Anti-corruption

Representing at least five percent of world GDP, corruption is a great challenge in general, and especially associated with sustainable business, both nationally and internationally. In this chapter, I reflect on how to address corruption and anti-corruption practices. Typical forms of corruption include bribery, facilitation payments, gifts, hospitality and expenses, political contributions, charitable contributions, sponsorships, voluntary community contributions, trading in influence, and conflict of interest and impartiality. Corruption leads to weak institutions and injustice, less respect for rights, denial of basic services, and several of the world’s environmental damage and tragedies are associated with corruption. Studies show that almost half of the companies have experienced fraud over the past two years. This chapter addresses the corruption challenges, how they are addressed by corporations and key international laws as well as challenges associated with norms and behaviors. How to detect and avoid corruption receives key attention. As anti-corruption is a huge topic and challenge, and only limited space is available in the book, the topic is presented from a more general perspective. Even though anti-corruption is often left out when talking about sustainability, it cannot be excluded from a book on sustainability and responsibility.


Abstract
Scholars tend to agree and evidence has shown that domestic businesses adapt to the local type of corruption, but little is known whether large multinational corporations also adapt to the local forms of corruption. Institutionalist theories of corruption and of international political economy would suggest that this would be the case, but the hypothesis has not, to our knowledge, been systematically tested. This paper, drawing on investigative materials about the activities of one such multinational, the German corporation Siemens AG, examines how it used corruption and bribery to advance its business around the world. We extrapolate from the logic of four "syndromes of corruption", as Michael Johnston terms them, to develop specific hypotheses about the kind of behavior multinational corporations would be expected to exhibit when doing business in each of the four kinds of syndromes. We examine and compare Siemens' activities in the United States, Italy, Russia and China. We find that Siemens did adapt to the local corruption form (or "syndrome") and used, among others, different types of intermediaries to approach the local elites. The evidence from these case studies supports the institutionalist argument that multinationals distinguish between corrupt environments and further supports the argument that there exist different types, or syndromes, of corruption.  Does the local (host country) corruption environment make any difference to an MNC's business tactics, or does it engage in corruption the same way in every country in which it operates? 1

Introduction
Anecdotal evidence suggests that MNCs vary their approaches across countries, however, to our knowledge, no systematic analysis has yet been done. This paper takes a first step in doing so, by assessing whether an MNC uses different corruption mechanisms depending on the corruption pattern in the host country.
A large and growing body of literature has investigated the role the domestic institutions play in determining a firm's business strategies. Numerous scholars have argued that a country's institutional framework shapes a firm's corporate structures and thus their activities through the incentives particular laws and regulations provide (Hall and Soskice 2001;Gourevitch and Shinn 2005). In their seminal study, Hall and Soskice argued that the way in which firms coordinate their behavior is strongly influenced by the domestic institutional configuration of markets and firms.
However, much of the theory is based on evidence from firms operating within their own advanced economies, neglecting to investigate the role the institutions of a host country play when an MNC invests in foreign markets. We extend the logic of the institutionalist argument to ask whether the local corruption pattern impacts a firm's strategic use of corruption.
When the political economy literature on MNCs has looked at corruption, it has generally expected that MNCs would compel corrupt foreign governments to "clean house". NAFTA was celebrated partly for its alleged potential to bring clean business practices to Mexico. One need look no further than Wal-mart in Mexico to see that these expectations are not born out in practice Barstow and von Bertrab 2012;Clifford and Barstow 2012;Stewart 2012;Moreton 2009;cf Warner 2007).
Only a few business scholars have looked at how different corruption patterns in the host countries shape an MNC's corrupt behavior (i.e., Bray, 2005;Uhlenbruck, Rodriguez, Doh, and Eden 2006;Rodriguez, Uhlenbruck, and Eden 2005). For instance, some have scaled corruption on two dimensions -arbitrary, referring to the randomness of when a firm can be confronted with bribery demands, and regular, referring to level of routine and predictability of bribery demands (Uhlenbruck et al. 2006;Rodriguez et al. 2005). Even though this approach captures better the variations in local corruption patterns, it lacks the conceptual and theoretical background to adequately explain the varying corruption patterns across countries and how they might affect a firm's use of corruption.
This may explain why little attention has been paid to whether an MNC's corruption tactics differ across foreign markets.
A common view is that corruption is a uniform phenomenon, the magnitude of which can be ranked on a single low-high scale (see for instance, the annual Corruption Perceptions Index by Transparency International, or the World Bank's Worldwide Governance Indicators (WGI) on anticorruption efforts in a country). This view has been reinforced by a number of studies that have used these indices (e.g., Mauro 1995;Treisman 2000;Montinola and Jackman 2002). When scholars have differentiated among types of corruption, they have largely categorized corruption according to he level of public office at which corruption occurs -such as "petty" and "grand" corruption (Moody-Stuart, 1997) or "legislative," "bureaucratic," and "political" corruption (Rose-Ackerman, 1999). However, such clear-cut categorizations of public offices can rarely be made, since corruption at the lower levels often finances the corrupt dealings at the higher levels and the same corruption case may implicate legislatures, bureaucracies and politicians and parties at the same time (Chayes 2015;Warner 2007).
In response to limitations in the literature, Michael Johnston (2005) proposes categorizing countries by their "syndrome" of corruption -the political and economic institutional arrangements that allow for and that produce corruption, and that characterize how power and wealth are attained. We extrapolate from Johnston's (2005) typology of the four ideal types of corruption structures patterns to argue that MNCs will employ different mechanisms of corruption depending on the host country's syndrome. As a first step in assessing this argument, we focus on the German MNC Siemens AG (hereafter Siemens). Siemens is a globally operating MNC that is active in a variety of industries. With an annual gross profit of around 20 billion euro, it appears that the company would be able to dictate the terms of business in its transactions -including walking away from corrupt deals. Instead, investigations have shown that it has engaged in bribery and other forms of corruption to land contracts and advance its business interests in countries around the globe. Our analysis draws on publicly available material, in particular the reports by investigative journalists and law enforcement authorities from around the world.
In the following section, we will provide a brief review of what is known about corruption and will develop hypotheses about how the foreign corrupt environment might affect an MNC's actions inside the country. The second section explains the selection of Siemens as the focal MNC, and countries and corruption-cases within them that have been chosen for the study. The third section analyses the cases. The fourth section discusses the results and concludes.

Theory
Scholarly studies of management and of political economy have largely ignored the question of MNC's corruption strategies. Both literatures, if they focus on corruption, view it as a homogeneous phenomenon that affects political-economic institutions. In contrast, much of the corruption literature has focused on the causes, types and forms of corruption, in particular examining historical, economic and socio-political determinants within a country or set of countries (i.e., Golden and Chang 2001;Mauro 1995;Treisman 2000;Montinola and Jackman 2002;Karklins 2005;Shleifer and Vishney 1993;Sun 2004). By looking at these broad sets of literature, we can tease out the main factors that might affect an MNC's use of corruption: the level of competition in the market; the corporate culture; the domestic and foreign institutional framework, including an MNC's assessment of the risks and costs of getting caught; and the local corruption patterns existing in the foreign country. Each of these explanations will be discussed in more detail below.

Explanation: Competition reduces overall corruption in a country
Several studies have looked at how increased competition affects corruption levels, either among bureaucrats (Shleifer and Vishy 1993) or within markets (e.g., Mauro 1995;Ades and Di Tella 1999), finding that more competition reduces opportunities for corruption (Rose-Ackerman 1978; Ades and Di Tella, 1997;Bliss and Di Tella 1997;Gerring and Thacker 2005). The presumed mechanism is that the increase in competition that MNCs bring lead to a reduction in corruption in the host country. More competition breaks up collusive arrangements, making corrupt deals hard to conclude, and more interstate economic competition means domestic firms and politicians can no longer afford the "tax" that corruption adds to economic transactions.
If this argument obtains, an MNC would refuse to engage in corruption, in particular if it were large enough to dictate the conditions of the transactions. However, recent experiences with NAFTA and also the 2004/2007 accession of twelve new member states in the European Union suggest that corruption patterns may change but corruption does not necessarily decrease (Warner 2015). Some management scholars also note the negative effects of competition, finding that if firms face greater financial and competitive pressure, they are more likely to resort to bribery (Martin, Cullen, Johnson, and Parboteeah 2007). A reason for this could be that an MNC will consider the behavior of their competitors in the new market. If they fear that their competitors are paying bribes, the likelihood that a firm is willing to engage in bribery increases (Bray, 2005). In addition, facing competitive pressures, an MNC may decide that corruption is the most expedient business strategy in a tight market (Warner 2007).

Explanation: Corporate Culture affects whether MNCs employ corruption.
The management literature proposes that the culture within the firm, and the home country culture, determines the firm's willingness to use corruption. First, the corporate culture within the firm impacts its perception and willingness to use corruption to further its business abroad (Bray 2005;Uhlenbruck et al. 2006

3.3.1) Law Enforcement
The literature on corruption has argued since several decades that the risks and costs of getting caught impact corrupt behavior strongly. The 'corruptor' will assess the likelihood of getting caught by law enforcement authorities and the costs of the possible punishment (Klitgaard, 1988;Rose-Ackerman, 1999). As to the likelihood of getting caught, in most Western countries bribing foreign public officials for business purposes has been criminalized only recently. In particular, 'export corruption', that is, a firm providing some form of remuneration to foreign officials outside of normal business activities, has rarely been investigated. Only in the late 1990s did the OECD countries ratify a convention against corruption. On the side of the host countries, investigating international corruption is difficult as it demands large resources in both time and personnel, which even richer countries cannot always afford.
The enactment and enforcement of anti-corruption laws for domestic and international corruption has varied as well. For instance, while China is now on a major anti-corruption drive, in the previous decades since economic reforms began, paying bribes was a regular part of doing business (Sun 2004

3.3.2) General Institutional Framework
A large and growing body of literature has investigated the role the domestic institutions play in determining a firm's business strategies. Numerous scholars have argued that the behavior of firms is strongly affected by the institutional structures they are in (Hall and Soskice, 2001). For instance, in their seminal study, Hall and Soskice (2001) illustrated that how firms coordinate their behavior is strongly influenced by the domestic institutional support they receive. In addition, some proponents of the institutional approach suggest that the institutional framework shapes a firm's corporate structures and thus their activities through the incentives particular laws and regulations provide (Gourevitch and Shinn 2005).
However, because they focus mostly on a firm's behavior at home, these arguments do not address to what extent the host's country institutional framework matters for a firm's behavior in that country.
The institutionalist perspective, however, might hold that MNCs operating in a variety of countries may be constrained in how much they can adapt their internal organizational structure to a new environment and thus may be ill-equipped to do business in a different context. However, while firms are constrained in their formal structures, their informal structures can be more flexible by decoupling formal and informal practices (Meyer and Rowan 1977;Helmke and Levitsky 2004).
Extending this view to our topic, it would suggest that MNCs would employ different corruption tactics abroad, once their headquarters determines the overall business strategy.
Some accounts have indicated one way that this decoupling occurs in practice -through the use of intermediaries, which puts distance between the act of corruption and the official business of the MNC. For this, intermediaries are commonly employed as consultants of some sort -as commercial agents, actual business consultants, joint venture partners or as managers of the local subsidiaries (Warner 2007: 50). Their common skill is that "[t]hey can get things done" (Bray 2005: 155). A second advantage is that intermediaries can act as "door-openers" and so function as a "buffer" against bribery demands and offer the claim of plausible deniability (Bray 2005; Moody 97).
Naturally, operating through an intermediary carries risks as well. Such intermediaries need to be trustworthy and the wrong intermediary raises the transactions costs if the deal fails. In such cases, the company risks being blackmailed by the intermediary themselves . In addition, the company runs the risk of their intermediary losing their access to the local elites. While employing intermediaries is not a stable long-term investment, it may be a tactic that enables MNCs to successfully negotiate the host country's institutional context.

Explanation: Local corruption pattern affects (local) business'
behavior Michael Johnston argues that the combination of political and economic institutions that affect how people attain power and wealth produce different patterns of corruption across countries. A country's governing and economic institutions are assessed for their capacity to sustain markets, and enforce the rule of law, and for how they allow and foster citizens' participation in economic and political life. Extrapolating from Johnston's model we would expect that firms would have to engage in different corruption practices in, say, China than in France to be successful. While that observation may seem trivial once stated, the corruption literature has tended to treat all forms of corruption equally: corruption is bribery; what varies is how much and how often a firm bribes. There is an implicit assumption in the literature that when a country that ranks as "highly corrupt" on international corruption perception indices, a firm will have to engage in more bribery than it has to when it operates in a country that is assessed as having little corruption. However, Johnston's key insight is that the corruption pattern, the behavior that constitute corruption, varies across countries.
As noted, he suggests there are four main patterns worldwide (see table 1).

Oligarchs and Clans
Transitional regimes

Official Moguls Undemocratic New markets Weak Weak
Johnston's argument is centered on domestic political and economic interests. He does not consider whether MNCs would also be affected by these patterns when operating abroad. The logic of his argument, however, implies that they would be.
Overall, these scholarly works suggest that an MNC's use of corruption and the variation in these approaches are affected by several factors: the corporate culture and domestic institutional framework, the MNC's assessment of the risks and costs of getting caught, and the local corruption patterns that exist in the foreign country. The puzzle remains, however, of whether firms bring to countries their own corruption strategies or adapt to country-specific contexts. While political economists stress that the institutional context in which MNCs operate is significant for explaining MNC business strategies, they have not yet applied that logic to what we are calling MNC "corruption tactics." We suggest it could be a fruitful line of analysis.
The argument we propose combines elements of comparative political economy's emphasize on a home country's institutional framework and Johnston's focus on local corruption patterns, both influencing a business' behavior. We suggest that while an MNC's headquarters may decide on the strategy, in particular whether the use of corruption is acceptable, the tactics are influenced by local corruption structures. This means that corruption as a business strategy is set by the senior management but is then adapted to the local corruption environment. In the case of Siemens and its use of corruption as a strategy, this is exemplified by the existence of 'cash-desks' situated in the Headquarter as well as its system 'off-the-book' accounts that were used for larger bribe payments Relevant for this study is how foreign firms can get access to host country politicians. Apart from campaign contributions, an increasing number of firms also resort to having lawyers and lobbying 5 Here it is important to note that it is often hard to distinguish between appropriate service to constituencies and political corruption (Harris 2003;Thompson 1993). This is also visible in the ranking of the corruption indices that focus on bribery (Johnston 2005: 60).
firms on their retainers that act as intermediaries on their behalf. This allows the MNC to stay within the legal limits (or at least within the gray areas of law) and still influence the political decision making process. We therefore propose: Hypothesis 1: In countries characterized by influence market corruption, the MNC will resort to employing lobbying firms and contribute to campaigns to further their business interests.
The second syndrome Johnston identifies is Elite Cartel corruption, found in such diverse countries as Argentina, Botswana, Italy, and South Korea. The major function of this type "is to protect, as well as to enrich, networks of higher-level elites" (Johnston, 2005: 89). These networks of elites consist typically of high-level politicians, business people, higher level bureaucrats as well as military and ethnic group leaders.
Elite Cartel corruption is characterized by its relative predictability and organization. Large-scale corrupt activities are commonly restricted to the elites and focus on sustaining cohesion among the group. Anti-corruption efforts, when they exist, are relegated to fighting petty corruption at the lower levels of bureaucracy and rarely reach the elites. This is also reflected in the weak formal institutions that often face pressure from undue influence from the outside. Law enforcements and judiciaries might be hesitant to investigate any criminal cases that involve an important member of the domestic elites, or banks will provide cheap loans to firms connected to elites. In general, the domestic elite networks act as a substitute for the formal institutions, but with the aim of protecting their network's interests.
This environment is challenging for foreign firms because they constitute a potential threat to the interests of the domestic elites and the firms they favor. An MNC's business activities also might challenge the business interests of some members of the elites. Moreover, the weak formal institutions mean that in case of disputes the MNC will have to rely on other means to protect its interests. Finally, because of the centralized form of elite cartel corruption, gaining access to the elites is the main hurdle for investors. For any MNC that enters such markets, this means that it will have to establish some form of connection with the elites.
In such instances, a firm would have to provide some form of compensation -most commonly through bribes, but it could also be through employing particular persons as a favor -when they enter an Elite Cartel market. For an MNC, it would be important to find business consultants who have personal connections to the elites to help distribute covertly the bribe payments and establish stronger ties with the domestic elites. The personal ties matter in so far as bribe payments cannot be made openly, and in so far as trust is a necessary prerequisite to ensure that all parties to the transaction remain quiet (Lambsdorff and Teksoz, 2005).
We therefore expect that: in Oligarch and Clan countries, corruption is rarely centralized and does not act as a substitute for weak formal institutions. Instead corruption increases the instability in the country. Corruption, together with targeted violence, acts as a means to increase profits and generally do business in an uncertain environment. In these countries: "A relatively small number of individuals use wealth, political power, and often violence to contend over major stakes, and to reward, their followers, in a setting where institutional checks and legal guarantees may mean little. Their power is neither clearly public nor private -often, it is both -but it is definitely personal." (Johnston, 2005: 120) Such clans can stretch "across business, state agencies and political parties, law enforcement, the communications media, and organized crime" (Johnston, 2005: 120) that at varying times cooperate, collude or compete among each other. A particularly well-known example would be Russia, but both Mexico and the Philippines also exhibit the syndrome.
In addition, the business literature suggests that in such an inherently uncertain environment, an MNCs would prefer short term gains over long term investments (Clague, Keefer, Knack, and Olson, 1996), to minimize its risk exposure. For instance, a firm might only bid for particular projects instead of building wholly owned subsidiaries (Uhlenbruck et al, 2006). Also, by contracting part of a project out to subcontractors, a foreign firm can pay their 'dues' to particular oligarchs and their clans. We therefore posit that: have in common is that elites plunder the state and society, and while economic opportunities exist, few individuals can make use of them without the acquiescence of the local official authority who acts as a patron within their district. The patronage networks span across the entire level of state bureaucracy. These well-placed officials can act almost without impunity in their territory as long as they do not threaten the interests of their superiors.
However, the lack of restrictions on the activities of the local mogul also mean that a "mogul's personal agendas matter a great deal." (Johnston, 2005: 183). Hence, because the success of economic activities strongly depends on the acquiescence of the ruling "mogul," there is widespread variation in how corruption operates on the local levels -some top figures view corruption as a challenge to their rule while others are more lenient, some oppose corrupt transactions in which they are not involved, while others do not care. What is common, however, is that the mogul has the ability to create major obstacles for any firm operating on its territory.
Charges of tax evasion or money laundering are just two of the tools at a mogul's disposal that can harm businesses.
As the business environment and operation of corruption depends mostly on the official mogul, the challenges foreign investors face vary greatly. Therefore any expectations on the obstacles a MNC faces and its response can be stated only in very broad terms. Johnston's framework does not indicate how a MNC would act in such environments. It seems plausible however, that a MNC would have to provide some kind of compensation in the form of private benefits for the local mogul in return for being able to operate in their territory. Gaining access to an official mogul and later on providing the necessary favors would again require an intermediary. We posit that: Hypothesis 4: In countries identified with Official Mogul corruption, MNCs will rely on intermediaries with personal ties to a high-ranking public official to further their business interests and their bribery activities will be extensive.
In conclusion, the use of intermediaries can be expected in all four cases, the differences is in whom an MNC will approach to act as intermediary -lawyers and lobbyists in influence market countries, business consultants with established ties in elite corruption countries, and intermediaries that are part of the oligarch and clans or the officials moguls network in the last two cases.
Expectations for the last two cases are similar as Johnston's theory offers little detail on the particular characteristics of an intermediary.

Research Design
We conduct a qualitative case study of one major MNC as a means of exploring the plausibility of the hypotheses. While there are thousands of MNCs one could choose from, we propose a wellknown, major corporation that is so large it presumably could dictate terms to the host country. If we find, instead, not just that the MNC is engaging in corruption, but that its tactics change according to the host country and correspond to the host country's domestic corruption syndrome, we increase our confidence in our hypotheses. There are of course a number of large, well-known MNCs, some more specialized in certain sectors than others. We chose Siemens AG for a variety of reasons.
First, Siemens is a large multinational corporation that operates across a variety of sectors -among others, industry, energy, healthcare, and infrastructure -in almost every country in the world, meaning it operates in countries that fall into each of Johnston's four categories. The company was founded in 1847 in Germany by Werner von Siemens -an inventor and businessman -and soon went international. Its main specialization at that time were technically, highly complex and risky projects, including building a telegraph network from London to Calcutta. While Siemens has expanded into other industries since then, its core strength remained that of doing complex and risky projects all around the world.
Second, given its size, it is precisely the kind of MNC that the literature expects can dictate its own terms to host countries, and thus the kind that would be expected to reject, rather than adapt to, To test our hypotheses, we analyze several cases in which Siemens has been charged with corruption. Each case has been chosen based on whether Siemens has been investigated on charges of corruption, whether Johnston identified the corruption syndrome within the country, and whether he also analyzes the country in more depth. We included the last condition because starting with a set of Johnston's paradigmatic syndrome countries facilitates analytical clarity. For a list of all the cases in which Siemens has been implicated as well as Johnston's categorization of the specific country's corruption syndrome, see the appendix. We assess Siemens cases in the US, Italy, Russia and China, representing, respectively, all four corruption syndromes.
One limitation to our study is that we do not know what the universe of corruption cases by Siemens (or any other MNC) is: corruption is notoriously difficult to detect (Jain 2001;Treisman 2007;Sampford, Shacklock, Connors, and Galtung 2006). While often the only corruption cases that come to light are those that can be said to have at some point failed (if they had been successful, they would not have been discovered), and thus the subject of study might not be the typical, and generalizable, cases of corruption but merely the type that is most easily discovered. However, in the case of Siemens, the large-scale, worldwide investigations meant that several cases have been found that otherwise might have remained hidden. We are able to see some corruption cases and patterns that were discovered not because they "failed" but because investigations were farreaching and persistent.
A second limitation is that there is variation in the economic sectors in which Siemens has been implicated for corruption. To our knowledge, there is not (yet) enough information available about corruption cases to make any comparison between countries in different corruption syndromes by the same firm in the same specific economic sector.

Analysis
This section examines whether a multinational corporation adapts its corruption strategy to the corruption syndrome of the host country. We describe at least one corruption case for each corruption syndrome Johnston identified in which Siemens has been found guilty of corruption. We note that this is just one interpretation, as we have neither space nor information on more than one case per syndrome. Hence, we will discuss in the conclusion alternative interpretations as well as additional limitations of the study.

Influence Market Corruption in the United States
The first hypothesis proposes that in countries characterized by Influence Market corruption syndromes, MNCs will resort to employing lobbying firms and contribute to campaigns to further their business interests. Siemens did all of the above and indications exist that they even resorted to 'influence-peddling'. While a minor case of outright bribery did occur, overall Siemens relied on lobbying activities to advance their business interests within the US.

Figure 1: Siemens' Lobbying Spending in the US between 1998 and 2007.
Lobbying is an expensive endeavor and not all companies in the health care industry do it (Israel, 2011). Siemens employs several lobbyist firms that are owned by former government officials and so makes use of the `revolving door' phenomenon. The `revolving door' phenomenon refers to former government officials that get employed in the private sector and lobbying firms in the hopes to gain access to the former official's colleagues at federal agencies and committees. One of these lobbyist firms on Siemens Corp's retainer is Riccetti Inc. that is owned by Mr. Ricchetti, who was Deputy Chief of Staff under President Clinton (Ismail, 2003). Another way of influencing policy that Siemens used over the years was to fund PACs. In 2000, for instance, Siemens Corp and its subsidiary contributed 14,500 USD to political campaigns. In the 2002 election cycle that number already increased to over 118,000 USD (Israel, 2011).
Last, Siemens is a member in the for-profit think tank, Center for Health Transformation. From 2003 to 2011 Newt Gingrich, the former Speaker of the US House of Representatives, has been linked to the Center, giving talks and presentations on the behalf of its clients (Schulte and Eaton, 2012).
Gingrich claims that he did not act as a lobbyist; however, annual membership in the Center gives clients the opportunity to have Gingrich present white papers and other policy briefs at government panels or other media outlets. Gingrich's claim of not having been a lobbyist is within the narrow definition of lobbyists that the federal law stipulates. This makes it easy for former politicians to influence policy issues without having to register as lobbyists. The law only stipulates that someone has to register as a lobbyist if they are retained by a client in exchange for compensation, have more than one lobbying contract and within a three-month time period has worked at least twenty percent of their time lobbying on their client's behalf. And while the membership of the Center is not public, a partial list of their 2009 members became public -Siemens is among them. Some academics and politicians view this type of behavior of companies as 'influence-peddling' -using ones reputation and influence, for a price, to assist someone who is seeking preferential policies (Schulte and Eaton, 2012). Siemens' activities seem to support Hypothesis 1.

Elite Cartel corruption in Italy
The second hypothesis states that MNCs will employ business consultants with previous ties to the elites to enter into business in countries categorized by Elite Cartel corruption. The main difference to Influence Market corruption is that formal institutions fall under the influence of domestic elites.
Their main motive is to keep competition limited and so ensure that their political and business interests are not threatened by outsiders. As mentioned, Johnston classifies Italy as an Elite Cartel syndrome country, and Siemens has been the subject of anti-corruption investigations there.
In 2003, Italian law enforcement unraveled an extensive network of hidden bank accounts that the power company ENEL, partially owned by the Italian government, held to provide off-the-books payments. While the investigation focused at the beginning only on suspicious payments to officials in Oman, Qatar and the United Arab Emirates, it soon uncovered that ENEL used a wide network of hidden accounts to distribute bribes to decision-makers around the world to win contracts (SEC vs Siemens, 2008: 8-9). In addition, the investigations implicated several European corporations such as the French firm Alstom Power, the Italian firm Ansaldo Energia, Siemens and the Swedish-Swiss firm ABB among others (Ansa 2006).
As the second hypothesis expects, Siemens employed intermediaries, in this case, Horst Vigner and Wolfgang Becker. Both were negotiating and distributing the bribe payments. Horst Vigener was a former Siemens manager whom Siemens contracted as a consultant after his retirement because of his long standing relationships in Greece and Italy. Wolfgang Becker was at the time the director of sales at the subsidiary Siemens PG (Leyendecker, 2009;Ansa, 2004). 8 According to them, the payments were necessary to ensure that Siemens would not only win the contract but also entrance into the Italian market for power-generation equipment (Dougherty 2007).
Vigener and Becker denied having offered bribes and even claimed of not having known that the accounts they were using to transfer the money were illegal. While it might not be clear who started the bribe payment, it is apparent that both sides of the deal were familiar with the process and had established covert financial networks through with the money could be transferred -from accounts Siemens also used another approach to gain access to local knowledge and the market when it formed a consortium with the subsidiaries of the Italian conglomerate Finmeccanica. Finmeccanica is similar to Siemens in that it operates across a wide span of sectors, among them Defense Systems, Energy and Transportation. By aligning itself with two of Finmeccanica's subsidiaries, Siemens ensured that it would not threaten Italy's leading energy corporation. 9 Furthermore, the contract that was the subject of the bribes gave Siemens a monopoly on the link with Finmeccanica for twenty years, much to the concern of several of the latter's executives (Ferrarella 2003). ENEL had already been the source of enormous domestic kickbacks to the Italian political class, as the Tangentopoli investigations discovered. Hence, our second hypothesis finds support as Siemens not only employed intermediaries that had connection with Siemens and the foreign market but also formed a consortium with two firms from the host country. This would ensure that it wouldn't threaten the business interests of any local business elites while also providing political elites with the electoral benefit of market stability.

Oligarchs and Clans in Russia
The third hypothesis proposes that in countries characterized by Oligarch and Clan corruption, Siemens would use intermediaries that are connected to the relevant oligarchs or clans to secure their business and would be very cautious about their long-term investments in a country.
US investigators revealed numerous separate corruption cases in Siemens' dealings in Russia. For instance, in one case Siemens deployed a bribery scheme to win the contract for a project funded by the World Bank for the design and installation of a traffic control system with a value of around 27 million USD. For this, OOO Siemens, the Russian subsidiary, ensured that one of their business consultants would be a technical consultant on the committee that designed and wrote the tender.
In addition, Siemens hired two of its competitors as subcontractors, forming a sham consortium (SEC vs Siemens, 2008: 25-26). Although this case illustrates our third hypothesis, there is not enough publicly available information to examine the details on how Siemens made use of its intermediaries. However, a more substantive case has come to light that allows us to analyze Siemens' use of intermediaries in Russia in more detail. 10 In recent years, Siemens' Russian medical branch, Siemens MED, has been at the center of several corruption investigations -both foreign and domestic. In general, senior officials at Siemens estimated that almost 80% of all Siemens MED business transactions in Russia involved bribes in the first decade of 2000. The smaller bribery cases appear to be only of limited scope. For instance, Siemens paid around 288,000 USD, including bribes, to ensure the sale of a 2.5 million USD computer tomography system to a public hospital in Yekaterinburg (SEC vs Siemens, 2008: 27).
Russian authorities arrested the former deputy head of Yekaterinburg city health department, Alexander Shastin, on suspicion of fraud in November 2007. While a court later found Mr. Shastin not guilty of having taken bribes from Siemens (but did from other businesses), the charges did include the accusation that Mr. Shastin formulated a tender in such a way that Siemens' equipment was the only one fitting the criteria. Together with Shastin, the authorities also arrested and charged the regional manager of Siemens medical equipment -Lev Dubnov -with corruption. He was later convicted (Dubicheva, 2010).
Already in 2005, a former senior manager of Siemens MED came forward with claims that Siemens required him to bribe Russian officials. Sam Tsekhman, a Canadian citizen originally from Russia, led the sales department for medical equipment in Russia from 1996 to 1999. In this capacity he often would receive large sums of bagged cash by another Siemens manager to distribute in Russia. In one particular instance, the renowned medical Burdenko Institute of Neurosurgery in Moscow ordered an entire radiology unit and other equipment -in total around 28.2 million EUR worth -from Siemens in November 1999. To get the permission of the Russian ministry as well as 10 The US Security and Exchange Commission (SEC) investigated Siemens also in connection with the supply of gas turbines for the Yamal-Europe natural gas pipeline. Swiss prosecutors investigated the Swedish Siemens subsidiary, Siemens Industrial Turbomachinery (SIT), on corruption charges. It turned out that SIT paid bribes to high-ranking officials of a large state-owned Russian gas-producing firm. And while the Swiss authorities did not release the name of the involved Russian company, the only state-owned gas producer in Russia is Gazprom. It is also involved in the construction of the Yamal-Europe gas pipeline (Bankier.pl, 2013). However, while the entire bribery scheme already started in the early 1990s, Siemens only bought the gas turbine producing subsidiary from the French company Alstom in 2003. In recent years, Alstom also faces charges of corruption in several countries. It appears that Siemens not only inherited the contracts for the gas turbine supplies but also the existing bribery scheme (Shleinov, 2013). However, Siemens acquired SIT in 2003 and seems to have incorporated SIT's already established corruption patterns into its overall corruption strategy.
win the contract, Siemens included in its tender bribe payments of around 7 million EUR. Mr.
Tsekhman states that two Siemens managers warned him to not distribute the bribes himself, but use a middleman for it, a common practice for Siemens. To distribute the bribe payments to Russian politicians and officials, Siemens employed Lazar Papernick, a Russian businessman situated in Boston. Both men filed affidavits in support of their stories (Tillack, 2005). It appears that in neither case the intermediaries were connected to any oligarchs -however, this might be explained by the relative minor sums involved.
In 2010 in Moscow in the mid-90s, the latter two took over the company in full and continued to supply medical equipment to Russian hospitals (Grey, Bush and Anin, 2014;Bryant and Sylvers, 2012 alone (Grey, Bush, and Anin, 2014;Belton, 2011).
We do not suggest that Siemens already knew in 1992 that Putin would one day become a major power player in Russia, however, in the early '90s public officials and members of the security services played a major role in post-Soviet Russian politics. By employing a sales representative close to the deputy mayor of St. Petersburg and having sold their equipment to a firm whose owners were also closely associated with the regional key figure, Siemens found two ways to buy "security" in a highly instable and unpredictable environment. Investigations indicate Nikolai Shamalov was linked, by the early 2000s, to much higher level officials (Grey, Bush, and Anin, 2014). 12 Confirmation came in 2014, when the EU, in response to Russia's annexation of Crimea, put Shamalov on the sanctions list that targeted members of Putin's inner circle (Dalton 2014). 13 It appears Siemens exercised caution in its long-term investments in Russia to some extent: it diversified investments across several industries, though often with the same state-owned companies. It developed, for instance, several business ties to Gazprom in various industries; these ties continued when Gazprom's CEO changed in 2003. As the hypothesis expects, in cases with high stakes Siemens used intermediaries with ties to the political power-holders of important cities and regions. In addition, it minimized its exposure by having business ties across several sectors. In other words, the third hypothesis finds at least partial support.

Official Moguls in China
In countries with the Official Mogul syndrome, we expect that MNCs will attempt to employ intermediaries with established ties to high ranking public officials. In such countries, corruption is used to support the ruling regime by rewarding loyal followers. This system of patronage and structural corruption ensures that any economic and political opportunities can only be used with the consent of the political elites (Johnston, 2013: 86-118 While the report by the Security and Exchange commission entails several references to projects in China that are suspected to be the result of bribes, the following case was not mentioned. It is, however, the case with the most publicly available information. The cases investigated by the US law enforcement agency appear to follow a similar corruption pattern but the SEC report does not offer enough details to identify the individuals involved and thus does not allow for more in-depth examination of the case. 14 The SEC report does not mention any bribery on part of Siemens in the Chinese telecommunication sector, but documents forwarded by US investigators to the Chinese authorities made it possible for the latter to start its own investigation (Smith, 2011

Conclusions
This section comes back to the questions posed at the beginning of the paper: what kinds of corruption do MNCs engage in foreign markets? Do they take their home country practices to the host country or do they adapt to host country patterns?
In established market democracies with strong institutions, Siemens did not resorted to overt corruption and instead has operated within the legal limits. Campaign contributions and lobbying activities are its chosen modus operandi. This modus operandi, however, changes when Siemens operates in countries characterized by Elite Cartel corruption. The prevailing stability and predictability in such countries means that Siemens' business activities are safe as long as the firm does not threaten the business interests of the domestic elites. In the Italian case, Siemens used two tactics to ensure that they a) gain necessary local knowledge to access to the power generation market, and b) show that they do not harm the interests of the elites. For the first condition, Siemens employed as consultant a former manager of theirs, who had extensive knowledge of the Italian market. For the second condition, Siemens formed a consortium with two subsidiaries of an Italian conglomerate.
The case of the Oligarchs and Clans as well as Official Mogul syndrome follows a different corruption tact than which occurred in the Elite Cartel syndrome. However, the tactics are somewhat similar to each other. Siemens employed individuals as sales representatives who were close to the regional oligarch-network or high-ranking public officials. These sales representatives acted as intermediaries for Siemens and enabled the firm to establish ties with the ruling elites. In addition, Siemens used the same mode of entry for entering the markets. It established subsidiaries in almost all provinces in China and Russia and relies on locals to manage these subsidiaries. One explanation for this would be that corruption is, in the end, still the same and it only differs in the scale of corruption.
This inference, however, may be premature, as there exist differences in the syndromes. First, the ultimate recipients of the bribe payment differed -in Russia, several individuals along the way received `compensation' for their efforts of securing lucrative deals for Siemens. This is in line with Johnston's (2005) suggestion that the decentralized and instable corruption dominant in Russia means that several figures have to be bribed along the way. In contrast in China, where corruption is centralized within the bureaucracy -Siemens relied on its contact with a local high-ranking official for its business and the bribes and only had to be split between the official and the intermediary. In addition, Siemens tapped into an existing corruption network in China, which is similar to the Elite Cartel syndrome -in both cases, established corruption networks provide stability for businesses where the formal institutions cannot provide of it. In the case of Russia the MNC played a role in establishing the network. Last, the level of economic opportunities available and the security and stability formal institutions provide differs strongly across the three countries. Siemens thus had to use different corruption tactics in each to ensure protection for its investments.
In conclusion, we find that hypothesis one, two and four are supported by the cases, and that hypothesis three is partially confirmed. These findings support Johnston's argument that there exist different forms of corruption and a simple high-low measurement is ill-suited to capture the variations. Siemens also stands as an example for the recent rise in investigations by the domestic authorities. The number of recent corruption scandals surrounding large MNCs suggests that these firms might want to reevaluate the cost andbenefits associated with corruption. While investigating the corrupt activities of MNCs often stretches the limited resources of host country law enforcement authorities, countries do at times investigate, prosecute, convict and fine.
This study contributes to the literature on corruption by arguing that MNCs have recognized something that scholars have not yet fully realized -corruption operates differently in various countries and therefore requires different business tactics, depending on the formal and informal institutional structures in place, the "syndromes of corruption." The political economy and international business literatures need to take into account that corruption, and the risks of being caught, have serious implications for MNCs and their business strategies and tactics. Though Siemens is just one case, indications from other investigations indicate that it is exceptional only in the reach of its business operations, not in its practices.
Future research is still necessary as several questions remain unanswered. Our study's reach is modest: we have only covered corruption cases in four countries, while Siemens alone was being investigated by authorities in approximately 25 different countries. We have only covered one country in each syndrome. Our claim, based on an exploratory case study, is modest as well. A possible next step is to group the full set of known Siemens cases by corruption syndrome of the country in which they occurred, and assess whether similar patterns obtained, as expected by the hypotheses. Expanding the study to other MNCs, such as General Electric, would also be necessary in order to have greater confidence in findings. Comparing Siemens' tactics across countries that are in the same corruption syndrome would also be useful, to see if features of those syndromes do, consistently, affect in the expected manner the way Siemens (or another MNC) uses corruption. In addition, we have not been able to address the question of whether and if so how, Siemens' tactics have changed since it had been prosecuted and fined by the US and some other countries. If the country syndromes have not changed, one would expect Siemens to search for even more camouflaged means of participating in corrupt "markets", or that it would quit doing business in those countries. As the latter does not appear to have happened, it may have found new means of adapting. By extension, this would be something that could be investigated comparatively among several MNCs. While we have drawn analytical attention to a possible correspondence between syndromes of corruption and MNC corruption strategies, much further research needs to be done.