Towards a Holistic Framework of MNE–State Bargaining (PDF)

Journal of World Business 53 (2018) 15–26 Contents lists available at ScienceDirect Journal of World Business...

Journal of World Business 53 (2018) 15–26 Contents lists available at ScienceDirect Journal of World Business journal homepage: www.elsevier.com/locate/jwb Towards a holistic framework of MNE–state bargaining: A formal model and T case-based analysis ⁎ Jakob Müllner, Jonas Puck WU Vienna, Vienna University of Economics and Business, Austria A R T I C L E I N F O A B S T R A C T Keywords: In this paper, we develop a holistic framework for MNE–state bargaining based on power-dependence theory. We Multinational enterprise contribute to IB theory by moving beyond a static, bilateral conceptualization of MNE–state bargaining towards Obsolescing bargaining a dynamic, multi-party framework. MNEs can shift the power balance in their favor by (a) reducing sunk costs Power-dependence theory from the investment, (b) improving access to alternative investments, (c) increasing the host government’s sunk Extractive industries costs, or (d) impeding the host government’s access to alternative investors and buyers. Subsequently, we apply Foreign direct investment Sovereignty-at-bay our framework to the Venezuelan oil industry, identifying 12 MNE micro-strategies to achieve a sustainable power balance. 1. Introduction obsolescing bargaining nor sovereignty at bay captures the inter-tem- poral dynamics of the bargaining process. Second, neither concept MNE–state bargaining has been at the core of IB research since its sufficiently accounts for the multitude of stakeholders and their socio- early years. From the seminal works of Vernon (1966) to classic economic context (Ramamurti, 2001; Stevens, Xie, & Peng, 2016). transaction-cost theorists (Teece, 1986; Williamson, 1967) to the ob- Third, both frameworks are MNE-centric and fail to account for possible solescing-bargaining literature (Ramamurti, 2001; Vachani, 1995), all mutual dependence in power relationships (Emerson, 1962). Our hol- have recognized the potential costs that host governments can impose istic MNE–state bargaining framework fills these theoretical gaps and on foreign MNEs. Whenever specific assets are deployed in a foreign serves to categorize, frame, and relate novel contributions to the topic. sovereign territory, the literature has argued, the investing MNE is As a second and more empirical contribution, we apply the bar- locked into the transaction and is vulnerable to ex-post opportunism by gaining framework to a specific case in the extractive industries, the host government (and other stakeholders) (Makhija, 1993).1 seeking to identify micro-strategies that MNEs can use to establish a At the same time, and in apparent contradiction to the obsolescing- sustainable power balance. Existing empirical explanations for con- bargaining argument, theorists have pondered over the sovereignty-at- straints on host-government bargaining predominantly focus on in- bay phenomenon: a gradual erosion of government power vis-à-vis dustry-, or MNE-level, attributes. A thorough exploratory analysis of MNEs (Kobrin, 2001; Vernon, 1971, 1981, 1991). Despite their shared bargaining strategies on a micro, or investment level, is missing.2 heritage, both arguments have developed somewhat in isolation, We use classic economic analysis in the tradition of Moran (1974), without theoretical integration or formalization. As a result, there has Jenkins (1986), and Woodhouse (2006) combined with an in-depth been little investment-level research on potential micro-strategies to qualitative case study of the Venezuelan oil industry to identify micro- influence the MNE–state power balance. In this article, we seek to fill strategies. Triangulating data from multiple sources, we categorize 12 these gaps, making two specific contributions. micro-strategies that have influenced the MNE–state bargaining process First, and on the theoretical side, we use the formalized metrics of and integrate them into our framework. Besides the above-mentioned power-dependence theory (Emerson, 1962) to formulate a dynamic, theoretical contributions, these empirical findings provide valuable multi-party framework of MNE–state bargaining that addresses three practical guidance for MNEs investing in high-risk contexts. theoretical limitations of existing conceptualizations: First, neither ⁎ Corresponding author. E-mail addresses: [email protected] (J. Müllner), [email protected] (J. Puck). 1 The relevance of obsolescing bargaining is supported by empirical data. Between 1972 and 2000, on average 1.5 MNE–state disputes were taken to the International Center for Settlement of Investment Disputes (ICSID). Since 2000, this figure has risen to 31 cases per year (ICSID, 2015). 2 This is surprising since other disciplines have proposed micro-strategies related to finance (Esty, 1999; Hainz & Kleimeier, 2012; M & llner, 2016; Vaaler et al., 2008), international law (Neumayer & Spess, 2005; Poulsen, 2010; Shadlen, 2005), or (non-market) strategy (Blumentritt & Nigh, 2002; Henisz & Zelner, 2003; Henisz, 2000; Henisz, 2011, 2014). http://dx.doi.org/10.1016/j.jwb.2017.07.003 Received 31 May 2016; Received in revised form 19 July 2017; Accepted 31 July 2017 Available online 09 August 2017 1090-9516/ © 2017 Elsevier Inc. All rights reserved. J. Müllner, J. Puck Journal of World Business 53 (2018) 15–26 Table 1 Literature overview of obsolescing bargaining and sovereignty at bay. Author (year) Methodology & perspective Contributions relevant to paper (Vernon, 1971) Book Origin of sovereignty at bay and obsolescing bargaining. (Moran, 1974) Multiple case study; Chilean copper industry Technological capabilities as a source of MNE bargaining power. (Vernon, 1981) Theoretical paper; macro-economic perspective Globalization, technological advantage, communication and transport efficiency improve bargaining power of MNEs. Governments respond by creating multilateral governance structures. (Fagre & Wells, 1982) Descriptive analysis; U.S. subsidiaries in Mexico Technology, product differentiation, exported intensity as sources of bargaining power for MNEs. Industry competition as a source of bargaining power for states. (Jenkins, 1986) Case study; Canadian National Oil Energy Program Local allies and home-government political support as sources of MNE bargaining power. (Kobrin, 1987) Empirical analysis; U.S. manufacturing subsidiaries in developing Technological advantages and global integration of host country can prevent countries obsolescing bargaining. (Vernon, 1991) Theoretical paper; macro-economic perspective Obsolescing bargaining and sovereignty at bay are co-evolutionary, dynamic processes in which both actors cooperate, conflict, and compromise. (Dicken, 1994) Theoretical paper; macro-economic perspective MNEs’ relational networks and competition between nations as sources of MNE bargaining power. (Thomas & Worrall, 1994) Formal model using game theory for sequential bargaining Future investments as a source of MNE bargaining power (repeated games). episodes (Vachani, 1995) Large sample empirical study; U.S., U.K., and European MNE-level determinants on bargaining success and the role of the country subsidiaries in India context. Introduces home–host-country ties and provides evidence for MNE- level sources of bargaining power. (Moon & Lado, 2000) Theoretical paper: Integration of resource-based view in Firm-specific resources provide the basis for a sustainable bargaining power for obsolescing bargaining the MNE. (Ramamurti, 2001) Theoretical paper; introducing two-tier bargaining Obsolescing bargaining is not a dyadic process but a two-tier, multi-party bargaining process (home countries of MNEs, multilateral institutions). (Kobrin, 2001) Theoretical paper; macro-economic and legal perspective Jurisdictional asymmetry (local governments vs. global MNEs). International investment law as a source of MNE bargaining power. (Doh & Ramamurti, 2003) Multiple case study; power projects in India Introduces financial approaches (project finance, financial engineering), multilateral agencies, and contractual strategies (trade agreements, fixed-term contracts) as sources of power. (Woodhouse, 2006) Multiple case study; multinational energy investments Legal and financial tools to counter obsolescing bargaining. Financial markets and project creditors as sources of bargaining power. (Gould & Winters, 2007) Case study; state–World Bank bargaining in the Chad–Cameroon Access to outside options are key to bargaining outcomes for parties. Dynamic oil pipeline project bargaining process with different power imbalances between different episodes. (Nebus & Rufin, 2010) Theoretical paper; Integration of network theory & Case Participants in international bargaining include nation-states, MNEs, NGOs, and study;privatization energy utility in Dominican Republic multilateral organizations. Introduces coalitions as strategic instrument. (Li et al., 2013) Multiple case study; natural resource investments in Tanzania Documents home-country support for Chinese MNEs (loans, social and financial investment programs). Two-tier and one-tier bargaining strategies. 2. Obsolescing bargaining and sovereignty at bay: two sides of a country ties (Vachani, 1995), repetitive investment rounds coin (Thomas & Worrall, 1994), or additional actors such as multinational or- ganizations and home-country governments (Ramamurti, 2001). Collec- Two partially conflicting streams of literature that date back to Vernon tively, the obsolescing-bargaining literature has identified a variety of (1971) dominate IB discourse on MNE–state bargaining. The obsolescing boundary conditions, but none of the frameworks amounts to a holistic bargaining argument describes the fundamental shift in power to a host- perspective, and, more importantly, there is a scarcity of evidence on country government once an investment is made in a foreign territory. As micro-strategies on an investment level.3 soon as irreversible assets are deployed, the host government can subse- Vernon (1981, 1991) initiated a second stream of literature often re- quently engage in hold-up (Woodhouse, 2006). The decisive sources of ferred to as sovereignty at bay that is somewhat juxtaposed with obsolescing power imbalance in obsolescing bargaining are the potential sunk costs for bargaining. The core argument is that the supra-territorial nature of MNEs an investing MNE. This sunk-cost argument is implicitly static in that it allows them to bypass the risk emanating from a local sovereign. Unlike does not consider potential changes in bargaining power over the lifetime host governments that are tied to their sovereign territory, MNEs can in- of an investment, or across several rounds of investments. Also, obsoles- vest in alternative locations, which results in superior bargaining power. In cing bargaining is dyadic and disregards other parties that might affect the sovereignty at bay, the primary source of power for MNEs is access to power balance between the focal actors (Li, Newenham-Kahindi, alternative investment locations. This access constitutes the theoretical Shapiro, & Chen, 2013; Ramamurti, 2001). Finally, the framework im- counterweight to the sunk-costs argument in obsolescing bargaining. plicitly assumes that the host government is not dependent on the in- The sovereignty-at-bay literature remained even more on a macro or vesting MNE and thus ignores possible mutual dependencies in the conceptual level, with little empiricism (Vernon, 1981, 1991). After its early MNE–state power relationship. successes, Vernon (1981) proclaimed a reemergence of the sovereign state These restrictive assumptions have somewhat impeded IB research brought about by inter-governmental cooperation in the 1980s. Later, Kobrin from theorizing on MNE bargaining strategies, especially on a micro, or (2001) picked up the discussion and introduced international legal frame- investment, level. Some boundary conditions to host-government bar- works as an important counterbalance to the sovereign-power advantage. gaining identified in early literature were related to industry- or MNE-level Table 1 summarizes the obsolescing-bargaining and sovereignty-at- aspects such as technological intensity (Kobrin, 1987; Moran, 1974), bay literatures. Overall, studies on boundary conditions in both product diversity, market access, capital availability (Fagre & Wells, 1982), ties to local business community (Jenkins, 1986), and home-country po- litical support (Jenkins, 1986; Li et al., 2013). Other authors extended 3 Notable exceptions are Doh and Ramamurti (2003) and Woodhouse (2006), who obsolescing bargaining to include dyadic elements such as home–host- argued that financial and contractual strategies on an investment level can protect from obsolescing bargaining. 16 J. Müllner, J. Puck Journal of World Business 53 (2018) 15–26 literature streams have remained fragmented and lack theoretical in- its own dependence on the host government is lower than vice versa. tegration. A holistic framework that captures both theoretical per- This results in a negative PB coefficient (sovereignty at bay). spectives, the multiplicity of actors, and the dynamics of the bargaining Alternatively, in the classic obsolescing-bargaining situation, govern- process is still missing (Stevens et al., 2016). ment power exceeds MNE power and the coefficient is positive. A state In the following, we propose power-dependence theory as a theo- of power imbalance may be present at a certain time (t0), but exerting a retical link between obsolescing bargaining and sovereignty at bay and power imbalance may compromise the future ability to access alter- as a solution to the limitations of both frameworks. native investments (AIt1) and even create sunk costs (SCt1). For ex- ample, an actor may compromise its ability to access alternative in- 3. Power dependence theory: a missing link vestments (AIt1) if it acts upon the incentive to capitalize on a power imbalance for a given investment (Gould & Winters, 2007). Excessive Power dependence theory (Bae & Gargiulo, 2004; Cook, Emerson, bargaining may also trigger retaliatory action from the counterpart and Gillmore, & Yamagishi, 1983; Emerson, 1962) offers a uniquely valu- other stakeholders, which result in considerable sunk costs (SCt1). The able and yet unexplored link between sovereignty at bay and obsoles- model suggests that changes in any of the four parameters can induce cing bargaining. The main strength of power-dependence theory is the power imbalance. The ultimate distribution of bargaining power is a formalized conceptualization of dependence: function of all four parameters, which is contextual in that it is de- pendent on who invests (AIMNE) what types of assets (SCMNE) in which The dependence of actor A upon actor B is (1) directly proportional host country (SCGOV & AIGOV). Achieving a sustainable bargaining po- to A’s motivational investment in goals mediated by B, and (2) in- sition requires that all four components be balanced. versely proportional to the availability of these goals to A outside As outlined above, we know little about the micro-strategies that the A-B relationship. MNEs use to influence the different parameters of the framework to Assuming that actor A is an MNE investing in country B, the moti- achieve a sustainable power balance. There is a pressing need for in- vational investment of A is determined by (a) the sunk costs of the depth, qualitative, and exploratory analysis of micro-bargaining stra- investment (SCMNE) and (b) its ability to access alternative investments tegies, which Jenkins aptly described in 1986: outside of country B (AIMNE). The first argument, sunk costs, closely While the obsolescing bargaining may explain why a host govern- relates to obsolescing bargaining, as it determines the degree to which ment feels compelled to exert control over MNCs, it disregards op- the host government can engage in hold-up. The second argument, erative factors that may prevent governments from implementing access to alternative investments, is the cornerstone of sovereignty at such plans. (p. 165) bay. Following Emerson’s (1962) definition, an MNE’s dependence on the host government can be calculated according to Eq. (1): In the next section, we apply our holistic MNE–state bargaining t framework to the case of the Venezuelan oil sector to identify such t SCMNE DMNE − GOV = t micro-strategies of MNEs, providing the second, more empirical con- AIMNE (1) tribution. Eq. (1) integrates obsolescing bargaining (SCMNE) in the numerator and sovereignty at bay (AIMNE) in the denominator. Therein, it closely resembles the original intent of Vernon (1971). Including access to al- 4. Methodology ternative investments (AIMNE) opens up the dyadic nature of MNE–state bargaining by acknowledging socio-economic restrictions resulting 4.1. Empirical context from third parties. Additionally, acknowledging that access to alter- native investments (AIMNE), and sunk costs (SCMNE), may vary over time Extractive industries are a fitting context for the purpose of this (t) allows for a dynamic interpretation of bargaining. The equation study. They require large-scale investments that amortize over several highlights that forces of obsolescing bargaining and sovereignty at bay decades. Often, investments are in highly uncertain host countries that coexist within a particular investment and that the ultimate power do not provide institutional safeguards (Li et al., 2013). The ecologic, balance is inherently contextual. social, and economic impact on the host countries results in intense However, at this stage, the equation does not acknowledge potential scrutiny (Doh & Ramamurti, 2003). Domestic public ownership collides differences in the host government’s power properties. Formally, the with private foreign commercialization, making foreign MNEs easy same MNE (AIMNE) making two investments with similar sunk costs targets for political populism and stakeholder activism. Extractive in- (SCMNE) in two potentially different host-country contexts results in the vestments yield rather stable cash flows that are relatively easy to same power balance. Thus, the framework does not include the pivotal market (Doh, McGuire, & Ozaki, 2015). Consequently, stakeholders argument of power-dependence theory: mutual dependence have a strong incentive and a superior ability to engage in obsolescing (Casciaro & Piskorski, 2005). Essentially, it is assumed that host gov- bargaining.4 ernments themselves do not “put into place specialized investments To identify MNE micro-strategies, we select the case of the dedicated to the multinational enterprise” (SCGOV) (Teece, 1986, p. 42). Venezuelan oil industry, which saw repeated expropriations (Cuervo- In addition, it disregards possible differences in access to alternative Cazurra, Andersson, Brannen, Nielsen, & Rebecca Reuber, 2016). We investments for host governments (AIGOV). Power is “a property of so- focus on a single country to avoid problems from cross-country varia- cial relation; it is not an actor attribute” (Emerson, 1962, p. 32), and tion in the political, economic, and social environment. Venezuela any bargaining conceptualization must include the counterparty’s de- ranks among the countries with the lowest political stability (Doh, pendence (Eqs. (2) and (3)): Lawton, & Rajwani, 2012) but disposes of the largest known natural oil reserves worldwide (OEPC, 2015). Most importantly, the second round t t PB t = DMNE − GOV − DGOV − MNE (2) of expropriation affected numerous foreign oil MNEs, which pursued t t very different bargaining strategies and achieved very different out- SCMNE SCGOV PB t = t − t comes. This provides a quasi-natural experiment for the analysis of AIMNE AIGOV (3) micro-strategies. The extreme nature of the case means that the Unlike the obsolescing-bargaining and sovereignty-at-bay litera- tures, this holistic MNE–state bargaining framework does not assume a 4 In practice, 25% of ICSID disputes were recorded in extractive industries (oil, generalizable a-priori bargaining-power advantage for any actor. At any gas, & mining), and 30% involved governments of South and Central America (ICSID, given point in time (t), the power balance is in favor of the MNE when 2015). 17 J. Müllner, J. Puck Journal of World Business 53 (2018) 15–26 Table 2 Data sources and triangulation of micro-strategies. Micro-strategy/Source type Academic papers Court documentation Interviews Policy sources WikiLeaks cables Industry and press sources Total Sunk cost MNE 38 3 12 12 12 31 126 Financial engineering 19 1 6 2 4 5 55 Legal recovery strategies 15 2 3 2 3 10 35 Home-country CPA 4 3 8 5 16 36 Access to alternative investments MNE 20 0 12 3 5 43 83 Technological advantage 11 4 3 4 37 59 General CPA & CSR 9 8 1 6 24 Sunk cost GOV 26 3 11 9 14 35 98 Local partners 8 1 4 4 4 21 Local CSR by MNE and home country 8 2 6 1 9 26 Technology drain 3 1 3 2 6 14 29 Legal retaliation 7 1 2 1 3 8 22 Access to alternative investments GOV 23 2 10 13 23 30 101 Coalition strategies 15 6 12 20 21 74 Legal collateral 6 1 1 8 Distribution dependence 2 1 3 1 3 9 19 Total 107 8 45 37 54 139 408 Note: CPA = corporate political activities; CSR = corporate social responsibility; GOV = government; MNE = multinational enterprise. intensity of the phenomenon is higher and the mechanisms within it are diplomacy. However, it is important to concede that we have no means more observable and multifaceted. As a result, the “process of interest of verifying the accuracy of WikiLeaks data. In sum, we can rely on [obsolescing bargaining] is transparently observable” (Eisenhardt, more than 9900 pages of documentation for our qualitative analysis. 1989, p. 537). 4.3. Data analysis 4.2. Data collection The two authors independently screened data for instances of micro- Identifying investment-level micro-strategies in an exploratory strategies, in an iterative process. Following our holistic MNE–state manner requires a detailed, qualitative approach (Birkinshaw, bargaining framework, we coded them as either (a) a sunk-costs Brannen, & Tung, 2011). To capture the complexity of bargaining, we strategy or (b) an access strategy for (1) the MNE or (2) the host gov- use a case-study approach, as we focus on “understanding the dynamics ernment, thereby creating a systematic micro-strategy portfolio from present within single settings” (Eisenhardt, 1989 p. 534). Fortunately, the data. Following discussions, some micro-strategies contained ele- the case of the Venezuelan oil industry is unique in the degree to which ments of two or even three mechanisms, making them particularly it is publicly documented. This allows us to draw on a large variety of important from a strategic perspective (i.e., possible win–win strate- information sources and to achieve a triangulated and “synergistic view gies). For each strategy, we sought to identify instances in different data of evidence” (Eisenhardt, 1989, p. 533). sources for cross-validation (Yin, 1994). Overall, our analysis yielded First, we obtained court documentation from the International 408 instances of observed micro-strategies, which we merged into 12 Center for Settlement of Investment Disputes (ICSID) and the micro-strategy categories. Table 2 provides a distribution of micro- International Chamber of Commerce (ICC) amounting to more than strategies and data sources. Tables 3 and 4 in the online Appendix in- 1400 pages. Second, we identified scholarly work on the Venezuelan oil clude exemplary quotes for each of the micro-strategies and an illus- sector using journal databases such as Web of Science, Social Science trative example of data triangulation. Research Network, and Google Scholar (Boué, 2014; Esty, 1999; Manzano & Monaldi, 2008, 2010; Monaldi, 2001; Monaldi, Gonzales, 4.4. Case reporting Obuchi, & Penfold, 2006; Schill, 2012). By the nature of the source, data from such publications adhere to scientific principles of objectivity and Following Eisenhardt and Graebner (2007), we first summarize the integrity. Third, we complement this with publications from interna- case chronologically. We seek to link changes to bargaining power to tional organizations such as the World Bank and the Inter-American the variables used in our bargaining framework. In a second step, and as Development Bank (UNCTD, 2015a, 2015b) as well as national political recommended by Yin (1994), we use our framework to present and documentation provided, for example, by the U.S. Congressional Re- discuss the micro-strategies identified in our case. search Service. Fourth, we use the Factiva Press database to collect 5336 newspaper articles (e.g., Washington Post, The Guardian, The 5. The case of the venezuelan oil sector Financial Times), and industry reports (e.g., Petroleum Review, Oil Market Watch, Moody’s Investor Service, Hydro Carbos Technology). 5.1. Episode 1: the golden age Fifth, we interviewed independent experts. Interviewee 1 was a con- sultant in several arbitration cases. Interviewee 2 was a former advisor In the 1930s, Venezuela became the largest exporter of oil. MNEs in- to the President of the United States on natural resource issues, and a vested considerable amounts, incurring extensive potential sunk costs long-time government relationship manager at an oil and gas MNE. (SCMNE). Encouraged by the dependence of the Allied forces on oil during Interviewee 3 was an academic scholar with a strong focus on oil and the Second World War, the bargaining position of the Venezuelan gov- gas in Latin America, and Interviewee 4 was a retired consultant to ernment improved for lack of alternative sources (AIMNE) (Monaldi, 2001). foreign MNEs and Petróleos de Venezuela, S.A. (PDVSA) in Venezuela. In 1943, the Hydrocarbons Law established the concession scheme, under Finally, we searched WikiLeaks for references to companies, oil fields, which revenues were evenly split between foreign MNEs and the state or government officials. Diplomatic cables contain confidential and (Espinasa, 1995; Monaldi, 2001). Oil revenues quickly became the largest privately communicated information that is almost impossible to obtain income source for the government of Venezuela and gradually increased through other sources of data. They are free of social, normative, and its dependence on foreign MNEs (AIGOV), initiating a period of sustained interviewer biases and allow us to reconstruct behind-the-scenes power balance (Monaldi, 2001). 18 J. Müllner, J. Puck Journal of World Business 53 (2018) 15–26 5.2. Episode 2: obsolescing bargaining I – La gran Venezuela 635.5 million USD using tax increases, interruptions of production, and commercial fines (Cable, 2007d) (SCMNE). In January 2007, the Vene- Between 1943 and 1958, government revenues from oil stagnated. zuelan government announced the second round of nationalization of With the first democratic election in 1958, the newly established gov- oil assets. ernment unilaterally increased its share of oil revenues to 65%. The incoming government had no previous access to oil revenues and no 5.5. Episode 5: epilogue – International arbitration sunk costs (SCGOV), which tilted the bargaining power in its favor. As a result, a long series of bargaining rounds began, and by 1975 the share Within days, U.S. oil producers met with Venezuelan oil minister of government revenues climbed to 95%. Raffael Ramirez and with U.S. officials (Vidal, 2005; Washington Lacking sustainable profit margins, MNEs neglected to maintain & Post, 2005). Eventually, Statoil, BP, Chevron, ENI, and Total agreed wells and actively sought alternative investment destinations (AIMNE). to the terms, but ExxonMobil and ConocoPhillips refused (Interview 3, By 1971, Venezuela lost its position as largest oil exporter to Saudi 2016; Stratfor, 2007). The combined dispute value reportedly Arabia (The Shift Project, 2015). In need of fiscal revenues, Venezuelan amounted to 45 billion USD (Boué, 2013). president Carlos Andrés Pérez nationalized oil operations as part of his ExxonMobil and ConocoPhillips followed several legal strategies to economic plan La Gran Venezuela, closing the first obsolescing-bar- recover their investments (SCMNE) (ICC, 2011). ExxonMobil was gaining episode. PDVSA was founded in the hope that it could profit- awarded 1.6 billion from the ICSID for Venezuela’s breach of the ably run oil operations without foreign profit sharing. PDVSA, however, Netherlands–Venezuela BIT in 2014 (SCMNE), and ICC ordered another had poor technological capabilities and continued to face capital con- fine of 747 million USD (SCMNE) (Ellsworth & Parraga, 2012; straints as the Venezuelan government skimmed the profits necessary Pitts & Kurmanaev, 2013). In 2012, ICC awarded ConocoPhillips a 66.8 for long-term investment. Over more than 10 years PDVSA struggled million USD settlement (SCMNE). internally, which eventually led to the political realization that access Apart from the legal aftermath, important economic developments to foreign investment (AIGOV) was a precondition to re-establishing the took place. In 2006, Venezuela saw a net decline of 40% in incoming Venezuelan oil sector’s competitiveness. FDI compared with 2005 (AIGOV). Venezuela’s political isolation re- duced its access to downstream buyers of oil (AIGOV). Eventually, 5.3. Episode 3: sovereignty at bay I – Apertura petrolera Venezuela attracted some FDI—however, on quite unfavorable terms. In one WikiLeaks cable, ENI officials are quoted pressuring Venezuelan From 1989 to 1999, different governments tried to establish a new Oil Minister Rafael Rodriguez before the official deal-closing ceremony: institutional framework to re-attract foreign investors from a con- “Take it or leave it, I can get on my plane and move on” (Caroll, 2010). siderably weakened bargaining position. The new framework bypassed Most recent developments in Venezuela do not indicate a return to a the Venezuelan government and established PDVSA as a private partner stable equilibrium. Rather, the Venezuelan economy has spiraled of incoming FDI. This, in theory, allowed foreign MNEs to recover in- downward towards, possibly, another episode of sovereignty at bay. vestments from PDVSA in case of government intervention (SCMNE) The historical description of the case allows us to make several (ICC, 2011; Monaldi, 2001). Such legal collateral proved particularly theoretically intriguing observations about the bargaining process: fruitful after PDVSA acquired the U.S. downstream company Citigo First, and as supported by our theoretical arguments, the MNE–state (Sullivan, 2008). In three auctions (1991, 1992, 1997) 34 of 41 oil power relationship was not static over time. Rather, episodes of mutual fields (Monaldi, 2001) were sold to foreign MNEs, including Total and dependence (episode 1) were followed by episodes of power imbalance Statoil, ConocoPhillips, ExxonMobil, BP, ENI, and Petrocanada (Cable, for the host government (episodes 2 and 4) and for MNEs (episode 3). 2007b). The new framework using PDVSA as collateral reestablished a The most radical changes in bargaining power were initiated by state of mutual dependence. changes in governments, whereby the sunk costs (SCGOV) were not transferred to the incoming regime. Second, and also in line with our 5.4. Episode 4: obsolescing bargaining II – Bolivarian revolution theoretical framework, the bargaining process was not restricted to the two parties but involved a multitude of secondary actors such as Late in 1998, Hugo Chavez seized power and initiated the PDVSA, misiones bolivarianas, and international courts. Fig. 1 sys- Bolivarian revolution. At first, the government concentrated on con- tematically illustrates the bargaining dynamics in the case, important solidating power over PDVSA. In a diplomatic cable, PDVSA executives actors, and the mechanisms that contributed to altering the power complained of “contracts being canceled capriciously” (Cable, 2004). balance. PDVSA called strikes that lasted for months and that led to social unrest. Most importantly, the micro-strategies used and the subsequent In 2002, Chavez replaced the board of PDVSA and fired approximately outcomes of bargaining appear to differ between MNEs, providing a 19,000 workers (Sullivan, 2008). The controversy involved high sunk quasi-experimental setting for in-depth qualitative analysis. In the fol- costs, public outrage, and considerable knowledge drain (SCGOV), but lowing, we present evidence on micro-strategies from the case, linking eventually it secured the government’s access to oil resources (AIGOV). them to our holistic bargaining framework. One diplomatic cable reads: “the new PDVSA is staffed by fervent supporters of the ‘revolution’ or by opportunists who would be un- 6. Micro-bargaining strategies in the case of the Venezuelan oil willing to lose their access to the spoils of corruption” (Cable, 2004). industry President Chavez used increasing oil revenues to launch social devel- opment programs, the so-called misiones bolivarianas, to maintain sup- 6.1. Micro-strategies reducing sunk costs for MNEs (SCMNE) port among his electorate (Sullivan, 2008). Tensions between foreign oil MNEs and the Venezuelan government Strategies to reduce sunk costs for the MNE aim at ex-ante or ex-post increased. The U.S. openly discussed economic sanctions in retaliation mechanisms to recover sunk costs of the investment. We observe three (SCGOV), and Chavez threatened to cut the U.S. off from Venezuelan oil different types of sunk-cost micro-strategies: financial engineering, (Cable, 2004). At the same time, U.S. oil MNEs intensely reached out to legal recourse, and home-country corporate political activities (CPA). U.S. politicians to discuss political levers to avoid expropriation Financial engineering allows MNEs to recover sunk costs (SCMNE) by (Interview 4, 2016). In 2004, the Venezuelan government unilaterally creating financial structures that mitigate the risk of expropriation and revoked the preferential Royalty Reduction Agreement, and in 2005 it minimize the loss given expropriation. In episode 3, we observe a new imposed additional extraction taxes (ICC, 2011) (SCMNE). Foreign MNEs investment strategy: project finance. This involves the creation of a reported attempts of creeping expropriation amounting to more than separate legal entity protecting MNEs from financial distress in the 19 J. Müllner, J. Puck Journal of World Business 53 (2018) 15–26 Fig. 1. Dynamics of MNE–state bargaining in the case of Venezuela. Note: PB = Power-balance; SC = sunk cost; AI = access to alternative investments; GOV = government; MNE = multinational enterprise; PDVSA = Petróleos de Venezuela, S.A. event of expropriation (Brealey, Cooper, & Habib, 1996; Gatti, 2013; bondholders maximized the financial sunk costs (SCGOV) for the Vene- Nevitt & Fabozzi, 2000). The main intent of project finance is to reduce zuelan government and restricted Venezuela’s access to further fi- sunk costs for the MNE (SCMNE) but it also has an indirect, risk-miti- nancial funding (AIGOV). Secondary data strongly support the validity of gating effect due to an increase in the host government’s sunk costs financial recovery strategies: In a study of global project finance ac- (SCGOV) and ultimately a reduction in its access to alternative invest- tivity, Moody’s (2014) found that the average recovery rate for loans ments (AIGOV). Non-recourse loans align incentives between debt pro- was 80% and the most likely outcome was a 100% recovery. viders and the investing MNE. Without the opportunity to reclaim loans Second, legal recovery micro-strategies reduce sunk costs (SCMNE). from MNEs, creditors focus on recovering debt from the expropriating Similar to financial engineering, legal recoveries are win–win micro-stra- government (SCGOV) (Beidleman, Fletcher, & Vesbosky, 1990; tegies, as a violation of international agreements can be accompanied by Doh & Ramamurti, 2003; James & Vaaler, 2014). In the case of Vene- considerable sunk costs for the host government (SCGOV) and may threaten zuela, the otherwise very nationalistic government was, and remains its access to future investments (AIGOV). ConocoPhillips and ExxonMobil very eagerto accommodate project creditors in order to maintain access were awarded substantial amounts in international arbitration. Such ar- to financial markets (Economist, 2017; The Financial Times, 2007). As a bitration opportunities are not available by default, and MNEs structured result, a large share of the capital was recovered rather quickly investments ex-ante in a way to secure ex-post arbitration (Interview 1, (Interview 3, 2016). 2016; Pinkham & Peng, 2017). ExxonMobil, for example, financed the Among the oil fields expropriated was the 2.4 billion USD investments via holding companies in the Netherlands, giving them access Petrozuata oil field. It included numerous financial engineering micro- to a bilateral trade agreement, which was not available from the U.S. strategies, such as a provision for international arbitration and an (Interview 1, 2016; Weiniger & Villaggi, 2014).5 escrow account designed to keep oil revenues out of Venezuela (SCGOV). Last, the case provides indications that home-country CPA activities At the same time, the foreign MNEs’ balance sheets were protected by can reduce sunk costs for the MNE by generating home-country support non-recourse provisions (SCMNE). Petrozuata was financed using public and potential financial compensation. CPA refers to companies’ con- bonds that included severe penalties in the event of a change in ma- nections with political actors, such as governments, political institu- jority ownership. ICC court documentation includes a full section on the tions, administrations, or NGOs (for a review, see Boddewyn, 1994; costs resulting from the immediate need to refinance the oil projects for Rajwani & Liedong, 2015; White et al., 2015). Because of the strategic the government of Venezuela (ICC, 2011, p. 447) (SCGOV). Unlike Pet- nature of the oil industry, close cooperation between governments and rozuata, the 3.5 billion USD oil field Hamaca was financed using syn- MNEs benefits both parties, resulting in possible bailouts in the most dicated bank loans. ConocoPhillips assembled a global syndicate in- extreme case (Interview 1, 2016).6 In the case of ExxonMobil, an ex- cluding Royal Bank of Scotland and Barclays, BNP Paribas, ING Bank, plicit promise was issued by U.S. officials that “the option of drawing WestLB and Bayrische Landesbank, and Bank of Tokyo-Mitsubishi. They obtained guarantees from Ex-Im banks in the U.S. and Canada. The political leverage amassed in the syndicate was impressive. Together 5 In a similar case in the mining industry, Crystallex was awarded 1.386 billion USD in the home countries of banks accounted for more than 70% of FDI in compensation based on the violation of the Canada–Venezuela bilateral trade agreement Venezuela and 73% of incoming foreign aid (Aid Data Beta, 2015; (ICSID, 2016; Interview 1, 2016). 6 James (1990) showed how implicit government guarantees can increase the bar- UNCTD, 2015a, 2015b). Essentially, the coalitions of banks and gaining power of banks and MNEs in sovereign defaults. 20 J. Müllner, J. Puck Journal of World Business 53 (2018) 15–26 upon the Strategic Petroleum Reserve would be available to calm the Equator Principles, a voluntary set of CSR practices (Interview 1, markets” (Cable, 2004). The State Department formally acknowledged 2016). In summary, technological advantages, CPA, and CSR may de- full support of Exxon’s legal actions against the Venezuelan government crease obstacles to alternative investments for MNEs (AIMNE). (Gelsi, 2008) and actively encouraged cooperation between oil produ- cers (Interview 2, 2016; Interview 4, 2016). In sum, micro-strategies reducing sunk costs allow MNEs to recover 6.3. Micro-Strategies increasing sunk costs for the host government (SCGOV) part of their investment from the foreign location through financial engineering, through legal recourse, and by creating a network of MNEs may also seek to influence the power balance by increasing support among creditors and home-country politicians. Some of these the sunk costs of the host government (SCGOV). We find four distinct micro-strategies can also affect the government’s sunk costs (SCGOV) micro-strategies in this case: local partners, local CSR, technological and its opportunities to attract new investment (AIGOV), making them dependence, and legal retaliation. particularly valuable for foreign MNEs. Foreign investors can allow local partners to buy into the investment (Henisz, Dorobantu, & Nartey, 2014; Jenkins, 1986; Vernon, 1983).7 In 6.2. Micro-strategies improving access to alternative investments for MNEs the case of Venezuela, PDVSA strongly acted as a local ally, calling (AIMNE) strikes and spearheading political demonstrations. The government’s overpowering of PDVSA resulted in considerable economic and political Beyond reducing the sunk costs of an investment, MNEs can im- costs, substantial interruptions of oil production, and considerable loss prove their access to alternative investments (AIMNE). In the case, we of human capital (SCGOV) (PIW, 2008). Even after the government ta- find strong evidence for two such access strategies: technological ad- keover, former PDVSA personnel actively supported MNEs through civil vantage and foreign CPA and corporate social responsibility (CSR). society groups such as Gente de Petroleo (Cable, 2004). In the case, the MNEs can develop technological capabilities and resources (Barney, local partner also proved helpful in recovering sunk costs for the MNEs 1991; Henisz & Williamson, 1999; Ramamurti, 2001) that allow them to in the legal aftermath (SCMNE). While government assets in foreign access resources or destinations that are inaccessible to their competi- countries are legally protected, the assets of a private company can be tors. In the case, exploration technology is an explicit strategic focus of used as legal collateral (Interview 1, 2016). A-priori contracts with ExxonMobil (2016): PDVSA specified compensating MNEs for economic damages due to government intervention. Hence, ExxonMobil could file separate claims The strength of our global organization allows us to explore for and against both the government of Venezuela and the local partner PDVSA capture all resource types, across all geological and geographical (Interview 1, 2016). environments, using industry-leading technology and capabilities. A second means of generating sunk costs for the host government The horizontal wells developed by both ExxonMobil and (SCGOV) is local CSR. One of the interviewees strongly stressed the ConocoPhillips allowed them to access oil fields that were inaccessible MNEs’ need to obtain social license to operate since it determines the to other oil MNEs, such as the Arctic Sakhalin Island in Russia, which degree to which the general public supports the MNEs’ operations or, in involved a 7.7-mile horizontal well (ExxonMobil, 2013). Technological the worst case, pushes the government to renege on contracts leadership also opened up alternative investment destinations in Ca- (Interview 1, 2016). In the case, there is documentation of CSR activity nada, where specialized technology was used to refine oil sands by the U.S. government, through the U.S. Agency for International (Jones & Haggett, 2007). Development and the National Endowment for Democracy (Sullivan, Beyond the technological advantage, CPA and CSR can be used as 2008). We also observe evidence of CSR on the part of the MNEs. Ex- access strategies (AIMNE) (Boddewyn, 1994; Doh et al., 2012; Faccio, xonMobil established a CSR presence in 1945 (Creole Petroleum Cor- 2006; Hillman, Keim, & Schuler, 2004; Marquis & Raynard, 2015; poration), and Shell is credited with the first major CSR incentive in Sawant, 2012). As such, CPA and CSR constitute another win–win Venezuela, in 1952. In one instance, a Shell official is quoted saying strategy, offering both domestic benefits for recovering investments that CSR efforts “delayed the nationalization of the oil industry by at (SCMNE) and benefits for accessing foreign investment locations least a decade” (L’Etang, Lugo-Ocando, & Ahmad, 2011. p. 182). The (AIMNE). In the case, there is substantial evidence of home-country case also highlights a limitation of CSR. In a strategic move, Chavez political support in getting access to alternative investments. Following closed CSR as a micro-strategy for foreign MNEs by nationalizing social the expropriation, the U.S. government actively supported MNEs to policy under the misiones bolivarianas (Interview 3, 2016; Rodríguez, diversify to other regions, most prominently in the Caspian region, Morales, & Monaldi, 2012; Sullivan, 2008). Russia, and West Africa. A congressional report stated: “The U.S. gov- Additional sunk costs for the host government (SCGOV) were related ernment can use its political influence to try to encourage nations […] to the loss of technological know-how. MNEs actively attempted to as a condition for trade and aid agreements” (Pirog, 2007, p. 1). In the maintain vital competencies and technologies within MNE boundaries. same year, the U.S. Congress (2007) approved a bill to “increase co- After nationalization, ExxonMobil and ConocoPhillips withdrew their operation on energy issues between the United States Government and core personnel (SCGOV) (Millard, 2007a). Within days of nationaliza- foreign governments” (p. 1). The bill established the Western Hemi- tion, several oil fields reported staffing problems. PDVSA, on the other sphere Energy Cooperation Forum, intended to coordinate energy hand, had trouble convincing operating personnel to accept lower sal- policy across importing countries (Sullivan, Seelke, & Rush, 2008). Po- aries under the state-owned regime (Millard, 2007a). There were also litical access strategies were also strongly supported by two of our in- reports of hostile hiring attempts of key PDVSA staff by U.S. and Saudi terviewees who were aware of, or involved in the political support oil companies (Millard, 2007b) (SCGOV). Government sunk costs were given to, U.S. MNEs in Venezuela (Interview 2, 2016; Interview 4, not limited to engineering staff but also included costs from reduced 2016). marketing capacity. A diplomatic cable concluded: “PDVSA’s current Evidence on CSR as a means of improving access to alternative in- marketing personnel would probably not be up to the task of placing vestment locations (AIMNE) is found in academic research (Du & Vieira, over a million barrels of oil a day onto the spot market” (Cable, 2004). 2012; Henisz & Zelner, 2005; Li et al., 2013) and among industry ex- Most importantly, however, MNEs maintained control over refining perts (Cort, 2008). Many low-risk destinations, such as Norway and know-how. A diplomatic cable quoted Chevron Latin America President Canada, have strict social and environmental criteria, making CSR a precondition to accessing such resources (Interview 1, 2016). At the 7 Jenkins (1986) described in detail how U.S. MNEs used local opposition, business same time, the CSR track record affects MNEs’ access to funding from communities, and federal governments to generate opposition to the National Oil Energy major creditors or the World Bank (AIMNE), which require adherence to Program in Canada, eventually achieving a gradual erosion of the government decree. 21 J. Müllner, J. Puck Journal of World Business 53 (2018) 15–26 Ali Moshiri saying that in the event of nationalization, “Production to buyers (AIGOV). The Chalmette refinery, which specialized in refining would decrease by 10% in volume terms due to inefficiency but by 25% heavy crude oil, was partially owned by ExxonMobil and outside in terms of value” (Cable, 2007c). One of the interviewees acknowl- Venezuelan jurisdiction (ICC, 2011; Interview 1, 2016). PDVSA lacked edged similar micro-strategies also in the mining industry, in which the necessary in-house capabilities, and diverting oil to other refineries MNEs refine raw materials on site only to a stage of mining concentrate, would have required oil tankers, storage, suitable port facilities, and which is easier to transport but requires specialized smelting technol- local distributers (Cable, 2004) (SCGOV). Through refining technology, ogies outside host-government jurisdiction (Interview 1, 2016; MNEs were able to block the host government’s access to buyers Pinkham & Peng, 2017). (AIGOV), or at least force them to sell unrefined oil at high discounts Finally, some oil MNEs followed legal retaliation micro-strategies to (Cable, 2007a, 2008; Interview 1, 2016). Similarly, the supply and increase sunk costs for the Venezuelan government. Legal retaliation offtake agreements involved in project finance (Beidleman et al., 1990; strategies differ from legal recovery strategies in that their primary Farrell, 2003; Miller, Lessard, & Floricel, 2001; Sorge, 2004; objective is to block operations of the host government rather than to Subramanian & Tung, 2016; Vaaler, James, & Aguilera, 2008) had to be recover sunk costs of the MNE (SCMNE). In the High Court of Justice in renegotiated, which blocked access to important components of the London, ExxonMobil obtained a temporary freezing injunction on al- supply chain for PDVSA. most 12 billion USD of PDVSA assets (Cable, 2008; ICC, 2011, p. 21) In summary, access micro-strategies are aimed at restricting the host and blocked 300 million USD of PDVSA funds in escrow accounts government’s access to alternative investors and buyers. They involve (SCGOV). coalitions with alternative investors, establishing legal collateral, or In summary, government sunk-cost micro-strategies aim to increase exercising control over distribution channels. A summary and descrip- sunk costs for the host government by creating local allies (partners or tion of all identified micro-strategies is provided in Table 3. the general public), maximizing damages resulting from the drain of key technological capabilities or through legal retaliation. 7. Discussion In this paper, we contribute to IB research in two ways. First, we 6.4. Micro-Strategies restricting access to alternative investments for the develop a dynamic, multi-party MNE–state bargaining framework Host Government (AIGOV) based on power-dependence theory. The framework combines argu- ments from the obsolescing-bargaining and sovereignty-at-bay litera- The objective of the last micro-strategy is to block the host-country tures into a formal model that recognizes mutual dependence. government’s access to future investors and buyers by using macro According to the framework, there are four mechanisms which MNEs dependencies of host governments (Alcacer & Ingram, 2013; can influence to alter the power balance. The first two mechanisms aim Rangan & Sengul, 2009). at reducing the MNEs’ dependence on the host government by means of MNEs can establish coalitions with economic and political actors (1) reducing sunk costs from the investment (SCMNE) or (2) improving (i.e., inter-governmental organizations) or tap into existing multilateral access to alternative investments (AIMNE). In addition, the power bal- frameworks such as bilateral trade agreements (Alcacer & Ingram, ance is determined by the host government’s dependence on the MNE 2013; Chowdhry, 1991; Pinkham & Peng, 2017). The case shows strong due to (3) sunk costs (SCGOV) and (4) its own access to alternative in- political support in the U.S. Congress (2008) for isolating Venezuela vestments (AIGOV). The four mechanisms vary over time and jointly internationally.8 In May 2011, the U.S. officially sanctioned PDVSA determine the power balance between the parties. Any variation can (AIGOV) (Quinn & Daniel, 2011). More importantly, on the investment create shifts that initiate bargaining episodes in either direction. Thus, level, similar coalition strategies were used. MNEs built strong coali- no single strategy confers a sustainable power balance in a dynamic tions with foreign partners and banks to create a “counterforce that may setting. be able to enlist the support of a number of different governments” Our holistic MNE–state bargaining framework lends itself to theo- (Vernon, 1983, p. 203). The most important means of establishing rizing on boundary conditions of bargaining in different contexts. coalitions was related to co-financing. Investments were financially Industries differ in the degree to which investments involve sunk costs structured to increase global political leverage and incentivize other (SCMNE), and the MNEs within these industries differ in their access to nations to pressure Venezuela. Using project bonds and global banking alternative investment locations (AIMNE). At the same time, host gov- syndicates, MNEs ensured that the projects were deeply embedded in ernments vary in terms of possible sunk costs (SCGOV) and access to financial markets, impeding the host government’s access to new ca- alternative sources of investment (AIGOV). The key implication of the pital and investors (AIGOV) (Hydrocarbons Technology, 2007). Banks, as framework is that, in isolation, none of the mechanisms can fully ex- opposed to equity investors, have superior capacity to impose sanctions, plain MNE–state bargaining. As indicated by Kobrin (1987), technolo- seize collateral, disrupt international capital-market transactions, and gical intensity in a specific industry may lead to a situation in which track down foreign assets (James, 1990). In the case, the government of MNEs can maintain a sustainable state of superior bargaining power Venezuela faced severe refinancing problems. In 2016, PDVSA issued a (sovereignty at bay) because of their preferential access to alternative liquidity warning and faced bankruptcy after failing to place necessary investment locations (AIMNE). Equally, full control of distribution in an debt-swaps in financial markets (Dye, 2016). industry (SCGOV) or complete economic dependence on the part of the Another MNE strategy to restrict the host government’s access to host country (SCGOV) may prohibit a host government from engaging in investment is to access existing legal investment frameworks. Bilateral any sort of bargaining. In such contexts, traditional obsolescing-bar- or multilateral investment agreements provide legal collateral that can gaining frameworks may overstate the ability of host governments to be leveraged by investing MNEs. In the case, ExxonMobil financed its alter the rules of the game. On the other extreme, governments with investment via holding companies in the Netherlands, reducing abundant access to alternative investors (AIGOV) and low sunk costs Venezuela’s ability to attract future investments under the agreement (SCGOV) may be able to create a sustained power advantage (obsoles- (AIGOV) (Weiniger & Villaggi, 2014). cing bargaining), especially in industries with high sunk costs (SCMNE) A particularly effective access strategy in the case was the control of and limited alternative investment opportunities (AIMNE) (e.g., ex- distribution channels by MNEs, restricting the host government’s access tractives). Our holistic framework can also be extended to firm–firm contexts. For example, bargaining with a foreign distributor is equally 8 Similarly, the U.S. government introduced several retaliatory pieces of legislation determined by mechanisms of sunk costs and access to alternative in- against Canadian companies and an official condemnation based on the GATT agreement vestments. However, in a firm–firm bargain, neither party has sovereign following the 1980 National Energy Program introduction in Canada (Jenkins, 1986). status and both must operate within the regulatory and institutional 22 J. Müllner, J. Puck Journal of World Business 53 (2018) 15–26 Table 3 obsolescing bargaining is not a function of bilateral dominance but a Summary of MNE micro-strategies. multi-party process that needs to be analyzed within its socio-economic context (Stevens et al., 2016). Any bargaining framework needs to ac- Micro-strategies reducing sunk costs for the MNE SCMNE Financial engineering High debt ratio (project finance) count for the possibility of secondary power relations with local part- Non-recourse provisions (project finance) ners, legal institutions, suppliers, customers, banks, financial markets, Bond market exposure (ownership and, of course, the MNE’s home government influencing the bargaining covenants) outcomes over time (as described by Gargiulo, 1993), which can in- Export guarantees (EX-IM, etc.) Legal recovery Bilateral trade agreements (Netherlands) troduce a state of mutual dependence as described by Emerson (1962). strategies International arbitration (ICSID, ICC) In this particular case, financial market actors, for example, have been Extraterritorial collateral (PDVSA assets, shown to exert considerable influence on the host government’s ability Escrow accounts) to bargain excessively. Such macro-economic financial dependencies, Home-country CPA Bilateral diplomacy (US DOE, US Foreign we believe, provide a fruitful and novel avenue for future IB research. Affairs) Operational support from home government In addition, the dynamic nature of the bargaining process in the case (US strategic reserve) supports our second theoretical claim that MNE–state bargaining is not Financial compensation from home government static, as assumed by traditional frameworks. Rather, we observe per- (bailouts) iods of obsolescing bargaining, but also episodes in which bargaining Public initiatives & support (boycotts) power is reversed. In particular, we find that regime changes strongly Micro-strategies improving MNE’s access to alternative investments alter the parameters and actors after an investment is made. This has AIMNE Technological Extraction technologies (vertical wells) important theoretical and managerial implications. Feeding the inter- advantage Refining technologies (oil sands, heavy crude oil refining) ests of an incumbent host government and capitalizing on weak in- Exploration technologies stitutions inevitably carries with it a mortality cost, fueling opposition General CPA & CSR Political support (Energy and sunk costs for the MNE (SCMNE) when the regime changes. In si- Diplomacy & Security Act) tuations of government change, relational contacts and political capital Reputation for CSR can become liabilities for the investing MNE (Fernández-Méndez, Micro-strategies increasing sunk costs for host government Garcia-Canal, & Guillen, 2015). Rather, MNEs investing in countries SCGOV Local partners Joint-venture partners (PDVSA strikes) Local personnel (predatory hiring) with weak institutions and strong partisan tendencies are well advised Local suppliers, banks to depart from neoclassical shareholder value maximization (Friedman, Local CSR Social programs (misiones bolivarianas) 2007), towards more stakeholder-oriented approaches to preempt such Opposition groups (gente de petroleo, NGOs) sunk costs (Wang, Tong, Takeuchi, & George, 2016). MNEs should seek Foreign aid (USAid, NED) to fill institutional voids by striving for a broader social license (Su, Technology drain Wells maintenance Refining capabilities Peng, Tan, & Cheung, 2016; Vachani, Doh, & Teegen, 2009). As opposed Legal retaliation International arbitration (ICSID, ICC) to political licenses, which are inextricably linked to an incumbent Asset freezes (PDVSA) government, social licenses are, by definition, based on a broad con- International sanctions (US sanctions) sensus and thus less sensitive to regime change (Fernández-Méndez Micro-strategies reducing host government’s access to alternative et al., 2015; Porter & Kramer, 2006). Therein, our study also contributes investors/buyers to the broader discussion on whether stakeholder interests are moral AIGOV Coalition strategies Trilateral diplomacy (EU, OPEC, Western imperatives or economic rationale in the CSR literature. Evidence from Hemisphere Cooperation) Global creditor syndicates (bond markets) the case positions CSR as a possible win–win strategy in extractive in- Industry coalitions & cooperation dustries. CSR not only reduces the likelihood of obsolescing bargaining Multinational organizations (World Bank, IMF) for a specific investment (SCMNE) but also creates valuable real options Legal collateral International agreements (BITs Netherland) in terms of accessing alternative investment locations (AIMNE). As such, International arbitration Distribution Refining facilities (Chalmette Refinery) CSR and stakeholder orientation can become a firm-specific advantage dependence Distribution channels (Citigo USA) that transcends single investments and country borders (Freeman, Marketing capabilities 1994, 2010; Freeman, Wicks, & Parmar, 2004). While we do find some evidence for this relationship in our empirical setting, we strongly en- Note: Strategies marked in bold were particularly important in the case or provided courage future research to follow up on this line of reasoning. Doing so win–win situations for MNEs. Abbreviations: SC = sunk cost; AI = access to alternative investments; would provide important implications for MNE–state bargaining stra- GOV = government; MNE = multinational enterprise; PDVSA = Petróleos de Venezuela, tegies and would improve our theoretical understanding of the role of S.A.; EX-IM = Export − Import Bank of the United States; ICSID = International Centre CSR in extractive industries and beyond. for Settlement of Investment Disputes; ICC = International Chamber of Commerce; US In addition to regime changes, we also find evidence for more subtle DOE = US Department of Energy; CPA = corporate political activities; CSR = corporate changes in power balance related to other parameters in our bargaining social responsibility; NGO = non-governmental organization; USAid = United States framework. All four components vary over time and can alter the power Agency for International Development; NED = US National Endowment for Democracy; EU = European Union; OPEC = Organization of the Petroleum Exporting Countries; balance between parties. By capitalizing on a temporary power im- IMF = International Monetary Fund. balance, the bargaining party may compromise its ability to bargain effectively in the future (AIMNE) and trigger retaliatory measures framework present. Hence, while the mechanisms of power dependence (SCMNE) from third parties (e.g., governments, MNEs, financial markets, are similar, the boundary conditions may be different in a firm–firm opposition). Thus, the shadow of the future becomes an important bargaining context and merit future research. Overall, and thanks to its consideration in repetitive bargains (Baker, Gibbons, & Murphy, 2002; general applicability, our bargaining framework provides abundant Gibbons & Henderson, 2012). Diving deeper into the long-term dy- opportunities for empirical analysis of bargaining dynamics across namics of bargaining therefore provides rich opportunities for future different industries, countries, and bargaining contexts. research. We believe it might be specifically valuable to analyze the In a second, empirical contribution, we apply our holistic contingencies of those dynamics. While we found strong dysfunctional MNE–state bargaining framework to the case of the Venezuelan oil tendencies in the case, our empirical setup does not allow us to claim industry. The historical analysis of the case speaks to the importance of the external validity of this finding. We thus strongly encourage future multi-party considerations in foreign investments. In the case, research to analyze long-term dynamics of bargaining in different set- tings. 23 J. Müllner, J. Puck Journal of World Business 53 (2018) 15–26 The notion of instable, multi-party power relations adds an im- References portant element to IB theories of the MNE, which most commonly focus on the advantages of internalization without in-depth consideration of Aid Data Beta (2015). Open data for international development. Washington, DC: the broader socio-political nature of an investment and the sustain- Development Gateway. Alcacer, J., & Ingram, P. (2013). Spanning the institutional abyss: The intergovernmental ability of these advantages. This, of course, is partly due to the specifics network and the governance of foreign direct investment. American Journal of of extractive industries, which involve high sunk costs, location-specific Sociology, 118, 1055–1098. resources, and a highly politicized redistribution of revenues. However, Bae, J., & Gargiulo, M. (2004). Partner substitutability alliance network structure, and firm profitability in the telecomunications industry. Academy of Management Journal, the notion that location advantages within a host country cannot be 47, 843–859. internalized unilaterally without a sustainable socio-political license is Baker, G., Gibbons, R., & Murphy, K. J. (2002). Relational contracts and the theory of the likely to apply to other contexts, too. In the broadest sense, it suggests firm. The Quarterly Journal of Economics, 117, 39–84. Barney, J. B. (1991). Firm resources and sustained competitive advantage. Journal of that classic unilateral rent internalization by MNEs needs to be ba- Management, 17, 99–121. lanced with a more bilateral equilibrium perspective to capture the Beidleman, C. R., Fletcher, D., & Vesbosky, D. (1990). On allocating risk—The essence of sustainability of MNE strategies. Strengthening the link between in- project finance. Sloan Management Review, 31, 47–55. Birkinshaw, J., Brannen, M. Y., & Tung, R. L. (2011). From a distance and generalizable to stable power relations and MNE strategy, both empirically and theo- up close and grounded: Reclaiming a place for qualitative methods in international retically, thus represents an important avenue for future research, as it business research. Journal of International Business Studies, 42, 573–581. would enhance our understanding of the long-term consequences of Blumentritt, T. P., & Nigh, D. (2002). The integration of subsidiary political activities in MNE strategizing. multinational corporations. Journal of International Business Studies, 33, 57–77. Boddewyn, J. J. (1994). International business political behavior—New theoretical di- Finally, we use the holistic MNE–state bargaining framework to rections. Academy of Management Review, 19, 119–143. identify and describe MNEs’ micro-strategies used to tilt a power bal- Boué, J. C. (2013). Conoco-Phillips and Exxon-Mobil vs. Venezuela: Using investment arbi- ance in their favor. Existing MNE–state bargaining conceptualizations tration to rewrite a contract. Investment Treaty News—September 20, 2013. Winnipeg, Canada: Institute for Sustainable Development. have not sufficiently acknowledged MNE micro-strategies to structure Boué, J. C. (2014). Enforcing pacta sunt servanda?: Conoco-Phillips and Exxon-Mobil investments in a way that optimizes power balance between the MNE versus the Bolivarian Republic of Venezuela. Journal of International Dispute and the state along the four parameters of the framework. Using ex- Settlement, 5, 438–474. Brealey, R. A., Cooper, I. A., & Habib, M. A. (1996). Using project finance to fund in- tensively triangulated data, we identify 12 micro-strategies that MNEs frastructure investments. Journal of Applied Corporate Finance, 9, 25–39. used to influence bargaining power in their favor. In particular, we find Cable (2004). 04CARACAS731_a: How real is Chavez’s oil threat? WikiLeaks. evidence that MNEs used ex-ante and ex-post investment recovery Cable (2007a). 07CARACAS83_a: BRV raises stakes in the Faja. WikiLeaks. Cable (2007b). 07CARACAS411_a: Oil nationalization decree raises questions. WikiLeaks. micro-strategies to reduce their sunk costs (SCMNE). These involved fi- Cable (2007c). 07CARACAS1314_a: Chevron, Exxon, and CNPC comment on recent events. nancial engineering, legal recovery, and home-country political activ- WikiLeaks. ities. In addition, MNEs attempted to reduce dependence by improving Cable (2007d). 07CARACAS1449_a: Politics and taxes. WikiLeaks. Cable (2008). 08CARACAS168_a: ExxonMobil moves to freeze PDVSA’s worldwide assets. access to alternative investment destinations (AIMNE), mostly by means WikiLeaks. of technological advantages but also through CSR and political support. Caroll, R. (2010). WikiLeaks cables: Oil giants squeeze Chávez as Venezuela struggles. The More indirectly, MNEs used micro-strategies to increase sunk costs for guardian December 9. London. the host government (SCGOV) by cooperating with local partners, en- Casciaro, T., & Piskorski, M. J. (2005). Power imbalance, mutual dependence: and con- straint absorption: A closer look at resource dependence theory. Administrative Science gaging with local communities (CSR), and maximizing human capital Quarterly, 50, 167–199. losses from expropriation, and by legally retaliating ex-post. Finally, the Chowdhry, B. (1991). What is different about international lending? Review of Financial case provides evidence of financial and political means of reducing the Studies, 4, 121–148. Cook, K. S., Emerson, R. M., Gillmore, M. R., & Yamagishi, T. (1983). The distribution of host government’s access to future investments (AIGOV) by maintaining power in exchange networks: Theory and experimental results. American Journal of and accessing country- or investment-level alliances and investment Sociology, 89, 275–305. agreements, and by ensuring control over downstream distribution Cort, T. (2008). CSR—An environment of accountability? Petroleum review, may 12. Irving TX: Energy Institute. channels. The outcome of our case illustrates that such micro-strategies Cuervo-Cazurra, A., Andersson, U., Brannen, M. Y., Nielsen, B. B., & Rebecca Reuber, A. can be used to reduce the MNE’s expected loss and the probability of (2016). From the Editors: Can I trust your findings?: Ruling out alternative ex- expropriation but they cannot fully prevent MNE bargaining, especially planations in international business research. Journal of International Business Studies, 47, 881–897. in extractive industries. Some of these micro-strategies have not been Dicken, P. (1994). The Roepke lecture in economic geography: Global-local tensions. addressed in the IB literature and provide promising avenues for future Economic Geography, 70, 101. research on an investment level. In addition, they provide valuable Doh, J. P., & Ramamurti, R. (2003). Reassessing risk in developing country infrastructure. Long Range Planning, 36, 337–353. managerial implications for MNEs in extractive industries. Doh, J. P., Lawton, T. C., & Rajwani, T. (2012). Advancing nonmarket strategy research: There are several limitations to our case study and to the identifi- Institutional perspectives in a changing world. Academy of Management Perspectives, cation of micro-strategies. The most obvious is related to the sampling 26, 22–39. strategy. While the triangulated data from official and leaked sources Doh, J. P., McGuire, S., & Ozaki, T. (2015). Global governance and international non- market strategies: Introduction to the special issue. Journal of World Business, 50, provide us with strong internal validity, the use of a single case ne- 256–261. cessarily reduces external generalizability. Hence, unlike our holistic Du, S., & Vieira, E. T., Jr. (2012). Striving for legitimacy through corporate social re- MNE–state bargaining framework, the micro-strategies identified in this sponsibility: Insights from oil companies. Journal of Business Ethics, 110, 413–427. Dye, J. (2016). Crunch time for Venezuela’s PDVSA. Financial times [October 13. New case may not be generalizable to other industries and countries. York]. However, financial engineering, socially responsible investment, legal Economist T. (2017, June 1). Socialism or debt − Goldman Sachs is criticised for buying protection, and CPA are clearly not limited to this context. As a result, Venezuelan bonds. New York. Eisenhardt, K. M., & Graebner, M. E. (2007). Theory building from cases: Opportunities we strongly encourage research on micro-strategies in other interna- and challenges. Academy of Management Journal, 50, 25–32. tional investment contexts. However, we emphasize that other MNEs Eisenhardt, K. M. (1989). Building theories from case study research. Academy of can benefit from the lessons of extractive industries, which have been at Management Review, 14, 532–550. Ellsworth, B., & Parraga, M. (2012). Exxon wins less than expected from Venezuela dispute. the short end of many MNE–state bargaining episodes. As such, our Reuters newswire January 1. Caracas. micro-strategies offer a managerial blueprint for all MNEs investing in Emerson, R. M. (1962). Power-dependence relations. American Sociological Review, 27, high-risk contexts.