r r r r r r r r r re e e e e e e e e e e e e ea a a a a a a a a a a a a a Po Po Po Po Po Po Po Po Po Po Po Po Po Por r r r r r r r r r r r r rt t t t t t t t t t t t t tu u u u u u u u u u u u u ug g g g g g g g g g g g g ga a a a a a a a a a a a a allllllllllllll Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Finl nl nl nl nl nl nl nl nl nl nl nl nl nland and and and and and and and and and and and and and Po Po Po Po Po Po Po Po Po Po Po Po Po Por r r r r r r r r r r r r rt t t t t t t t t t t t t tu u u u u u u u u u u u u ug g g g g g g g g g g g g ga a a a a a a a a a a a a allllllllllllll Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Fi Finl nl nl nl nl nl nl nl nl nl nl nl nl nland and and and and and and and and and and and and and Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Kor r r r r r r r r r r r r re e e e e e e e e e e e e ea a a a a a a a a a a a a a Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Kor r r r r r r r r r r r r re e e e e e e e e e e e e ea a a a a a a a a a a a a a Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Po Po Po Po Po Po Po Po Po Po Po Po Po Por r r r r r r r r r r r r rt t t t t t t t t t t t t tu u u u u u u u u u u u u ug g g g g g g g g g g g g ga a a a a a a a a a a a a allllllllllllll Po Po Po Po Po Po Po Po Po Po Po Po Po Por r r r r r r r r r r r r rt t t t t t t t t t t t t tu u u u u u u u u u u u u ug g g g g g g g g g g g g ga a a a a a a a a a a a a allllllllllllll S S S S S S S S S S S S S Spai pai pai pai pai pai pai pai pai pai pai pai pai pain n n n n n n n n n n n n n Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Kor r r r r r r r r r r r r re e e e e e e e e e e e e ea a a a a a a a a a a a a a Po Po Po Po Po Po Po Po Po Po Po Po Po Por r r r r r r r r r r r r rt t t t t t t t t t t t t tu u u u u u u u u u u u u ug g g g g g g g g g g g g ga a a a a a a a a a a a a allllllllllllll Sp Sp Sp Sp Sp Sp Sp Sp Sp Sp Sp Sp Sp Spa a a a a a a a a a a a a aiiiiiiiiiiiiiin n n n n n n n n n n n n n S S S S S S S S S S S S S Spai pai pai pai pai pai pai pai pai pai pai pai pai pain n n n n n n n n n n n n n S S S S S S S S S S S S S Spai pai pai pai pai pai pai pai pai pai pai pai pai pain n n n n n n n n n n n n n Sp Sp Sp Sp Sp Sp Sp Sp Sp Sp Sp Sp Sp Sp Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan a a a a a a a a a a a a a aiiiiiiiiiiiiiin n n n n n n n n n n n n n Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko S S S S S S S S S S S S S Spai pai pai pai pai pai pai pai pai pai pai pai pai pai r r r r r r r r r r r r r re e e e e e e e e e e e e en n n n n n n n n n n n n n a a a a a a a a a a a a a a S S S S S S S S S S S S S Spai pai pai pai pai pai pai pai pai pai pai pai pai pain n n n n n n n n n n n n n Ir Ir Ir Ir Ir Ir Ir Ir Ir Ir Ir Ir Ir Ire e e e e e e e e e e e e elllllllllllllla a a a a a a a a a a a a an n n n n n n n n n n n n nd d d d d d d d d d d d d d S S S S S S S S S S S S S Spai pai pai pai pai pai pai pai pai pai pai pai pai pain n n n n n n n n n n n n n S S S S S S S S S S S S S Spai pai pai pai pai pai pai pai pai pai pai pai pai pain n n n n n n n n n n n n n Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Ir Ir Ir Ir Ir Ir Ir Ir Ir Ir Ir Ir Ir Ire e e e e e e e e e e e e elllllllllllllla a a a a a a a a a a a a an n n n n n n n n n n n n nd d d d d d d d d d d d d d Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Kor r r r r r r r r r r r r re e e e e e e e e e e e e ea a a a a a a a a a a a a a Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne New Z w Z w Z w Z w Z w Z w Z w Z w Z w Z w Z w Z w Z w Ze e e e e e e e e e e e e ea a a a a a a a a a a a a alllllllllllllla a a a a a a a a a a a a an n n n n n n n n n n n n nd d d d d d d d d d d d d d Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan I I I I I I I I I I I I I Ir r r r r r r r r r r r r rel el el el el el el el el el el el el elan an an an an an an an an an an an an and d d d d d d d d d d d d d Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Kor r r r r r r r r r r r r re e e e e e e e e e e e e ea a a a a a a a a a a a a a Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Ir Ir Ir Ir Ir Ir Ir Ir Ir Ir Ir Ir Ir Ire e e e e e e e e e e e e elllllllllllllla a a a a a a a a a a a a an n n n n n n n n n n n n nd d d d d d d d d d d d d d I I I I I I I I I I I I I Ir r r r r r r r r r r r r rel el el el el el el el el el el el el elan an an an an an an an an an an an an and d d d d d d d d d d d d d Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Ko Kor r r r r r r r r r r r r re e e e e e e e e e e e e ea a a a a a a a a a a a a a I I I I I I I I I I I I I Ir r r r r r r r r r r r r rel el el el el el el el el el el el el elan an an an an an an an an an an an an and d d d d d d d d d d d d d N N N N N N N N N N N N N Ne e e e e e e e e e e e e ew w w w w w w w w w w w w w Z Z Z Z Z Z Z Z Z Z Z Z Z Zeal eal eal eal eal eal eal eal eal eal eal eal eal ealan an an an an an an an an an an an an and d d d d d d d d d d d d d I I I I I I I I I I I I I Ir r r r r r r r r r r r r rel el el el el el el el el el el el el elan an an an an an an an an an an an an and d d d d d d d d d d d d d Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne New Z w Z w Z w Z w Z w Z w Z w Z w Z w Z w Z w Z w Z w Ze e e e e e e e e e e e e ea a a a a a a a a a a a a alllllllllllllla a a a a a a a a a a a a an n n n n n n n n n n n n nd d d d d d d d d d d d d d I I I I I I I I I I I I I Ir r r r r r r r r r r r r rel el el el el el el el el el el el el elan an an an an an an an an an an an an and d d d d d d d d d d d d d I I I I I I I I I I I I I Ir r r r r r r r r r r r r rel el el el el el el el el el el el el elan an an an an an an an an an an an an and d d d d d d d d d d d d d Au Au Au Au Au Au Au Au Au Au Au Au Au Aus s s s s s s s s s s s s st t t t t t t t t t t t t tr r r r r r r r r r r r r ra a a a a a a a a a a a a alllllllllllllliiiiiiiiiiiiiia a a a a a a a a a a a a a Au Au Au Au Au Au Au Au Au Au Au Au Au Aus s s s s s s s s s s s s st t t t t t t t t t t t t tr r r r r r r r r r r r r ra a a a a a a a a a a a a alllllllllllllliiiiiiiiiiiiiia a a a a a a a a a a a a a Au Au Au Au Au Au Au Au Au Au Au Au Au Aus s s s s s s s s s s s s st t t t t t t t t t t t t tr r r r r r r r r r r r r ra a a a a a a a a a a a a alllllllllllllliiiiiiiiiiiiiia a a a a a a a a a a a a a Au Au Au Au Au Au Au Au Au Au Au Au Au Aus s s s s s s s s s s s s st t t t t t t t t t t t t tr r r r r r r r r r r r r ra a a a a a a a a a a a a alllllllllllllliiiiiiiiiiiiiia a a a a a a a a a a a a a Au Au Au Au Au Au Au Au Au Au Au Au Au Aus s s s s s s s s s s s s st t t t t t t t t t t t t tr r r r r r r r r r r r r ra a a a a a a a a a a a a alllllllllllllliiiiiiiiiiiiiia a a a a a a a a a a a a a Au Au Au Au Au Au Au Au Au Au Au Au Au Aus s s s s s s s s s s s s st t t t t t t t t t t t t tr r r r r r r r r r r r r ra a a a a a a a a a a a a alllllllllllllliiiiiiiiiiiiiia a a a a a a a a a a a a a Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne New Z w Z w Z w Z w Z w Z w Z w Z w Z w Z w Z w Z w Z w Ze e e e e e e e e e e e e ea a a a a a a a a a a a a alllllllllllllla a a a a a a a a a a a a an n n n n n n n n n n n n nd d d d d d d d d d d d d d US US US US US US US US US US US US US US C C C C C C C C C C C C C Ca a a a a a a a a a a a a an n n n n n n n n n n n n nada ada ada ada ada ada ada ada ada ada ada ada ada ada C C C C C C C C C C C C C Ca a a a a a a a a a a a a an n n n n n n n n n n n n nada ada ada ada ada ada ada ada ada ada ada ada ada ada C C C C C C C C C C C C C Ca a a a a a a a a a a a a an n n n n n n n n n n n n nada ada ada ada ada ada ada ada ada ada ada ada ada ada Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne New Z w Z w Z w Z w Z w Z w Z w Z w Z w Z w Z w Z w Z w Ze e e e e e e e e e e e e ea a a a a a a a a a a a a alllllllllllllla a a a a a a a a a a a a an n n n n n n n n n n n n nd d d d d d d d d d d d d d Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne New Z w Z w Z w Z w Z w Z w Z w Z w Z w Z w Z w Z w Z w Ze e e e e e e e e e e e e ea a a a a a a a a a a a a alllllllllllllla a a a a a a a a a a a a an n n n n n n n n n n n n nd d d d d d d d d d d d d d US US US US US US US US US US US US US US C C C C C C C C C C C C C Ca a a a a a a a a a a a a an n n n n n n n n n n n n nada ada ada ada ada ada ada ada ada ada ada ada ada ada C C C C C C C C C C C C C Ca a a a a a a a a a a a a an n n n n n n n n n n n n nada ada ada ada ada ada ada ada ada ada ada ada ada ada Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne New Z w Z w Z w Z w Z w Z w Z w Z w Z w Z w Z w Z w Z w Ze e e e e e e e e e e e e ea a a a a a a a a a a a a alllllllllllllla a a a a a a a a a a a a an n n n n n n n n n n n n nd d d d d d d d d d d d d d C C C C C C C C C C C C C Ca a a a a a a a a a a a a an n n n n n n n n n n n n nada ada ada ada ada ada ada ada ada ada ada ada ada ada Au Au Au Au Au Au Au Au Au Au Au Au Au Aus s s s s s s s s s s s s st t t t t t t t t t t t t tr r r r r r r r r r r r r ra a a a a a a a a a a a a alllllllllllllliiiiiiiiiiiiiia a a a a a a a a a a a a a Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne New Z w Z w Z w Z w Z w Z w Z w Z w Z w Z w Z w Z w Z w Ze e e e e e e e e e e e e ea a a a a a a a a a a a a alllllllllllllla a a a a a a a a a a a a an n n n n n n n n n n n n nd d d d d d d d d d d d d d Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne Ne New Z w Z w Z w Z w Z w Z w Z w Z w Z w Z w Z w Z w Z w Ze e e e e e e e e e e e e ea a a a a a a a a a a a a alllllllllllllla a a a a a a a a a a a a an n n n n n n n n n n n n nd d d d d d d d d d d d d d C C C C C C C C C C C C C Ca a a a a a a a a a a a a an n n n n n n n n n n n n nada ada ada ada ada ada ada ada ada ada ada ada ada ada C C C C C C C C C C C C C Ca a a a a a a a a a a a a an n n n n n n n n n n n n nada ada ada ada ada ada ada ada ada ada ada ada ada ada US US US US US US US US US US US US US US Au Au Au Au Au Au Au Au Au Au Au Au Au Aus s s s s s s s s s s s s st t t t t t t t t t t t t tr r r r r r r r r r r r r ra a a a a a a a a a a a a alllllllllllllliiiiiiiiiiiiiia a a a a a a a a a a a a a C C C C C C C C C C C C C Ca a a a a a a a a a a a a an n n n n n n n n n n n n nada ada ada ada ada ada ada ada ada ada ada ada ada ada US US US US US US US US US US US US US US UK UK UK UK UK UK UK UK UK UK UK UK UK UK Au Au Au Au Au Au Au Au Au Au Au Au Au Aus s s s s s s s s s s s s st t t t t t t t t t t t t tr r r r r r r r r r r r r ra a a a a a a a a a a a a alllllllllllllliiiiiiiiiiiiiia a a a a a a a a a a a a a US US US US US US US US US US US US US US UK UK UK UK UK UK UK UK UK UK UK UK UK UK US US US US US US US US US US US US US US US US US US US US US US US US US US US US UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK US US US US US US US US US US US US US US UK UK UK UK UK UK UK UK UK UK UK UK UK UK US US US US US US US US US US US US US US UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK UK 1994 1996 1998 2000 2002 2004 Year Figure A1 Coordination index, 1995–2003. Journal of International Business Studies Index of Coordination 0 .2 .4 .6 .8 1 VOC and Comparative Advantage Michael A. Witt and Gregory Jackson Table A1 Coordination index for 22 OECD countries, 1995–2003 Country 1995 1996 1997 1998 1999 2000 2001 2002 2003 Australia 0.24 0.23 0.23 0.23 0.23 0.20 0.13 0.11 0.09 Austria 0.96 0.96 0.94 0.91 0.91 0.89 0.86 0.85 0.82 Belgium 0.92 0.88 0.86 0.80 0.75 0.72 0.69 0.69 0.67 Canada 0.17 0.17 0.17 0.15 0.12 0.11 0.12 0.14 0.13 Denmark 0.68 0.67 0.67 0.68 0.61 0.61 0.54 0.52 0.51 Finland 0.64 0.64 0.62 0.59 0.58 0.52 0.50 0.48 0.47 France 0.60 0.60 0.58 0.57 0.55 0.55 0.55 0.56 0.57 Germany 0.90 0.88 0.86 0.84 0.82 0.80 0.77 0.75 0.74 Greece 0.65 0.64 0.62 0.59 0.57 0.57 0.56 0.55 0.56 Ireland 0.37 0.34 0.31 0.28 0.25 0.25 0.26 0.27 0.26 Italy 0.76 0.72 0.71 0.70 0.68 0.66 0.64 0.62 0.60 Japan 0.46 0.43 0.43 0.37 0.36 0.34 0.33 0.32 0.30 Netherlands 0.79 0.77 0.76 0.71 0.69 0.65 0.63 0.65 0.65 New Zealand 0.33 0.26 0.25 0.19 0.17 0.16 0.13 0.13 0.13 Norway 0.71 0.69 0.62 0.62 0.61 0.56 0.59 0.59 0.58 Portugal 0.54 0.55 0.54 0.50 0.48 0.47 0.43 0.43 0.41 South Korea 0.52 0.49 0.46 0.45 0.41 0.37 0.33 0.31 0.27 Spain 0.40 0.41 0.39 0.37 0.39 0.37 0.36 0.37 0.36 Sweden 0.73 0.72 0.70 0.67 0.63 0.61 0.58 0.58 0.56 Switzerland 0.65 0.61 0.57 0.55 0.52 0.51 0.50 0.51 0.52 UK 0.10 0.08 0.06 0.03 0.01 0.02 0.01 0.04 0.06 USA 0.10 0.08 0.07 0.04 0.03 0.06 0.11 0.15 0.18 Note: Higher values indicate higher levels of CME-style coordination. We used this variable in its interaction with EU, NAFTA These country-level control variables radicality (which then provided the independent are dummy variables coded 1 if the respective variable of interest for testing H1) and its direct effect country was a member of the EU or NAFTA and 0 in the regression to control for the possibility that otherwise. It controlled for the possibility that coordination in itself affects revealed comparative membership in these free trade areas may have an advantages without interacting with radicality. impact on the structure of trade of a country and Control variables included the following: thus on the structure of contributions to the trade Number of patents This industry-level control balance. variable is the natural logarithm of the number of GDP per capita This country-level control variable patents in the given country, industry, and year is the natural logarithm of per capita GDP at (OECD, 2013). It controlled for the possibility that purchasing power parity and constant year 2000 greater contributions to the trade balance in a given US dollars (OECD, 2013). It controlled for possible industry may be the result of higher levels of structural effects in contributions to the trade innovation in general, as opposed to radical balance based on higher levels of economic innovation. development. For instance, it is possible that Corruption levels and regulatory quality These coun- higher GDP levels may be associated with greater try-level control variables are the corruption per- comparative advantage in higher technology ception index, obtained from Transparency industries. International (2013), and the regulatory quality GERD per GDP This country-level control variable measures from the World Bank Development is the ratio of Gross Expenses on R&D over GDP Indicators database (2013). They controlled for (OECD, 2013). It controlled for possible effects that the possibility that institutional shortcomings may higher levels of spending on R&D in an economy prevent patent owners from making full use of may have on contributions to the trade balance. their patents, which could reduce contributions to We further included dummies for all years in the the trade balance an economy can derive from sample. Table A2 shows descriptive statistics for all radical innovations. variables. Journal of International Business Studies VOC and Comparative Advantage Michael A. Witt and Gregory Jackson Model Specification We used hierarchical linear modeling (HLM), as implemented in the ‘‘mixed’’ command in Stata 13.1 (StataCorp, 2013). This choice was condi- tioned by the presence of multiple violations of the assumption that data are independent and identically distributed. Specifically, we expected clustering effects at two levels, country and indus- try. In line with the Varieties of Capitalism approach, we assumed that country-level charac- teristics supersede industry-level characteristics and consequently specified a three-level HLM model that nested observations first in industries and then in countries. We further specified robust standard errors with correction for AR(1) autoregressive effects. Robustness tests with AR(2) and AR(3) specification yielded qualitatively identical results. Results Models (1) through (3) in Table A3 show the build- up to our main model (3), which is a three-level HLM specification regressing the contribution to the trade balance variable on the interaction of coordination and radicality expressed in citations received. The coefficient estimate for the interaction effect is negative, as expected, and statistically significant. However, further analysis revealed that the result in model (3) is dependent on the presence of two specific sets of data points: the chemical and pharmaceutical industries in Ireland, both of which have received unusually large levels of inward foreign direct investment for production for export. Dropping these two industries for Ireland resulted in insignificant coefficient estimates, as model (4) indicates. No other countries or industries showed similar impact. The problem persisted when using OLS with clustered standard errors, either by industry or country, or when changing the lag to 2 or 3 years, as well as when using the alternative measures of radicality discussed in the main text. We conclude that the results of model (3) are contingent on 18 influential data points represent- ing said two industries in Ireland and thus not generally supported. Model 5 shows the results for the Balassa index as the dependent variable. The coefficient estimate for the interaction effect is negative, as expected, and statistically significant. Unlike in model 3, this result is not contingent on the presence of Irish outlier data points, and it is robust to inverting the Journal of International Business Studies Table A2 Descriptive statistics and correlations Mean SD Min Max 1234567 89 10 1 Contribution to the trade balance -0.119 1.717 -7.328 10.893 2 Balassa index 0.920 0.725 0.000 6.317 0.590 3 Coordination index 1.482 0.244 1.010 1.960 0.020 0.080 4 Radicality 0.029 0.038 0.000 0.417 0.110 0.050 -0.220 5 ln(no. of patents) 3.645 2.253 0.000 9.816 0.260 0.230 -0.180 0.270 6 Corruption 7.602 1.681 2.990 10.000 -0.050 -0.030 -0.150 0.160 0.040 7 Regulatory quality 1.374 0.408 0.258 2.065 -0.060 -0.040 -0.170 0.130 -0.050 0.790 8 EU membership 0.636 0.481 0.000 1.000 -0.010 0.000 0.500 -0.200 -0.290 -0.130 0.110 9 NAFTA membership 0.091 0.288 0.000 1.000 0.020 0.020 -0.470 0.320 0.390 0.140 0.140 -0.420 10 GERD/GDP 1.913 0.810 0.434 4.130 0.160 0.130 0.090 0.170 0.610 0.380 0.120 -0.170 0.130 11 ln(GDP/cap) 10.232 0.214 9.627 10.719 0.110 0.110 -0.020 0.290 0.360 0.500 0.450 -0.210 0.300 0.340 Correlations of abs(0.0403) or greater are statistically significant at the 0.05 level, and of abs(0.0501) or greater, at the 0.01 level. VOC and Comparative Advantage Michael A. Witt and Gregory Jackson Table A3 Regression results (1) (2) (3) (4) (5) Contribution to Contribution to Contribution to Contribution to trade Balassa trade balance trade balance trade balance balance, no outliers Controls ln(number of 0.0807*** 0.0804*** 0.0800*** 0.0775*** 0.000702 patents) (0.0184) (0.0184) (0.0184) (0.0170) (0.00614) Corruption levels -0.00944 -0.0100 -0.00840 -0.00844 0.000640 (0.0189) (0.0189) (0.0189) (0.0174) (0.00647) Regulatory quality -0.0438 -0.0436 -0.0330 -0.0118 0.00733 (0.0748) (0.0752) (0.0753) (0.0690) (0.0263) EU membership 0.103 0.103 0.0954 0.00765 -0.00803 (0.216) (0.223) (0.222) (0.215) (0.0923) NAFTA membership -0.275 -0.283 -0.327 -0.169 0.0623 (0.366) (0.373) (0.373) (0.361) (0.155) GERD/GDP 0.0296 0.0298 0.0308 0.0587 0.0217 (0.0593) (0.0593) (0.0593) (0.0549) (0.0211) ln(GDP/cap) 0.926** 0.904** 0.871** 0.451 0.213 (0.284) (0.286) (0.286) (0.268) (0.113) Year dummies Yes Yes Yes Yes Yes Coordination index 0.00524 0.109 0.101 0.242* (0.281) (0.284) (0.265) (0.107) Radicality 0.399 5.482** 0.587 1.603* (0.289) (2.073) (1.930) (0.699) Coordination 9 radicality -3.432* -0.482 -1.076* (1.386) (1.285) (0.468) Constant -9.760*** -9.552*** -9.394** -5.164 -1.675 (2.859) (2.865) (2.862) (2.692) (1.136) N 2772 2772 2772 2754 2772 AIC 3108.5 3110.6 3106.5 2627.6 -2821.2 BIC 3227.0 3241.0 3242.8 2763.8 -2684.9 Log likelihood -1534.2 -1533.3 -1530.2 -1290.8 1433.6 v 40.98 42.90 49.13 34.53 17.48 Note: Standard errors in parentheses. * p \ 0.05, ** p \ 0.01, ***p \ 0.001. nesting of the HLM specification. Graphing the these results even if the Balassa index represented a interaction showed that the higher Balassa scores of meaningful measure of revealed comparative CME countries decline with increasing radicality, advantage, which we doubt. whereas the lower Balassa scores of LME countries increase with radicality. However, the estimates of Balassa at high levels of radicality have wide confi- ABOUT THE AUTHORS dence intervals, so that we cannot say with confi- Michael A. Witt is Affiliate Professor of Strategy dence that LMEs outperform CMEs at high levels – a and International Business at INSEAD. He earned conservative statement would be to say the disad- his PhD at Harvard University and explores vantages of LMEs are diminished. Looking at other national institutions and their impact on manage- model specifications, the interaction term became rial and organizational outcomes. His papers have statistically insignificant if OLS with clustered stan- appeared in journals such as JIBS, SMJ, JMS, BEQ, dard errors by industry or by country was used. JBE, MOR, and SER. He has published four scholarly Using originality as a measure of radicality resulted books, including The Oxford Handbook of Asian in a statistically significant result with the wrong Business Systems (with Gordon Redding). (positive) sign, while the use of generality produced no statistically significant result. Taking all this Gregory Jackson is Professor of Management at the together, we see at best tentative support of H1 in Freie Universita ¨t Berlin and Einstein Research Journal of International Business Studies VOC and Comparative Advantage Michael A. Witt and Gregory Jackson Fellow at the Berlin Social Science Center. He other third party material in this article are earned his PhD in Sociology at Columbia Univer- included in the article’s Creative Commons license, sity, and his research interests include corporate unless indicated otherwise in the credit line; if the governance, corporate social responsibility and material is not included under the Creative Com- irresponsibility and comparative management. mons license, users will need to obtain permission from the license holder to reproduce the material. To view a copy of this license, visit http:// This work is licensed under a Creative creativecommons.org/licenses/by-nc-nd/3.0/ Commons Attribution-NonCommer- cial-NoDerivs 3.0 Unported License. The images or Michael A. Witt and Gregory Jackson have contributed equally to this article. The authors have a policy of alternating author order of joint papers. Accepted by Jaeyong Song, Area Editor, 18 April 2016. This article has been with the authors for three revisions. Journal of International Business Studies

Journal

Journal of International Business StudiesSpringer Journals

Published: Jul 22, 2016

There are no references for this article.