Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

1 Introduction

For a transnational rule to be implemented and interpreted on a national level it has to be understandable. In other words, the reasons for the rule to come about have to be expressed and appear legitimated in the local context where it is meant to be applied. When importing a rule to a national context, two interconnected processes of sensemaking occur. First, in order to make it feasible to import a rule, the local conditions must be understood in a way that makes the rule appropriate to apply in this context. This implies that in order to legitimize import of a rule, either some local problems and/or some material conditions must be perceived to exist to which the rule applies. In addition, the norms behind the rule have to be perceived as legitimate. Second, after the import, as the rule is implemented, it has to be explained and its content argued for. This second process of interpretation is partly dependent on the first one. If the rule is understood as appropriate for solving a problem in the specific national context, then its content is easier to explain and legitimize. While most of this book focuses on the second process of interpretation, in this chapter we will focus on the first process, i.e. how import of a rule gets on the regulatory agenda and how an understanding is built of the local context as appropriate for the import of certain rules. We will do this by introducing the case of the import of a code of corporate governance in Sweden and describe how the perception of ownership of Swedish public listed companies is sharpened in order to suit the transnational rules of minority protection and the effect of this on the content of the implemented rules.

Code of corporate governance is a British invention, which was widely diffused to different legal systems during the 1990s. It is one of several regulatory devices for gaining convergence of different corporate governance systems in order to ease the global integration of financial and product markets (McCann, 2007; Oxelheim, 1996; Oxelheim & Randoy, 2003). Through implementing similar rules for the governance of listed companies, the information and evaluation by international investors of these companies are facilitated. In this way, codes of corporate governance are similar to what Goode (1997 p. 2) defined as “transnational commercial laws”, i.e.

…law which is not particular to or the product of any one legal system but represents a convergence of rules drawn from several legal systems or even, in the view of its more expansive exponents, a collection of rules which are entirely anational and have their force by virtue of international usage and its observance by the merchant community…

Within the EU, the corporate governance code has been legally enforced, as each country is required to have such a code as part of its regulation of the stock market (2005/162/EG). The contents of these codes are, however, not legally regulated, but part of the national self-regulation system. Regulating the existence, but not the content, of a code opens up for variations in both the rules included and the understanding of why they are appropriate in a certain context. In other words, the development of codes of corporate governance includes both processes of interpretation mentioned in the first paragraph above.

Transnational regulations are not merely motivated by local demands, but furthermore imposed by negotiations and discussions on the international level. Our assumption is that import of regulatory reforms is supported by agendas emerging from the international arena about what is right and what is wrong. In other words, the international agendas are important for the process of making a rule appropriate to a specific context. Our intention is to investigate how these international agendas may make import of rules understandable and thereby influence local action.

The rest of the chapter is organized in the following way: First, we develop a frame of reference for how the agenda setting on the national level is influenced by the international development in the area, and how and why a code of corporate governance appeared as a solution to the regulative problems within the Swedish corporate governance system. Second, we investigate whether the assumptions behind the agenda setting that led to the code hold true. Third, we discuss the extent to which these assumptions are reflected in the code. We end the chapter with a discussion about the relation between perceived problem, actual situation and solutions in a situation of regulatory imports.

2 Frame of Reference

In order to investigate how an understanding is built of the local context as appropriate for the import of certain rules, we first have to gain knowledge of how the issue arrived on the regulatory agenda. To do this we apply the so-called agenda-setting theory developed within political science. This theory has been developed to describe political decision-making in the United States during the 1980s. In order to be applicable to the regulatory state of today, we complement the agenda-setting theory with the idea that central legal regulation has been exchanged for the regulatory state and the so-called programmes for regulating certain sectors of society. We take this notion a step further and claim that, regarding transnational regulation, these programmes are international rather than national. They imply a certain rhetoric that may be used to build an understanding of the appropriateness of a certain rule. In other words, the programmes give a basis for legitimizing import of a rule by defining either some problems and/or some material conditions that motivate this import.

2.1 Putting the Corporate Governance Regulation on the Agenda

One classic theory of how an issue gets on the regulatory agenda is the agenda-setting theory (Kingdon, 1984, 1995). This theory is based on the garbage-can model (Cohen, March, & Olsen, 1972), which is a general social theory of problem-solving processes. It is built on the assumption idea that problems and solutions are disconnected streams of ideas that occasionally meet in problem-solving. A problem-solving process may therefore start with a solution looking for a problem to apply to, just as well as with the definition of a problem. Applied to political processes this implies streams of problems and solutions; politics and actors are partly separated and connected on occasions when windows of opportunity are opening up and the different streams coincide. The window of opportunity may open up due to a change in material or other conditions that influence the ideology or legitimacy of politicians (Kingdon, 1995, p. 172). Politics in this context have to do with managing the flow of actions carried out by politicians in order to be perceived as legitimate or to be re-elected. In its simplicity, the theory may give the impression that a certain amount of haphazardness is involved in the agenda-setting processes. However, in his original development of the theory Kingdon (1984) emphasizes that the political entrepreneurs, who are “willing to invest their resources in return for future policies they favor” (Kingdon, 1995, p. 204), are important for which, and in what way different, issues enter the political agenda. These entrepreneurs are usually actors a bit offside the centre of the political power, but may, of course, also be people in the centre of political power. O’Malley (2007), for example, shows convincingly how Tony Blair most skilfully uses release of information and structuring of issues in order to achieve a decision regarding the United Kingdom’s involvement in the war with Iraq. The conclusion is that the influence a person may have on a political process rests on the ability to manage the different streams of problems and solutions in order to get an issue on the agenda.

The agenda-setting theory has been criticized for having concepts which are diffuse and difficult to operationalize and because profound qualitative, rather than quantitative, studies are needed in order to develop or confirm the theory (Soroka, 1999). Without criticizing the theory as such, recent research has pointed at a substantive complexity in the interactions between actors when different agendas are set (e.g. Hancher & Moran, 1989; Miller & Rose, 1992). In addition, one may question whether Kingdon’s theory is context-dependent and less suitable in a neo-liberal European landscape than an American governmental structure during the 1980s. Two aspects of the theory will be highlighted here. First, the perception of problems and problem-solving models as more or less disconnected streams, i.e. they may exist in relative independence of each other and have to be connected by an occurrence or an entrepreneur. Second, the perception that changes in existing conditions (i.e. a problem) or politics is the trigger for setting new agendas. The first aspect we will, hopefully, confirm by discussing how different streams create an agenda. The second one we will more or less reject, or at least redefine, by stating that it is not changes in existing conditions that create agenda-setting and the consecutive regulation, but the perception of problems that are salient for political entrepreneurs and in popular societal discourses.

2.2 Putting a Code of Corporate Governance on the Agenda

Reformation of the public firms’ governance through regulatory changes has been at the heart of the public agenda since the early 1990s. The EC company directives, and the changes therein, the OECD corporate governance code, the US Sarbanes-Oxley Act and the UK Combined Code, are examples of this wave of regulations. In Sweden, a new Companies Act and a first formal binding code of corporate governance have recently been issued. Much of the Swedish regulatory discourse in the area departs from a perception that a drastic change has occurred in ownership structures of the public firms and that Swedish ownership structures have gone from individual and local to institutionalized and global ownership. This in turn – it is argued – implies a need for regulatory reforms. From an academic point of view one might understand the perception of changing ownership as a basis for regulatory changes as a confirmation of Coffee’s (2001) conclusion that changes in the material condition for corporate behaviour are followed by new regulation.

A common international solution for the problem of regulating firms in a corporate governance system characterized by dispersed ownership is the introduction of a code of conduct. The code of corporate governance is a British invention, closely linked with the British context and legal tradition. In 1991, in the aftermath of several financial scandals and collapses, the Financial Reporting Council, the London Stock Exchange and the accountancy profession established the Committee on the Financial Aspects of Corporate Governance. The Committee was chaired by Sir Adrian Cadbury and their report, issued in December 1992, became widely known as “The Cadbury Report”. In Britain, the Cadbury Code has been succeeded by a number of new reports and codes covering new issues and parties in the corporate governance system. The code of corporate governance, thereby, is regulation descending from a corporate governance system (the Anglo-Saxon ones) where the ownership is dispersed.

In Sweden, a code of corporate governance was introduced in 2005. It was initiated by the governmental “Commission of Trust”. The commission was appointed by the government in order to rebuild the trust in the Swedish business community after a number of scandals during the late 1990s. To implement a code was not a part of the original directive from the government to the commission, but soon it became one of the commission’s main objectives. One important reason for putting a code of conduct on the agenda was to make Sweden interesting for foreign investors who were used to the existence of such a code in their home countries (SOU 2004:47, The Commission of Trust). The commission therefore formed a “code group” in cooperation with Swedish major business interests in order to develop a code of corporate governance. The problem to be solved by the code was, according to the preparatory work of the Swedish Code of Corporate Governance by Commission of Trust and the code group, to create potential for the exercise of an active ownership role as the ownership situation of the public firms shifted from individual and local ownership to institutional and international owners, i.e. when ownership became more dispersed (see The Commission of Trust, SOU 2004:47, also perfectly in line with The Company Committee, SOU 2001:1). Thus, the perception was that when ownership disperses there is a demand for new regulations protecting the dispersed ownership from a re-concentration, or what we might call ex ante minority protection (cf. Coffee, 2001).

The commission’s arguments in favour of formation of a specific code group were the advantages of self-regulation and market enforcement, supervised by, and under the constant threat of, state intervention. The establishment of the code group was also in accordance with the traditional Swedish way of developing regulation. Sweden has long been characterized as applying a “corporatism” mode of regulation (Puxty, Willmott, Cooper, & Lowe, 1987). This implies that problems and solutions leading to regulation are detected and discussed in cooperation between different interest groups and the state, and with considerations of the contextual situation in hand. To sum up the development that led to the Swedish corporate governance code: the problem to be solved was defined as active ownership (minority protection) in context with dispersed ownership, the solution was a code of conduct and the technique self-regulation under the supervision of the state. The question is why this specific solution appeared on the agenda.

2.3 Finding Problems and Problem-Solution Models

Both the problem defined and the solution discussed by the Commission may be seen as part of a larger agenda for converging different national corporate governance systems. The globalization of the financial markets is most often a salient argument for convergence of corporate governance systems and has been the subject of lively discussion on the international arena by politicians as well as by researchers. Corporate governance is a label for a number of actors and control mechanisms important for the control of the public corporation. The issue of convergence of different national corporate governance systems thereby includes a manifold of aspects and areas for regulation. It may be defined as a bundle of issues connected under one label, or what Miller and Rose (1992) labelled a neo-liberal programme for regulation.

A neo-liberal programme may be seen as a technique for the governance by states that has evolved during the last decades (cf. Moran, 2002; Lodge, 2008). According to Miller and Rose (1992) the neo-liberal government is characterized by on-distance governing. This implies that central legal regulation has been exchanged by the so-called programmes, ideas and solution models related to different societal problems that are supported by different actors in various organizations simultaneously and lead to different kinds of regulation, not merely laws. Miller and Rose (1992) describe New Public Management (NPM; Hood, 1991), the wave of reforms imported to the public sector from the private one during the late 1980s and the 1990s, as one such programme on an overriding level, and quality assurance in the health care as one more specific programme within NPM. In both cases, a manifold of actors and types of regulations were involved. This implies that even if problems, politics and actions are partly disconnected, they may be sorted in under different programmes.

Many of the neo-liberal programmes have been developed on the international arena and diffused to various national states (cf. Meyer, 2003; Meyer, Boli, Thomas, & Ramirez, 1997). For example, the idea of convergence of different national corporate governance systems is connected to the emergence of international financial markets. The emergent international financial markets are also closely connected to the more general globalization of the business society. According to Fligstein (2001), globalization (together with the neo-liberal discourse) is to be seen as an American programme up to now mostly materialized in the productions- and sale-organizations of the global company. In addition, one may claim that also the capital that finances these global companies is internationalized (Oxelheim, 1996). Besides this, Fligstein maintains (2001) that one should distinguish between globalization as an ideology, the idea of and the desire to make the business world global, and globalization as the material effects that really may be deferred to globalization of the businesses.

In this context, code of corporate governance can be seen as part of the international discourse, or ideology, of convergence of corporate governance systems and this discourse can be seen as emerging from an American programme. One condition that appears as important for this programme is ownership (cf. Gourevitch & Shinn, 2005). Scientifically, the point of departure has been the condition of diffused ownership for both empirical research and theoretical development (Bearle & Means, 1932; Alchian & Demsetz, 1972; Fama, 1980; Fama & Jensen, 1983). Theoretically a basic assumption is that diffused ownership implies an efficient capital market.Footnote 1 However, diffused ownership implies that no individual owners have the incentive to perform the needed control of the top management team.Footnote 2 It is usually more efficient for the shareholder of a company with diffused ownership to sell her/his share than to attempt to influence the top management. Instead, the control over the top management team has to be carried out by other mechanisms, for instance by means of the signals from the market and hostile takeover.Footnote 3 In contexts with ownership concentrated to one or a few strong owners, the strong owner is expected to control the top management, but another problem arises: how do we prevent the large owners from enriching themselves at the cost of the minority owners? The answer has been to suggest more comprehensive regulation to protect the minority (cf. La Porta, Lopez-De-Silanes, & Sheifer, 1999; Coffee, 2006). The American programme regarding owners may, thereby, be said to include problems and solutions related to various types of owners, market effectiveness and different relationships between owners (minority/controlling owners). The rhetoric or ideology of this programme indicates that the corporate governance system with newly dispersed ownership needs to implement a new regulation, for example, an internationally accepted corporate governance code, in order to deal with the new ownership situation. As seen above, this was the case for the Swedish commission for trust and its code group.

Being the basis of the international discussions in the area, the American programme regarding ownership may be viewed as a “worldwide model” (Meyer et al., 1997, p. 44) of corporate governance:

“Worldwide models define and legitimate agendas for local actions, shaping the structures and policies of national states, and often local actors in virtually all of the domains of rationalized social life – business, politics, education, medicine, science, even the family and religion.”

Worldwide models are created and diffused by global actors such as states, professional organizations, pan-national policy organizations, etc. To be internationally accepted and legitimated and in order to be perceived as modern and “good”, nations have to implement (at least formally) some of these models. Worldwide models, thereby, offer problems and solutions as well as legitimacy for the political entrepreneur to act on. The implementation of such a model is more closely connected to the stream of action made by politicians in order to be perceived as legitimate, rather than get on the agenda due to change in material conditions (cf. Kingdon, 1995). In the case of implementing a code of corporate governance, therefore, the condition for ownership in a specific context will not be the basis for putting the code on the agenda, but will be re-interpreted in the light of the dominant world model at the present time (cf. Jonnergård & Larsson, 2007).

To view the code of corporate governance as part of a worldwide model may explain the diffusion of the code around the world. Since the early 1990s a code of conduct has been one dominant solution to demands for regulation in the area. In particular, the Cadbury code, issued in 1991, has become the standard for codes implemented in other national contexts (see above p. 5). The idea of a code has spread quickly throughout the world. The European Corporate Governance Institute (ECGI) lists on its website (http://www.ecgi.org/codes/all_codes.php) over 140 different codes from all over the world and from different pan-national or international organizations. The initiators of the codes and the status of the codes vary substantially. Many nations have several codes that are issued by different bodies and enforced in different ways. What seems to be similar about the different codes (with a possible exception of the Australian one) is their structure. The codes include the same issues, even if the substantial regulation about them may differ. Apparently some issues have to be included in order to make the code legitimate in the eyes of the world society and the foreign investors. In this way it is a powerful model for problem-solving in different converging corporate governance systems.

2.4 Summarizing the Framework

In summary, if transnational rules are to be implemented, they have to be understandable in the national context. This is brought about through two processes of interpretation: one that explains why the rule is suitable in the specific context and one that clarifies the content of the rule. The first of these processes of interpretation may be viewed as agenda-setting, where the agenda may arise as a reaction to either a change in the material conditions in hand or a change in ideology or rhetoric in order to increase the legitimacy of the politicians or the political process. In the case of implementing a code of corporate governance we claim that the latter is valid, i.e. the implementation of the code is part of an ideology for supporting the globalization of the business society. This ideology arises from an American programme that has become dominant in the international discussion in the field; it has become a world model. The programme includes a number of perceptions, problems and solutions held together by some common assumption of the world (for example, the importance of diffused ownership and capital markets for economic growth). The worldwide model of diffused ownership offers a suitable description of a problem (and/or solution) and calls for a problem-solving model for regulation. A code of corporate governance offers such a model and prescribes at the same time the items that ought to be included in the code. To import a corporate governance code, thereby, implies that the first process of interpretation of the regulation is connected to the international discourse and understanding in the area, rather than to local material conditions.

3 The Perceived Problem of Ownership Change in Sweden

In the following sections we will illustrate the framework given above by describing how the ownership dispersion in Swedish firms was introduced as a problem (in Kingdon’s use of the word) and to be solved by the introduction of a code of corporate governance that appeared. In the first section, we trace the agenda in use by the regulators in various preparatory legal works proposing corporate regulations. It is argued here that the perceptions of ownership posed in these preparatory works limit which regulatory solutions could be applied and that this implies – in this case – a perceived correspondence between the local conditions of regulation and internationally set agendas. In the following two sections, this argument is developed by looking deeper into the actual statistics on ownership of the Swedish public firms. The final section presents an alternative way of viewing the ownership situation in Sweden that will be contrasted with the actual regulation developed in Sect. 4.

3.1 Regulators’ Perceptions of the Ownership of the Public Swedish Firms

As mentioned, the regulatory mode of Sweden has been defined as corporatism (see Puxty et al., 1987). This implies that economic regulation is developed through negotiations between the national elite in politics, labour and business, and that consensus has been established around solutions on important issues such as the division of the industrial surplus and the balance between self-regulation and the law. The close collaboration between different parties and the consensus solutions are fundaments in the so-called Swedish model that have been present since the 1930s (see Högfeldt, 2004; Stafsudd, 2009). This model has supported and been supported by a concentrated ownership among the Swedish companies, strong unions and a stable political situation.

The ownership of the Swedish industry goes back to the boom of entrepreneurship that took place between 1860 and 1931 when most of the large Swedish companies existing today were established. Through the financial crises and corporate scandals (the most notable being the Kreuger scandal) during the 1930s, the ownership was concentrated to the main Swedish banks. Legal measures, however, implied that the bank ownership was moved into investment companies, controlled by the bank. Together with some visible industrial families, these investment companies acted as parties in the negotiations with the state. The existence of few visible owners as opponents in the negotiations increased the efficiency of the Swedish model (Collin, 1998). To date, the concentrated ownership has been supported by different legal means such as the existence of dual class shares, pyramiding and cross-holdings.

The traditional political elite of Sweden have been the Social Democrat party. This party has held power most of the time from the 1930s to 2006. When the party gained power, large industrial companies were considered the most important unit of production, the ownership of which, in time, was to be transferred to the public, creating “social enterprises without owners” (Stafsudd, 2009). A move that according to Henrekson and Jacobsson (2003) would be eased by a large owner concentration as it would be easier to socialize large firms in a few hands than smaller firms in many hands. However, over time the Swedish Social Democrat party has been characterized more by democracy than by socialism and since the late 1970s and the early 1980s the party has supported private ownership and the freedom of incorporation (see Henrekson & Jacobsson, 2003).

Thus, one might say that the traditional Swedish model was formed by the tension between corporatism and socialism. Two main perceptions of ownership coexisted. The first one viewed concentrated ownership as advantageous as they were supposed to support the industrial development in collaboration with the state and the union. The second one viewed the present situation as a step towards a future stage with “social enterprise without owners”. Both these views, however, led to a perceived mandate for social engineering of the ownership situation of the firms, predominantly through (tax) regulation.

The different views on ownership have initiated a number of parliamentary task groups on desirable ownership structure. The latest one was the so-called Ownership Commission. This was initiated in 1985 as a response to criticisms from the Communist party’s leader CH Hermansson (1971) and his, in Sweden, famous naming and shaming book on the “fifteen families” who were in control of Swedish business.Footnote 4 In the Ownership Commissions, different experts were teamed with politicians from different parties. Their work was performed at a crossroad for Swedish ownership politics. When the result of the commission’s work was reported in 1988, East European communism was starting to collapse and a new area of deregulation and liberalization was well under way. Thus, there was a perceived need for a new way of describing the ownership situation of the Swedish listed firms and it became the main task for the commission to fulfil this need. The Ownership Commission made a very thorough analysis of the ownership situation of Swedish firms. In their conclusions, the commission highlighted an identified growth in internationalization and institutionalization of the ownership of the Swedish firms. This conclusion later became established as the truth, in the public and scientific discussion of Swedish ownership, and a new (third) perception of Swedish ownership emerged.Footnote 5

The notion that the ownership of the Swedish public firms has become internationalized and institutionalized prevailed when new regulation for the public firms was proposed. This is illustrated by the following quote from the preparatory work to a new Company Act (SOU 2001:1):

“Another aspect worth mentioning at this point is the large changes in the ownership structure amongst Swedish public firms that have taken place during the second half of the 1990s. A great many of our largest firms are today more than half owned by foreign legal entities. These foreign owners are often pension funds and institutions that invest a fraction of their assets in stock, and which very rarely interfere with the company management, but are rather more likely to sell their stock when displeased with management. However, it is not only the category foreign owners which to a large degree consist of institutions. Also the Swedish owners comprise a great number of funds, insurance companies, and other institutions, while the physical persons that just a few decades ago held a majority of our quoted firms, today have seen their positions greatly reduced.”

Put more succinctly by the Commission of Trust (SOU 2004:47) who is the issuer of the corporate governance code scrutinized here (our translation):

“The ownership of the public firms has to a great deal been institutionalized and internationalized.”

Besides a new perception of who owns the Swedish companies, these quotes illustrate a new attitude to the relation between the owners and the state. Following the Ownership Commission’s conception of ownership, the idea of social engineering by influencing firm ownership was dismissed; internationalized or not, the ownership development was not perceived as being able to change by political means. The objective of the regulation has instead turned into adopting the situation in hand, that is, how to facilitate active ownership in the face of internationalization and institutionalization rather than providing for more Swedish and private owners. However, as the Swedish financial markets have yet not proved themselves efficient in disciplining management – already the Ownership Commission noted the weaknesses of the Swedish market for hostile takeovers – there is a need for regulation compatible with the new dispersed ownership.

3.2 Foreign Ownership of Swedish Public Firms

It should be clear that the Ownership Commission (1988) only superficially treated the internationalization of the ownership. The commission noted that it rose from 4 to 8%, but it fell back to 6% during the surveyed period. Consequently, the commission only considered an increase in foreign ownership as a trend that had to be taken into account in the future and which may change if certain restrictions such as limitations of foreign ownership stakes in high profiled firms and sectors were removed. As the financial markets in Sweden were deregulated in the late 1980s, foreign ownership increased to about 30% and skyrocketed in 2000 to 40% (Statistics Sweden, http://www.scb.se). The importance of foreign ownership for the development of industry was highlighted in both academia (i.e. Henreksen & Jacobsson, 2003; Jonnergård & Kärreman, 2004; Agnblad, Berglöf, Högfeldt, & Svencar, 2001) and public commissions (i.e. The Companies Act Commission, SOU 2001:1 and The Commission of Trust, SOU 2004:47). The idea of these public commissions seemed to be that by issuing regulation familiar to foreign owners, they could be expected to act as responsible owners (see Sect. 2.2 above).

The figures from Statistics Sweden, the official Swedish statistics bureau, presented above, do not reveal the identity or the intentions of the foreign owners. In Fristedt and Sundqvist’s annual book “Owners and Power in Sweden’s Listed Companies”, however, the biggest foreign owners (around a third of all the foreign owners) are listed. This listing implies that foreign owners are not a homogeneous group. The list includes business firms (such as Renault, Volkswagen and MAN), foreign states (such as Finland and Singapore) and large international institution investors (such as Fidelity and Franklin-Templeton). The heterogeneity among the foreign owners may imply heterogeneity in the owners’ intentions as well. For example, it is likely that a global truck manufacturer, participating in the global consolidation of the truck market, has a different view of their shareholdings in VolvoFootnote 6 than a New York index fund manager.

As we do not know the identity of smaller foreign owners we cannot be sure of the foreign owners’ connections to Sweden. For instance, there are financial products only available to Swedish residents that for tax purposes use a middle country (all Swedish banks have so-called Luxemburg-funds in their offers to customer) when investing in the Swedish stock market. There are also certain holdings from Dutch foundations with connections to former Swedish citizens indicating that money moved from Sweden for tax purposes is being reinvested in Swedish firms.Footnote 7 These persons are still major owners of their family companies acting behind legal tax constructions and will probably relate to their ownership in a markedly different way than, for instance, a foreign government.

Finally, foreign ownership does not affect all public firms in the same manner. On the contrary, there is a tendency for foreign owners to invest in the largest Swedish firms. This is illustrated in Table 14.1 where we can see the foreign ownership of the firms, as mean, median and standard deviation for both capital and votes in the firms listed on the Stockholm Stock Exchange in 2007. From these numbers it becomes clear that not all public firms are affected equally by the foreign ownership.

Table 14.1 Foreign ownership on the Stockholm stock exchange 2007
Full size table

In other words, instead of being one common category, the international investors are of different kinds, probably with various intentions with their ownership and with different approaches to the control of the firm.

3.3 Institutional Ownership of Public Swedish Firms

The concept of institutional ownership has been at the heart of the Swedish political discourse since the 1970s.Footnote 8 On the one hand, some discussants claim that institutional investors are desirable from an efficiency perspective, since the firms are owned and controlled by “professional” owners, with no strong feelings for the family name or emotional connections to the manufacturing municipality or the business. On the other hand, when the social consequences of fast-moving capital have been noted – at plant closing, labour lay-offs or the sale of “firms of national interest” – the lack of “flesh and blood” owners has been portrayed as a hazard for the long-term development of the society. These standpoints can be viewed in politics (without any right–left implications), from the business society itself and in media; the institutionalization of the ownership is always a topic of interest in Sweden.

In the 1980s, the Swedish Ownership Commission put great emphasis on the constant decline of individual direct holdings of stock that had been taking place since the 1950s, when the Swedish households’ ownership accounted for almost 80% of total market value on the Stockholm Stock Exchange. Shortly after the Commission report was presented in 1988, the decline in individual stockholding started to level out and has since 1991 been between 10 and 15% of the total stock market capitalization (http://www.scb.se). The decrease in individual ownership has been interpreted as an indication of a corresponding increase in institutional ownership. Individual stockholdings are, however, affected by various issues, most importantly, taxation. Individual shareholdings, and stockholding by foundations or closely held firms, are treated differently by the Swedish taxation system. This has, over time, had an effect on the ownership patterns (Henriksen & Jacobsson, 2003). In the early 1990s, for example, transferring capital into a private corporation or in a private foundation that, in turn, invested in stocks, led to a tax reduction, while individuals who owned stocks were taxed. In such a situation it would be highly irrational to continue to own controlling blocks of shares as a private person instead of using a private corporation as a “vehicle of control”. However, it is still the same private person who controls the public firm even though the shares in the public statistics are held by a private corporation (i.e. an institution). In other words, it is rather the form of ownership than the ones who own the shares that has changed in Sweden over time. In this chapter we therefore do not treat private shareholdings through private corporations or foundations as institutional ownership. We base this argument on the notion that this kind of ownership solution does not distort any incentives for control and it does not create any new agency costs (cf. p. 11 above). As long as the families or the financial groups are the owners behind the institutional forms, we still define the stockholding of the institutions as family or “sphere” ownership. It is a matter of who controls the firm, not the tax-purpose constructs or the vehicle of control that is the heart of ownership.

What about the institutional ownership then? In this chapter, institutional ownership is considered when an additional layer of agents is placed between the public firm and its owner. That is, when person A hands over cash to person B at his/her discretion and the cash is used to buy shares in a firm controlled by person C. Corporate governance usually focuses on the relation between person B and person C, but that would only be half the problem if one did not expect person B to act in total alignment with the interests of person A. Thus, it is the double-layer agency problem that makes institutional ownership differ from physical ownership. The size of what we consider institutional ownership is found in Table 14.2 The figures are summarized from the Statistics Sweden categories “Banks, financial institutes and more”, “Funds”, “Insurance companies, pension institutes” and “Social security funds”. What is clear from Table 14.2 is that this ownership form has been constant between 25 and 30% since the 1980s. A large increase assumed in the public discourse is not at hand.

Table 14.2 Ownership of the public Swedish firms (% of market value),
Full size table

In summary, we may distinguish two types of owners that use institutions as a vehicle for control. First, controlling owners and, second, formal institutional owners. The ownership share of formal institutional owners has been more or less constant during the last 25 years.

3.4 Ownership of the Swedish Public Firms

The data in Table 14.2 only present ownership of the cash-flow rights of the firms. In Sweden, the political preference for large individual owners (see Bergström & Samuelsson, 2001; Collin, 1998; Högfelt, 2004) has led to a legal framework that provides a number of possibilities to separate voting rights from cash-flow rights, so-called dual class shares. These have made way for clearly identifiable dominant owners in many firms. Dual class shares have been used by traditional owner families and financial group to retain their control over firms, by companies investing for operational reasons in other firms as well as by foreigners investing in Sweden.In Table 14.3 the control of the firms of the Stockholm Stock Exchange in 2007 is presented, ranging from dispersed ownership (or management control) to foreign owner control. However, the largest group is still individual (family) or sphere control.

Table 14.3 The control of the firms listed on the Stockholm stock exchange 2007
Full size table

As may be seen in the table, 66% of all Swedish firms are still owned by an identifiable controlling owner. The myth of dispersed ownership may therefore – if not neglected – at least be put in perspective.

In conclusion, the public agenda used in politics and regulations of the ownership of the public firms has focused a good deal on the perceived increases in institutional and international ownership, and the new regulation that may be needed due to these changes. The perceived changes in the ownership situation have, however, slight bearing on the structure of ownership in Sweden. Much of the ownership remains local and individual and consists of owners who exercise control of incumbent management. The names of the owner families and other constellations may have changed over time, but we see new forms of concentrated ownership rather than a general trend towards dispersed ownership in the Swedish corporate governance system. Likewise, it is obviously true that the foreign ownership has increased substantially since the market was deregulated in the 1990s. Foreign owners are, however, a heterogenic group, of which some act as controlling owners and others do not. The former do not need corporate governance reform to control the firms and the latter who do not act as owners would probably favour minority protection over possibilities to get involved in actual control. In other words, we have described a situation where there is a perception of a “problem of ownership” in accordance with the worldwide model described in sector two, but where the material conditions are not in agreement with this “problem”. With this as a background we now turn to discuss the effect the “new ownership notion has had on the code of corporate governance”.

4 The Solution of a Swedish Corporate Governance Code

After defining “the problem” and relating this to the material situation in hand, it is time to look into the suggested solution. As mentioned in Sect. 2, the solution to changes in ownership structure according to the worldwide model for corporate governance is the import of a corporate governance code and this was also the solution implemented in Sweden. As mentioned above, a corporate governance code most often contains a package of rules defined by those in the British Cadbury report. In other words, the structure of a corporate governance code is based on governance problems experienced in Britain, constituted by the British ownership structures and applied in relation to other British regulation (such as the London City Code on takeovers or the Companies Act). The Swedish code has a structure similar to the British one and labels the rules in the same way. However, the content of the rules has to a great degree been adapted to local circumstances (Jonnergård & Larsson, 2007). To understand the codes’ effect on ownership we thereby have to look at the content rather than at the label of the rules.

Viewing the code as a solution to the problem of diffused ownership, we focus on the code rules relating to different types of owners. As a first step, we categorize the rules favouring majority or minority owners. Majority owners, in turn, could be divided into individual (family) owners and sphere ownership, Swedish industrial and foreign, following the classifications in Table 14.3 Minority owners are further divided into local institutional owners, international institutional owners and small individual minority owners following the different minority interest groups reported in Jonnergård and Larsson (2007). The whole analysis is found in Appendix 1 and the Swedish corporate governance code can be found at http://www.bolagsstyrning.se.

As can be seen in Table 14.4, the rules in the Swedish corporate governance code in the majority of the cases (45 out of 69) relate to the conflicts of interest that occur between different owners in the public firm. These rules almost exclusively support the minority owners. The most favoured owners are the small individual owners and the international institutional investors. However, given their local occurrence and the size of their holdings, one should not underestimate the favours given to the local institutional investors either. Thus, the Swedish corporate governance code is a piece of regulation directed at protection and in favour of minority owners at the expense of traditional controlling owners. The small individual owners as the primary group benefiting from the regulation may be an attempt to revitalize the ownership of individual households after a long period of decline connected to the traditions of the Ownership Commission. The other two benefiting groups, the international investors and the local institutions, are very much in line with the thinking of the Companies committee and the Commission of Trust. However, most of these groups’ new forward positions came at the expense of the traditional owners, and that might be seen as something very unresponsive following the picture given above of actual ownership and control of Swedish public firms. The code is, thereby, very much in accordance with the worldwide model discussed in Sect. 2 and gives less consideration to the actual ownership situation. This finding will be the starting point of the discussion below.

Table 14.4 Number of rules favouring different owners
Full size table

5 Discussion and Conclusion

We started this chapter by claiming that in order to make transnational rules understandable two processes have to take place on a local level. The first one, which we have focused on here, deals with understanding the rule as appropriate in the national context where it should be implemented. The second one deals with creating an understanding of the content of the rule. In this chapter we have described the first of these processes of interpretation as agenda-setting processes in which problems and solutions are loosely connected and put together through either a change in material conditions or a need for legitimacy by the politicians. We have suggested that one source of both defined problems and solutions is so-called programmes, bundles of ideas connected to a specific social area, which may emerge in a national context, but which in many cases emerge on the international arena in discussions and negotiations within different international or pan-national organization of activities. In the case of corporate governance, we have identified an American neo-liberal programme as a framework which, by means of international cooperation in the area, has emerged as a worldwide model for defining problems and solution for the convergence of corporate governance systems. What is left to discuss is (i) how the worldwide model of corporate governance may be used to explain the new regulation and (ii) the possible effects the understandings from such a worldwide model may have when applied in the Swedish corporate governance systems.

The worldwide model of corporate governance supplies, first of all, an overriding understanding for why new regulation is needed. According to this model, new regulation is needed in order to facilitate efficient flows of investment on a global capital market. Such efficiency has for a long time been on the agenda for international organizations such as the World Bank, international cooperation, for example, in the OECD, and pan-national collaborations such as the European Union. The idea is, therefore, well-established and embedded in international networks. In addition, it has been on the political agenda internationally as well as nationally for several decades. In Sweden, for example, the capital markets were deregulated in the late 1980s in order to facilitate such a development. To facilitate efficient investment on a global capital market is, however, not only a political agenda, but built on economic theory and consequently built on certain theoretical assumptions and logical deduction. As such, the world model provides normative as well as cognitive arguments for why a certain regulation should be implemented and through this it may serve as a basis for understanding why a certain rule should be implemented. In addition, when discussing the diffusion of the code, we are dealing with a complex structure of national (Swedish) agenda-setting, depending on a pan-national (EC) agenda-setting. The choice of which agenda to promote on the national level is thereby partly delimited by the agenda-setting process on pan-national and international levels. Therefore, the politics involved affect the legitimacy of the politicians both to their own voters and towards the international community. The application of a well-known worldwide model is a way to promote such legitimacy. In other words, the American programme for corporate governance is a powerful model for explaining the import of certain rules as it gives a normative-political basis and a logical-deductive explanation for the rule as well as legitimacy in relation to a certain political agenda.

What we have shown in this chapter is that it is not self-evident whether the logic of such a model has to agree with local conditions in order to be implemented. Rather, some of the assumptions behind the model get interpreted on a national level as if they apply. In our case, we have observed that the ownership structure has not gone through any major changes, even though such changes have been used to argue for new corporate governance reforms. On the other hand, the perception of such changes has been part of the political as well as scientific discourse for some time and it is not surprising that they have become taken for granted. With a less thorough analysis of the number than ours, the figures of ownership may also be interpreted in a way that they agree with the perception of changes in ownership structure.

The issue is what kinds of effect the new regulation will have on the Swedish system for corporate governance and for the Swedish ownership structure. On this we may only speculate. In the literature regarding the relationship between the development of the capital markets and its regulation, two major hypotheses have been formulated. First, Coffee (2001, p. 14) claims that material changes lead to changes in regulation. He states:

In short, if form follows function (that is, if legal rules are determined by the system of corporate governance that pre-exist those rules), then no similar rapid legal transition should necessarily be expected in the Continental economies in which concentrated ownership is still the norm

If Coffee’s (2001) discussion holds true, we would not expect any great effect of the discrepancy between the material condition and the content of the corporate governance code. As form follows function, without a function (i.e. a material change that has to be dealt with) new regulations are condemned to be more or less useless. In this case the process for understanding why a rule is appropriate in the specific context will not help with applying the content of the rule as such.

However, La Porta, Lopez-De-Silanes, Sheifer, and Vishny (1998) and La Porta et al. (1999) suggest that a change in ownership structure is a both intentional and desirable effect of changes in the national regulation for corporate governance. In other words, they suggest that a change in regulation for minority owners, regardless of the initial material conditions, will give incentives for a more diffused ownership structure in the future, and thus to adjust to the worldwide model of corporate governance implies in the long run more efficient global capital markets. In this case the process for understanding why a rule is appropriate will probably be supportive in explaining the content and for the obedience of the rule.

These hypotheses are built on different empirical studies and partly different normative approaches. Coffee (2001) is a description of the historical development of ownership structure, while La Porta et al. (1998, 1999) build their conclusion on contemporary analysis of the legal systems in different countries around the world. While Coffee (2001, 2005) acknowledges differences between corporate governance systems, La Porta et al. (1998, 1999) use the American model as their normative point of departure. These changes make it difficult to speculate on whether the one or the other applies in our case.

One conclusion we may draw, however, from this chapter is the importance of not only understanding the rule as such when implementing transnational rules but also how this rule is explained as appropriate in the national context. This process may both give the framework for explaining the rule and influence its material effects as it gets implemented.