Actors and interests

Olson and others have made the argument that greater control of economic resources translates into greater political power, which can be used to shape institutions in the future toward the goals of those in power. This link thus presents, they argue, an economic incumbency effect (Olson Reference Olson2000; Morck and Yeung Reference Morck and Yeung2004; Acemoglu, Johnson, and Robinson Reference Acemoglu, Johnson, Robinson, Aghion and Durlauf2005). The authority structure of a firm decides who can lay claim to the cash flow. Hence, corporate governance affects wealth creation and distribution, the fate of suppliers and distributors, the fortunes of pension funds and retirees, and the endowments of charitable institutions (Gourevitch and Shinn Reference Gourevitch and Shinn2005). And the distribution of political resources via lobbying affects who wins elections and what policy options are chosen. Further, “the players in the firm, as they turn to politics to get the regulations they prefer, have to appeal to a broad set of external stakeholders” through the system of political institutions (Gourevitch and Shinn Reference Gourevitch and Shinn2005: 10). The size of this external stakeholder group will differ as the ratio of small shareholders to large shareholders changes.

Whether small or large shareholders dominate will affect the choice of political institutions. For example, the concentration of ownership and control are related to the basic features of a country’s legal system, and in particular to the choice of minority shareholder protections (LaPorta, Lopez-de-Silanes, and Shleifer Reference Laporta, Lopez-de-Silanes and Shleifer1999). In other words, other scholars have identified how ownership distribution and corporate governance are critical components shaping the political economy and the link between firms and political outcomes.

Table 2.1 shows variation in ownership concentration by comparing the average number of largest shareholders in a firm in each country under examination. Firms were asked to report the number of shareholders holding the largest packet of shares. Poland has the highest number, indicating that ownership is spread across a relatively high number of stakeholders. Bulgaria is significantly lower, and Romania has the lowest value. In the latter, larger shareholders dominate. The variation here is sufficient to lead us to expect very different types of coalition forming among firms.

Table 2.1 Ownership concentration, 2000

Source: EBRD (2005b).

The prominence of small and large shareholders is not the only factor that affects the macro-structure of ownership. Owners in a firm vary also by how likely they are to hold shares in only a single other firm, in a set of firms in a particular sector of the economy, or in a broad assortment of firms. For example, banks and financial firms are more likely to have an interest in many different types of firms as part of a portfolio of diverse investments. Industrial or family firms tend to hold stakes in fewer firms and will often vertically integrate other firms in supply chains. Thus, when banks are prominent, networks will be broader, and this raises the likelihood of collective action. Because banks and financial firms have an ownership interest in a much more diverse group of firms, I hypothesize that banks are more likely to lobby for broadly conceived rather than narrowly distributive institutions.

The cases analyzed later in this chapter therefore examine the form of owner (or combination of owners) that dominates: bank, institutional investor, state, industrial, or family. Broad and narrow networks each impart a different flavor to the broader behavior of firms in a particular political economy, as each owner type can perform a different characteristic function. These are shown in Table 2.2 and explained in more detail below. Broad ownership indicates that there are many smaller shareholders and firms widely connected to each other. Narrow ownership means that large block holders dominate.

Table 2.2 Owners, ownership structures, and resulting functions

As shown in Table 2.2, owner types can be divided into two categories: those that promote a horizontal structure by virtue of their direct ownership of many firms (see Figure 2.1a) and those that promote a hierarchical structure, such as family groups and business pyramids (see Figures 2.1b and 2.1c). Banks and institutional investors, for example, will tend to acquire stakes across sectors to spread risk across a portfolio. These firms will sometimes seek input in firm governance but are less likely to own controlling shares. By contrast, owners that promote a hierarchical structure will tend to value control.

Figure 2.1 Ownership structures

Table 2.2 also illustrates how each type performs a different characteristic function. Banks and institutional investors serve as information conduits and facilitate monitoring (Useem Reference Useem1984; Mintz and Schwartz Reference Mintz and Schwartz1985; Davis and Mizruchi Reference Davis and Mizruchi1999). Business groups reduce contracting costs and facilitate alliances (Gilson and Roe Reference Gilson and Roe1993). In contrast, the two hierarchical types that create narrow networks protect and retain control for insiders while facilitating risk sharing with minority shareholders (Rajan and Zingales Reference Rajan and Zingales2003).

Each form affects how control votes of firms are aggregated in the economy and spreads risk differently at the macro level of the economy. Family pyramids, for example, tend and often seek to alienate small shareholders while benefiting from the use of their capital. Institutional investors, by contrast, promote the interests of diffuse shareholders as intermediaries by virtue of the fact that, when these investors are prominent, they mediate between a large economy of shareholders and the firms that are the target of their investment.

What are the advantages of each form? Each structure provides a different solution to the problem of ownership and control in contexts in which risk is high and financial resources are scarce. The pure hierarchical structure (Figure 2.1b) joins both ownership and control: those possessing funds have full control of the firm(s). The pyramidal structure (Figure 2.1c) separates minority owners from control of the firm. A range of intermediate outcomes are possible, for example when multiple, cross-cutting ties link firms (Figure 2.1a), with the possibility of reciprocal holdings of shares among a group of firms such that mutual ownership blurs the lines of ownership and control but can reinforce alliances between firms (Gerlach Reference Gerlach1992). This last type can be either loosely or tightly connected.

The question of ownership and control in its classic form has focused on management issues that arise when managers are separated from owners. Different structures of ownership also offer owners different access to the resources of firms in a group, however (Perotti and Gelfer Reference Perotti and Gelfer2001). Whether a business group will add a firm in a pyramidal or direct (horizontal) fashion depends on the business environment. Loose horizontal ties are viable when coalition governance is possible and trust is high. In periods of financial distress, instead, collective action can become more difficult. As coalition-enforced threats become less credible, hierarchical control is preferred (Berglof and Perotti Reference Berglof and Perotti1994).

Different structures also allow owners to benefit to varying extents from the cash flow of the firm. Ownership in a structure in which the family takes on direct ownership of the new firm (as in Figure 2.1b) is preferred when the controlling shareholder does not wish to share any of the cash flow rights of any firm in the group with other shareholders, as it would in the pyramidal structure. The drawback is that, in creating new firms, the controlling shareholder has access only to its retained cash flow in the original firm but receives all the benefits and costs from the new firm (Almeida and Wolfenzon Reference Almeida and Wolfenzon2006). By contrast, pyramids allow the already established firms in the group to finance the acquisition or creation of new firms with capital belonging to minority shareholders higher up in the pyramid.

Hence, network structure is a direct reflection of the level of cooperation that is possible among holders of capital. More hierarchical structures make collective action among firms less likely, while more dispersed structures raise the likelihood that firms can collectively agree on the structure of institutions.

One feature that determines the structure of ownership in the economy is the identity of owners. Some owners will tend to promote certain kinds of structures. For example, if financial institutions dominate, then the structure of the economy will tend to be more horizontal, as funds and banks invest in firms and in each other. When families dominate, a more hierarchical structure will tend to emerge. The next sections explain how the identities of owners impact the macroeconomy by tracing the incentives generated by different types of horizontal and vertical owners. Based on an analysis of these types, I argue that horizontal structures raise the likelihood of collective action among firms.

Horizontal

Three types of horizontal owners – banks, business groups, and institutional investors – raise the likelihood of collective action.

Hierarchical

Family firms and individually owned industrial groups or pyramids are two hierarchical forms of ownership that lower the likelihood of collective action.

Firm collective action

As the previous sections show, firms that tend to promote a more horizontal structure raise the likelihood of collective action by facilitating information flow, monitoring, and credible commitments. Firms that promote hierarchical structures are better placed to defend the status quo and narrow elite interests, and thus they lower the likelihood of broad collective action.

Having introduced these different incentives for different types of firms, it is possible to construct a framework for business political action in terms of the factors that shape the extent of firm cooperation. Two key features governing the emergence of coordinated political action are (1) agency problems (if owners and managers are the same person, their interests overlap; if owners and managers are different people, agency conflicts are likely) and (2) the diversity of assets, which reflects the breadth of ownership networks.

What are the expected outcomes as these two features vary? High asset diversity, meaning that firms across different sectors of the economy have links through common owners, raises the likelihood of coordination, because common ownership interests trump asset-specific interests. This is the case because common ownership of firms in different sectors will tilt collective action toward non-sector-specific preferences, rather than industry-specific demands. As a result, firms will be likely to find joint ground on which to broadly coordinate their political demands. High levels of the agency problem (found under concentrated ownership) instead lower the likelihood of firms coordinating their political demands, because high levels of this problem allow controlling owners to separate minority shareholders from control of their investments, making credible commitments to broad coalitions difficult. This relationship is summarized in Table 2.3.

Table 2.3 Agency problems, asset diversity, and the likelihood of collective action

Why does concentrated ownership lower and dispersed ownership raise the likelihood of collective action? Concentrated ownership is a sign of low levels of underlying social trust and increases the influence of a small and established elite that seeks to sustain the status quo. This elite also faces incentives to exercise political pressure in order to sustain the status quo. According to Morck and Yeung (Reference Morck and Yeung2004: 392), close ties “between members of a small elite magnif[y] the returns to political rent seeking by this elite.” Investors hold concentrated shares only in order to exercise control and gain a strategic input into the management of the firm (this block need not be a majority stake, because small shareholders often do not vote). If investors are not concerned with control, according to portfolio theory, passive portfolios should be diversified across industries and countries to reduce risk.

Owners are often concerned with control when they are faced with conditions in which passivity equates with risk. In other words, the ownership structure is in some ways a reflection of the extent to which there is social trust and elites expect to be able to cooperate. Dispersed ownership is possible in contexts in which the expectation is high, while concentrated ownership exists when common ventures among strangers cannot easily be sustained.

One example of concentrated ownership – family firms – illustrates how it undermines collective action. Concentrated ownership in family firms lowers the likelihood of collective action because, in order to maintain their competitive edge and prevent the erosion of their wealth, they operate in ways that are less socially responsible than nonfamily firms (Chrisman, Steier, and Chua Reference Chrisman, Steier and Chua2006). The owners of old wealth and corporate assets are the most likely funders of innovation, but they prefer to withhold investments in anything that threatens the status quo. In fact, economies with more family-owned assets spend less on research and development (R&D) and file fewer patents. Family firms also spend less on R&D than comparable nonfamily-owned firms (Morck and Yeung Reference Morck and Yeung2004). Concentrated ownership supports cooperation between a narrow elite of managers, capital and/or oligarchic families, and political elites who are likely to support the status quo or their own narrow interests at the expense of the broader political economy (Morck and Yeung Reference Morck and Yeung2004). In fact, higher levels of family control are associated with inferior quality in terms of bureaucracy, higher barriers to entry (making the entry of innovators more difficult), and more extensive regulatory burdens (Fogel Reference Fogel2006). Although it is possible that family firms are just more adept at dealing with such obstacles and do not push for institutional reforms, it seems more likely, based on agency theory and incumbency, that these firms are using their power to affect government policy and block competition (Fogel Reference Fogel2006). For example, financial markets in economies dominated by family firms are likely to be intentionally less advanced because corporate elites favor a weakened financial system to maintain the position of already powerful firms (Rajan and Zingales Reference Rajan and Zingales2003). Further, when ownership is concentrated, politicians tend to emerge from families that control the largest firms in an attempt to reinforce economic power with political power (Faccio Reference Faccio2006). Concentrated ownership thus bolsters the influence of a small and established elite seeking to reinforce its status through both the economic and political arenas.

In contrast, Morck and Yeung argue that dispersed ownership is related to a broad and dispersed elite. Dispersed elites have incentives to put their capital to the best use possible and seek the highest returns available by investing in innovative practices. Countries with this structure sustain high rates of growth because assets are used in a more efficient manner and are directed toward innovation (Morck and Yeung Reference Morck and Yeung2004). Because dispersed owners obtain economic benefits from innovation instead of by exercising pressure to obtain rents, defending the status quo, and creating barriers for new entrants in the marketplace, they are more likely to support political projects that improve market-supporting institutions.

These observations support the view that, when agency problems are high or asset diversity is low, firms will tend toward industry-specific demands but have a difficult time pursuing them because of the small space within which they are able to define goals. This dynamic may be reversed at the highest levels of ownership concentration, however. It is possible, if the principal shareholders are sufficiently large and sufficiently few in number, that they may be able to coordinate despite the presence of agency problems and regardless of the level of asset diversity. In other words, ownership concentration in such cases may trump asset diversity. Apart from these extremes, however, higher asset diversity and lower levels of the agency problem (i.e. lower ownership concentration) will raise the likelihood of coordination among firms.

The next section applies these arguments to empirical data. Scholarship that attempts to understand how economic structure affects politics has focused on ownership concentration (mean shareholder size) or the role of banks in ownership (because banks tend to hold smaller shares on average). Below, I use these two types of data as proxies for the level of agency problems (through ownership concentration) and the level of asset diversity (by considering the role of financial firms and public ownership, because both are more likely than other types of investors to spread their investments across sectors in order to mitigate risk).

Firms in central and eastern Europe

The remaining sections focus on patterns of business collective action in post-communist Europe. The argument thus far has been that the structure of networks between firms, which depends on the size of the average shareholder (ownership concentration) and the extent to which common owners link different sectors, shapes the window of opportunity for business to act collectively. In the next two sections I present two kinds of evidence. First, I use comparative data for Bulgaria, Romania, and Poland to show that decreasing ownership concentration and rising asset diversity are associated with an increase in business collective action. Second, I conduct case studies of business organizations in these countries to show how collective action organizations developed in Poland around broad and diverse coalitions of businesses. By contrast, Romanian organizations tended to be personal political vehicles. In Bulgaria, organizations were largely irrelevant and unable to organize business as a group.

Figure 2.2 compares the role of financial firms in Bulgaria, Romania, and Poland in 2005. The level of ownership by financial firms varies even across the more advanced economies; for example, banks are much less present as owners in Hungary than in Poland. This situation developed because banks and industrial firms were promoted into different roles by the policies of privatization pursued in each country. Thus, the absence of banks in Bulgaria and Romania is not the result of poor economic development or financial failures so much as the product of a conscious policy of promoting industrial firms as owners. In Poland, banks came to be owners of other firms not because they were flush with cash but because the state pursued a strategy of swapping debt for the equity of indebted firms and concentrating the debt in banks while recapitalizing firms. The banks were later partly, and then fully, privatized. This was a very different strategy from that pursued even by Poland’s neighbors. In the Czech Republic, for example, firms were privatized via a voucher system, which only later generated a concentration of ownership as shares came to be publicly traded and gradually landed in the hands of larger owners.

Figure 2.2 Bank and industrial ownership across east central Europe, 2005

Note: All shares over 5 percent.

Source: EBRD (2005b).

Table 2.4 shows the level of ownership concentration and asset diversity (from Table 2.1 and Figure 2.2, respectively) plotted against each other. To remind the reader, higher levels of ownership concentration are expected to reduce business collective action. Asset diversity is expected to have a positive effect on business collective action because, when the owners are financial firms, they are likely to also own other, dissimilar assets.

Table 2.4 Ownership concentration and asset diversity

In fact, Poland has high levels of asset diversity and low levels of ownership concentration. Romania and Bulgaria, by contrast, have high levels of ownership concentration and low asset diversity. The effect of these two factors leads to sparse networks for Romania and Bulgaria and broad networks for Poland. While Romania and Bulgaria share sparse networks, their institutional development differs as a result of differing levels of uncertainty, as will be examined in Part III of this book.

Business collective action

How do these features affect the emergence of collective action among firms in practice? There are numerous challenges to assessing firm collective action in emerging markets. First, official organizations are often status organizations with little impact. In post-communist countries, many firms continued to be members of such organizations simply because they had always been members. Secondly, much firm collective action takes place informally. For example, numerous informal business councils were created in Bulgaria for the purpose of coordinating lobbying and business alliances. There are, however, ways to try to understand the extent of business collective action using survey data on firms. One available avenue is to examine perceptions of owners and chief executives regarding the extent to which collective firm associations matter and are able to impact politics. Firm CEOs are at the front line of interactions with the state and are consequently one of the best sources of information about practices prevalent among firms.

I begin with membership in business associations – a poor indicator on its own, because firms can be members of such organizations for the purpose of status, in order to have access to information, or simply because the marginal cost of membership does not justify withdrawal. Nevertheless, Figure 2.3 shows that at least 40 percent of all medium- and large-sized firms are members of such organizations. On the basis of this evidence alone, we can conclude that a significant number of firms pay membership dues in the belief that membership might help and probably does not carry any negative consequences. As expected, membership is highest in Poland, where more than twice as many firms are members as are not members. In Romania, the majority of firms are not members, while slightly more Bulgarian firms are members than not. This is in line with the outcomes expected as a result of ownership concentration and the diversity of assets.

Figure 2.3 Membership of business organizations

Source: EBRD (2005b).

BEEPS (EBRD 2005b) makes it possible to also assess firms’ views of the value of business association or chamber of commerce membership and the impact that firms can have as a group, which is the outcome of interest here. Figure 2.4 shows how medium- and large-sized firms assess the value of business associations in resolving disputes with other firms, workers, and officials.

Figure 2.4 Perceived value of business associations in resolving disputes with other firms, workers, and officials

Note: Means of ordinal scale responses, higher values indicating greater impact.

Source: EBRD (2005b).

As expected, associations are seen to have the greatest relevance in the country with low ownership concentration and high asset diversity: Poland. They have the least impact in Bulgaria, the country with the lowest level of asset diversity. The level of influence is also significantly lower in Romania than in Poland. Further, because only a minority of Romanian firms are members of associations, a much narrower grouping answered this question than in either of the other two countries. Thus, Polish associations are much more likely to promote cooperation among firms than associations in either of the other two countries. According to the same question on the EBRD survey (EBRD 2005b), this holds for the role of such associations in performing other functions, such as distributing information about regulations: Polish firms find the most value, while Bulgarian firms detect the least value.

BEEPS (EBRD 2005b) asks a further question about the value of associations in lobbying. Polish firms find such associations less valuable when it comes to direct lobbying than do Bulgarian firms (see Figure 2.5). This result is also consistent with the argument, as it reflects the perception that associations do not serve the direct interests of firms. Further, because so many Polish firms are members of business associations (see Figure 2.3), when associations lobby they do so on behalf of a broad group. The opposite is true for Bulgarian firms. While they generally see little value in such associations along dimensions that foster group collective action, Bulgarian firms find them more valuable for lobbying. This lobbying takes place on behalf of a narrower set of firms, however. As we shall see in the case study of Bulgarian business associations, this is to be expected in a country where narrow groupings of elite firms have created their own powerful associations to pursue specific goals. In Romania, few firms are members, and they receive relatively little in the way of representation.

Figure 2.5 Value of business associations in lobbying government

Note: Means of ordinal scale responses, higher values indicating greater impact.

Source: EBRD (2005b).

Institutional factors

Any discussion of collective action needs to also address the institutional context and how it may promote collective action. Do the countries that achieve high levels of business collective action benefit from institutional structures that promote it? Some institutions designed precisely to represent business as a collective actor are present in central eastern Europe and the Balkans. The foremost examples across the post-communist area are collective bargaining arrangements, which bring together employers and employees.

Building on legacies of worker representation in socialist factories, tripartite arrangements linking employers, unions, and the state were developed in most post-communist countries. These were seen as a way to emulate western European arrangements and thus to establish what was understood to be an institutional precondition for acceptance into the community of European welfare states. Although worker–employer negotiations were commonplace, countries varied in the extent to which these negotiations were formalized. For example, in Poland, the law establishing the tripartite commission was passed in Reference Ikstens, Smilov and Walecki2001.

Despite their widespread presence, the general consensus is that such institutions have little impact. With the exception of Slovenia, where collective bargaining covers over 90 percent of all employees, institutional arrangements linking unions, the state, and employers have declined in importance over time and do not cover enough firms to make collective bargaining a meaningful forum within which business interests are brought to bear on government. Crowley (Reference Crowley2004) surveys east European countries and finds that only 44 percent of employees on average are covered by such agreements (this drops to 33 percent if the outlier, Slovenia, is excluded). The form that these arrangements take further reduces the credibility of joint employer–state–union action. Across the region, employer–worker negotiations often do not take place in a centralized fashion, even when they cover a significant number of workers. For example, in Hungary a relatively low coverage of 51 percent is rendered even less meaningful because 80 percent of such agreements are made at the individual firm level instead of the sectoral or central level (Crowley Reference Crowley2004).

Thus, tripartite arrangements are unlikely to serve as a platform on which business collective action can develop. Moreover, with the exception of Slovenia, there is scant evidence that variation in collective action outcomes is determined by the extent of corporatist arrangements in the post-communist context. In that country, high levels of collective action are probably the result of formal tripartite structures. Enthusiasm about the importance of such arrangements in other countries that emerged in the early 1990s has since faded. In the majority of cases, corporatist institutions in post-communist countries have been characterized as weak (Rutland Reference Rutland1993; Ost Reference Ost2000; Crowley Reference Crowley2004), and essentially a “political shell for a neo-liberal economic strategy” (Pollert Reference Pollert1999, quoted by Crowley Reference Crowley2004: 409).

Another approach to exploring firms’ collective action would be to relate it to privatization strategy. Yet the loose relationship between early privatization outcomes and subsequent ownership structures is itself a testament to the fact that post-communist trajectories create legacies, but that these legacies do not prevent countries from changing the path on which they initially embarked. Said another way, the development of ownership subsequent to and often independent of privatization outcomes is a sign that these countries have entered a phase of post-post-communist development. Hence, in order to understand the forms of capitalism that are developing there, we must look well beyond the initial trajectories of privatization. Take, for example, the case of the Czech Republic, where voucher privatization was used to distribute shares of state-owned enterprises to the public. Within a few years these shares had moved their way into concentrated holdings, with the result that the Czech Republic has ened up with the highest average shareholder size in the region. This is not to say that the legacies of early privatization and related policies do not continue to affect ownership structure. The widespread use of debt-for-equity swaps in Poland, for example, made banks into important shareholders. Subsequent exchanges of shares in all countries altered the initial paths, however, in ways that were often discontinuous with the policy choices of the early 1990s.

Three cases: Poland, Romania, and Bulgaria

The remainder of this chapter focuses on a discussion of the role of business associations and gives qualitative confirmation of the patterns seen in the data above. It also establishes a link between the preceding analysis and the following chapters, which focus on detailed within-country data, process tracing, and interview-based research conducted in Poland, Romania, and Bulgaria. The subsequent pages show how the organizational sphere that represents business developed differently in each country on the basis of the characteristics of the prominent actors. This discussion seeks to highlight the characteristics of the organizational landscape in each country, not to explain why the countries followed such different paths, beyond the brief discussion that may be useful to the reader here.

In Poland, the variety of firms that were bound together by common interests led to the development of broad organizations that were consequently able to assemble significant lobbying power. By contrast, in Romania, firm owners joined organizations that were led by business leaders who already had political power. These organizations were largely personal vehicles for their founders and reflected a tendency toward the formation of narrow clique groups. They were able to extract benefits and exert influence, but this ability depended heavily on the personalities of group leaders and their ties to state actors and political parties. Finally, in Bulgaria, the tendency toward self-interest undermined projects of building collective organizations and left a barren associational landscape.

Each of the major Polish business organizations united firms across sectors. Common owners also often joined firms across sectors. Ownership stakes tended to be lower than in other countries, reflecting a tendency among firms to take stakes in other firms. This was reinforced by the prominence of financial firms in the economy, which acquired stakes across sectors because of the common strategy of swapping debt for equity and concentrating this equity in what were initially state-owned banks that were then partly privatized. As a result, the organizations that became powerful in post-socialist Poland were broad rather than sectoral.

The major Polish organizations included the following.

1. The Polish Business Roundtable (Polska Rada Biznesu: PRB), an elite grouping of the owners and CEOs of the most prosperous private and foreign firms present in Poland, created in 1992. Its presidents have been the owners and CEOs of the largest Polish firms. It was formed to represent business interests and sits on the tripartite commission. Headquartered in a lavish villa in central Warsaw, it also acts as a social club for those holding vast wealth.

2. The Polish Confederation of Private Employers Leviathan (Polskiej Konfederacji Pracodawców Prywatnych Lewiatan: PKPP), an organization of private employers that was established in 1998 to represent the interest of the private economy. Henryka Bochniarz, a strongly pro-market figure who served as minister of industry in the government of Jan Bielecki, has led the PKPP. In Reference Ikstens, Smilov and Walecki2001 it joined the newly formed tripartite commission. It is also the only Polish member of BUSINESSEUROPE, the main business association at the EU level.

3. Employers of Poland (Pracodawcy Rzeczypospolitej Polskiej: Pracodawcy RP; before 2010 the Konfederacja Pracodawcow Polskich: KPP), an association, created in 1989, that represents mostly state-owned firms and was seen by members of Leviathan as an organization that emerged from the planned economy and exists to protect the state sector. The Pracodawcy RP is also represented on the tripartite commission. It is led by Andrzej Malinowski, a peasant party activist before 1989 and MP for the Polish Peasant Party (PSL).

4. The Business Centre Club, which represented the interests of a much larger number of medium-sized firms, as well as some large firms. It was established in 1991. It was created and led by Marek Goliszewksi, a journalist before 1989 and one of the early initiators of the negotiations that led to the roundtable talks.

Although these organizations are broad and diverse, cleavages still exist between groups. They developed around the principle cleavages in the early post-socialist economy: the emergent division between state and private industry, as well as an early division between managers of state-owned enterprises and private entrepreneurs. In other words, organizations tended to unite firms around a broader political struggle. Exceptions to the rule existed, but Polish business organizations tended to have their own political identities related to the post-communist/anti-communist cleavage.

By contrast, the Romanian business organization scene is much less developed. The role of business interest groups and associations in Romania is insignificant. By their own admission, business associations find it very difficult to even collect dues from members, not because member firms lack funds but because they simply do not see the value of group representation. Hence, it is nearly impossible to convince members that it is important for the business community to lobby the state in unison, even for issues on which there is no conflict of interest between firms. According to directors of all influential business organizations, wealthy firms refuse to contribute funds to joint lobbying campaigns or to fund political campaigns by the associations. The leader of one organization lamented the failure of multiple attempts to organize, commenting to me that “there were several attempts to form a consolidated group of firms but they failed.”

A critical question is whether this is in response to a perception that such a business organization cannot be effective because of Romania’s political structure. Or do firms simply prefer direct exchanges? When probed, the directors of the business organizations all attribute the situation to the preference of business to engage in direct exchanges of campaign funds for personal political goods. In an interview, the head of one organization attributed the difficulty of collectively organizing firms to the outlook of firm owners: “Owners don’t have the right formation to address their problems collectively and continue to use methods from before 1989; in other words, they work through political channels to address their problems. These owners have relationships directly with political parties, not through associations.”

The business organizations that do exist in Romania are symbols of failed attempts to build effective representative organizations. They are often “prestige groups,” or personal showcases for single entrepreneurs who are trying to further their political careers. Their other members are firms that would like to benefit from group representation but have not succeeded in causing such an organization to emerge.

The main organizations are as follows.

1. The Businessman’s Association of Romania (Asociatia Oamenilor de Afaceri din Romania: AOAR), which is largely a personal vehicle for a major businessman who allegedly managed Romanian firms and hard currency funds in Cyprus before 1989. Whereas many members are clearly committed to what they see as the only possible future for Romanian business, the organization is largely a status symbol for Dan Voiculescu’s political and business ambitions.

2. The General Union of Romanian Industrialists (Uniunea Generală a Industriaşilor din România: UGIR), which was largely associated with the Paunescu family and served as a personal status vehicle.

3. The General Union of Romanian Industrialists 1903 (Uniunea Generală a Industriaşilor din România 1903: UGIR-1903), a rival organization to UGIR that established itself as the legitimate heir to the original organization after a lengthy court battle.

These bodies are very different in character from those present in Poland. Businesspeople such as the Paunescus, who were closely connected to the Romanian Social Democratic Party, did not need formal organizations for any purpose other than the status attached to the organization itself. This status was considerable; a business organization called UGIR had been created in 1903 and operated to represent the interests of pre-war industrialists. This group was of particular interest because the Paunescus and many of their peers had allegedly emerged from Romania’s security apparatus and sought to deflect calls for investigation of their dealings by assuming the outward appearance of conventional businesspeople. For interest representation, however, they opted for a one-to-one strategy by using their links to the PSDR, which held power without interruption until 1996 and regained power in 2000 from an ineffectual opposition coalition. Interviewees thus repeatedly argued that the Paunescus do not need such an organization for lobbying purposes.

The desire for status organizations led to a bizarre conflict, however, over who the rightful heir was to the original UGIR. Multiple groups of businesspeople sought control of the name of a new organization that was to resume the activities of a powerful pre-war organization. The Paunescus and their allies eventually decided on a new name, UGIR-1903, to emphasize their claim as the continuation of the original UGIR. UGIR-1903 became less active after 1996 because the Paunescu family relocated to the United States in order to avoid prosecution once the Romanian democratic opposition came to power.

AOAR had a similar link to Voiculescu, one of the wealthiest new businesspeople of the 1990s. His dealings were the subject of much discussion because of this alleged links to businesses and accounts belonging to the Ceausescu regime abroad. Again, it became a vehicle through which a coalition of businesspeople sought to acquire a legitimate image.

The two organizations that undertook most lobbying on behalf of business, organize executive and entrepreneurial education, and provide a location within which to network were the alternative UGIR and the Romanian Chamber of Commerce (CCIR). These organizations have had very little actual impact on policy, however.

Overall, Romanian organizations formed around and worked to further the interests of particular powerful business groups, and thus they sustained a certain level of cooperation among their members, who probably sought to attach their fortunes to the individuals and firms that anchored each particular organization.

By contrast, the Bulgarian organizational scene was a barren landscape. Three large organizations were divided according to the historical identities of the firms they represent. These were as follows.

1. The Bulgarian Industrial Association (Българска стопанска камара), which was established in 1980 and retains an organizational structure that recalls socialist industrial organization. It continues to represent industrial “branches” and coordinates the organization of branch associations. Moreover, individuals who rose through the socialist industrial hierarchy staff it and sit on its management board. Since 1990 most of those who retain a management role have served as directors of state-owned enterprises or formed firms that have strong ties to the state sector.

2. The Bulgarian Industrial Capital Association (Асоциацията на индустриалния капитал в България), which was created in 1996 as a union of privatization funds and brings together high-profile managers from the state sector.

3. The Bulgarian Investors Business Association (BIBA), which was formed in 1992 to represent the interests of foreign investors in Bulgaria against some of the early attempts to keep foreign direct investment (FDI) at a disadvantage.

These groups have negligible influence, however, and were not able to make progress in formalizing the process of affecting the policy-making process. Describing the general approach to lobbying, even the leader of one association said in an interview, “Foreigners come here and tell us that we should paint in these colors. But we have our own colors. Why should a lobbying law be an important institution for us?” Another business organization leader added, “Business has no interest in investing in the main organizations.” Recounting how the business community was divided, a business owner said, “In Bulgaria, the different organizations that were created could not carry through their vision because each organization was working for its own survival and met with lots of opposition.” Summing up the actual interaction between parties and firms, one politician said simply, “The moral of the story is that they all interact behind the scenes.”

Not surprisingly, several additional, informal groups were formed in Bulgaria, whose members carried somewhat more influence than the above-mentioned organizations. These were the groups that formed around business clusters. The first of these was known as the “Group of Thirteen” (G-13), which emerged as a lobbying vehicle in 1993 and dominated the private sector until 1995 (Synovitz Reference Synovitz1996). This group, including subsidiaries, consisted of the owners of the largest banks, insurance companies, stock exchanges, large trading companies, newspapers, and private security firms. The group was formed with the intention of protecting Bulgaria from the entrance of foreign firms. Member firms aimed to work together in order to obtain preferences, licenses, and other preferential benefits from the state. The group even created its own business organization, the Confederation of Bulgarian Industrialists. Divisions began to appear among the members quite early, however, as individual firms had differing interests. Some of these divisions were driven by the narrow interests of business owners, most of whom focused on various primary activities ranging from telecommunications to banking. Some realized that they could benefit from doing business with foreign firms, while others benefited from protection, which clashed with the stated purpose of the organization. Even the identities of the individual business groups that made up the G-13 was grounds for disagreement beyond single policy areas; this disagreement reflected the inability of firms to elaborate preferences as a group. Thus, as firms tried to diversify their holdings, individuals also came into conflict around privatization deals. The one-to-one manner in which preferences were awarded to firms in such deals meant that the organization could not offset the divisions arising from them. Faced with conflicts, firms fell back on a dyadic logic.

BIBA also accused the group of supporting economically nationalist policies that might have served the interests of domestic private business but were in conflict with those of foreign investors. Those firms interested in joint ventures with foreign firms sought closer ties to BIBA, making the institutional landscape unstable. Finally, the firms of the G-13 began to lose their political protection as economic pressure on Bulgaria mounted.

As the Confederation of Bulgarian Industrialists crumbled and many of its member firms – the early winners of the transformation – failed or faced significant economic difficulties, it was replaced by other groups with new political protectors. These shifts in business coalitions are discussed in more detail in the following chapters. Their significance here is to underline the unstable nature of the organizational landscape in Bulgaria. In contrast to the Romanian organizational landscape, these organizations were undermined by the excessively individualistic nature of negotiation and interest expression. In other words, interests were conceptualized in such an individualistic fashion that they precluded the successful expression of joint aims.

The striking difference between these three countries is that state–business relations were conducted on fundamentally different bases. In Romania, individual leaders were able to form personal organizations that attained a measure of stability, accounting for a higher level of impact among business organizations. In sharp contrast, the Bulgarian business leaders were unable to make any significant progress in creating organizations (with the exception of BIBA, which owes its prominence to support by foreign investors). Instead, individual business owners and CEOs used their influence over political actors to control both the content of policy and individual decisions. Although attempts to form groups were made, they were doomed from the outset, undermined by the strong dyadic alternative available to those seeking influence in politics.

Conclusion

These three cases complement the survey data presented earlier to make the broader point that horizontally networked and heterogeneous groups are more likely to have success in articulating group goals than narrow groups or groups wherein individualistic ties interfere with the elaboration of group objectives. There is a middle range in which narrow groups can be held together by leadership, and the Romanian case is precisely a case of the latter – individuals who exerted enough authority were able to create lasting groups – but these organizations depend on individual personalities for their cohesion, and, although they are stable, their performance is very modest. Poland, by contrast, is a case in which heterogeneity and networks in a period of uncertainty and political conflict raised the incentives for firms to collaborate. Bulgaria represents the other extreme, where the lack of network ties among firms and a strong tendency toward dyadic relations undermined efforts to elaborate group goals.

This chapter has argued that there is a link between ownership structures and the development of firm collective action, drawing on a literature that focuses on the macro-effects of corporate governance. I have shown how broad networks raised the likelihood of collective action. More precisely, two dimensions of networks – (1) the diversity of assets and (2) agency problems in the management of the firm – predict when firms have an incentive and the ability to act collectively. Data on the average size of shareholders and the prominence of firms that tend to hold heterogeneous assets, together with survey data on membership in business organizations and the value that firms obtain from those organizations, illustrated the effect of these two dimensions on business collective action.

Ownership networks as maps of politics–business relations

Ownership matters because modern states derive a considerable portion of their power from the economic assets that they control directly and indirectly, as well as the assets that they regulate. States build this power by coordinating and promoting sectors and particular groups of firms. In other words, the configuration of ownership ties among firms is a key component of the politics of economic development.

In making this argument, I follow a tradition that has recognized the impact of the configuration of accumulated assets on state capacity and regime type (Moore Reference Moore1967). For Moore, where ownership lies, how it is configured, and the participation of landed and peasant classes explain the transformation of agrarian societies into either democratic or authoritarian (fascist or communist) ones. Moore’s focus is on class conflict, but similar arguments have been used to explain the development of institutions without such a focus on class. The point is that the configuration of assets structures conflict between classes and determines the trajectory of political development.

Carruthers (Reference Carruthers1996) makes a similar point: that elite cooperation on state-building projects is often based on exchanges between elite groups. Thus, as political elites are trying to resolve state-building dilemmas by soliciting support from powerful societal interests, the latter are also seeking to resolve their own institutional dilemmas. These projects can be complementary if the different parties involved are able to make credible commitments, although that itself is a complicated problem of political innovation. Carruthers details how the Crown in seventeenth-century Britain essentially invented the possibility of state borrowing from the public in order to push forward with the project of centralizing and expanding state power. The fundamental puzzle that the Crown overcame was to find a way that the sovereign could credibly commit to the repayment of loans obtained from its subjects. Economic linkages not only solved a financial problem but also served as the glue for a broad exchange of loyalty between different social groups. This resolved a dilemma for both parties by unleashing stocks of capital for productive use in early bond markets and generating desperately needed resources for state projects. What economic actors saw as a struggle to shape the rules of the economic game, the state and its political leaders viewed as the search for extractive capacity and control (Carruthers Reference Carruthers1996).

The British Crown’s success in developing financial markets was not just a result of collaborative bargains between economic and political elites. The period is interesting because the financial revolution was taking place just as political parties were created in the British parliament. Carruthers shows that the new parties, the Whigs and Tories, used the economy for political ends and imposed a political logic on the economy that shaped institutional outcomes such as the development of competitive and efficient capital markets.¹

Within these struggles, networks were deployed to offset uncertainty. Stark and Bruszt (Reference Stark and Bruszt1998), McDermott (Reference McDermott2007), and Stark and Vedres (Reference Stark and Vedres2012) have shown that early post-communist ownership reconfigurations were often geared toward achieving political ends, or attempts to recombine elements into new valuable property or to provide protection from external pressures. Network reconfigurations also served as an adjustment mechanism that helped firms cope with changing external conditions, because networks reorganize in reaction to external stimuli such as globalization or economic liberalization (Hamilton and Biggart Reference Hamilton and Biggart1988; Hall and Soskice Reference Hall and Soskice2001; Kogut and Walker Reference Kogut and Walker2001; Stark and Vedres Reference Stark and Vedres2006). These networks also provide support to firms in difficult times. In other words, networks – whether they are based on ownership, membership of boards of directors, or even friendship – are strategic assets, and ties are deployed as a part of the profit-making efforts of individual firms. Whether banks are allowed to merge with each other, and which firms are allowed or encouraged to acquire stakes in other firms, are similarly political decisions.

This chapter observes how ties of ownership between companies took form between 1990 and 2005. Decisions to configure or reconfigure the economy in a particular way and create joint links – to create ownership ties between two firms – are all reflections of the political organization of the economy (Mizruchi Reference Mizruchi1992; Davis and Mizruchi Reference Davis and Mizruchi1999; Uzzi Reference Uzzi1999). As decisions over economic policy were made in the early 1990s, banks, foreign investors, single large private owners, investment funds, holding companies, or large industrial firms began to emerge as more or less prominent stakeholders. These actors acquired different roles in each context. These early decisions, in turn, generated pressure for certain policies that favored the dominant type of stakeholder and thus reverberated through the process of institution building.

The ownership structure is one way of assessing the configuration of interests and their potential for constructive interaction (concertation). Economic transactions took place against the background of a political contest that instrumentalized firms for the sake of gaining an edge in politics. This dynamic also reproduced itself within the market, however, where competitive pressures rose in accordance with the needs of political actors. Firms also were pushed to perform in order to generate profits that were partly used to buy much-needed political capital. Said another way, when political competition was sharp, firms went from being the spoils to generating the spoils for political actors. Market institutions thus served and were shaped by political struggles. In some cases, as in Poland, these conflicts were sufficiently deadlocked to push elites toward the development of more stable rules of interaction. When elites were less balanced, as in the other two cases examined here, institutional outcomes were significantly poorer.

Ownership development in Poland, Romania, and Bulgaria

To understand the development of different state–business relationships, two types of data are considered in addition to a historical analysis of business networks: (1) data on the top owners and the number of stakes they hold in the top 200 firms (ranked by revenue); and (2) data on the characteristics of the networks among the top 200 firms themselves.

The following questions are posed with regard to the data. What types of owners have emerged in each country? How did privatization reshape firm networks? What subsequent changes restructured those networks? Are the networks between firms broad or deep? Are cross-holdings between firms common? Do they extend across business sectors? Do these networks include the state? What kinds of actors are the most central owners in the economy? Are institutional investors, banks, industrial groups, or family firms prominent?

Comparing data on firm ownership networks across countries shows the starkly different paths followed by each country in the political process of creating marketized economies.² This difference is based not only on the speed of privatization or the persistence of state ownership in particular sectors. In each country, different structural shifts took place across the economy, transferring ownership stakes from the state to different dominant types of actors. The following sections present data on the types of key owners emerging in each country. Table 3.1 shows the changes in the ownership structure of the 200 largest firms by revenue between 1995 and 2005 for each of the three largest categories (by number of ties) in any given year and country. It is compared to the status of the same owner type in the other countries to show how sharply different paths were taken. Examining the ties between the 200 largest firms by revenue and their owners captures the control structure of the powerhouses of economic activity in each country and effectively delivers a picture of the choices about restructuring that each government took. These paths were not the reflection of an unintended choice but largely conscious decisions on the part of policy makers and their allies in the economy. They reflect the dominant alliance between political and economic stakeholders.³

Table 3.1 Top owners by type and their ties, 1995, 2000, 2005

Note: Percentage of ties held by each type of owner.

As the table shows, state ownership – unsurprisingly – declined between 1995 and 2005 as a result of privatization but remained a prominent feature in each country until 2000. Partial privatization was a common strategy in the transition economies and was driven by the desire to retain some political control on the part of the state, inadequate capital on the part of new investors, and the division of risk between private investors and the state (Maw Reference Maw2002). The gradual pace of privatization in transition economies was an avenue for states to signal commitment, a logical way of showing that the government was willing to retain a share of the residual risk – that is, “a signal that it does not intend to redistribute value through a future shift in policy” (Perotti Reference Perotti1995). This is particularly important to potential private stakeholders in periods of future policy uncertainty that have the potential to bring about reversals that would reduce the future value of the privatized firm. Such uncertainty could come from the possibility that the party in power could change or simply because an insufficient amount of time has passed to allow investors to gauge the government’s longer-term intentions. All these factors were present in the transition economies, and thus it is not surprising that partial privatization persisted for such a long time.

Although privatization as a broad goal was common to all transition countries, they chose different paths in addressing uncertainty. Comparing the three countries shown in Table 3.1, it is quite apparent that privatization strategies and subsequent policy decisions privileged some potential owners over others, and different types of owners emerged as the largest stakeholders in each country. As early as 1995 state-owned banks were being pushed forward as stakeholders in the Polish network of large firms. By 2000 Poland’s ownership structure was heavily influenced by foreign banks and investment funds. Strikingly, four large banks linked the state to foreign capital, while one linked the state with private domestic capital. By 2005 foreign firms were heavily in control of assets in the Polish economy. This injection of foreign capital was, at least in part, a fruit of the early strategy of placing banks in control of industrial firms as part of the process of restructuring.

Throughout the Polish transformation, financial firms were key owners. The data show that the politics of industrial restructuring privileged the shifting of ownership to state-owned banks, even when these were partially privatized to foreigners, and resisted the transfer of industrial firms to other industrial firms. Later sections of this chapter show that banks held ownership ties both to other banks and to industrial firms. The network position of banks thus made them the decision-making centers for the management of the economy and rendered them influential in the operative decisions of a wide range of nonfinancial firms. Comparison with Bulgaria and Romania shows that this financialization of the economy was a quite peculiar characteristic of the Polish approach and distinguished it from other countries, which chose mass privatization, employee participation, or the creation of industrial groups by sector as their preferred strategies. Instead, the Polish path postponed the transfer of ownership rights in a significant way even to industrial firms until quite late – between 2000 and 2005. Only in 2005 did industrial firms appear as significant stakeholders there.

By contrast, in Romania, management and employee buyouts (MEBOs) and mass privatization funds were the key emergent ownership form until 2000, when industrial companies displaced them. In place of Polish banks, Romanian private owners and family groups, representing the very prominent new economic elite, ranked highly but were not in the top three. Foreign financial firms were present in 2000, but the lack of domestic financial firms meant that the overall presence of finance as an owner was much smaller. Instead, the overwhelming characteristic of Romanian ownership networks was that of direct individual or group ownership of assets. By 2005 private individual investors were the second largest owner type in Romania. Some of these were small stakeholders, but the majority held large stakes in the top firms.

A different strategy was pursued in Bulgaria – perhaps even one that can be defined as the opposite of the Polish strategy. Instead of financializing ownership by transferring debt in exchange for equity to state-owned banks, the Bulgarian state focused on giants in each sector to conduct restructuring. The hope was that agglomerating firms in the same sector under single holding companies would make the assets easier to restructure and more attractive for privatization down the line. Hence, in Bulgaria, state-owned firms actually increased their profile as owners, reflecting this transfer of ownership from the state to state-owned holding companies that took place during the first decade of transformation. The move effected an informal decentralization of power and loss of control of state-owned assets by putting large numbers of firms under the control of holding company managers instead of subjecting each firm to ministerial supervision. Consistent with this trend, management buyouts and mass privatization dominated, together with state ownership until a later period, when individual investors began to emerge.

Consider the contrast with the Polish case. In this, an alliance of financial capital with the state slowly transformed into a union of financial and industrial capital. In Bulgaria, a decentralization of ownership stakes to state-owned firms and employee privatizations transformed into an economy governed by individual stakeholders and industrial firms. Unsurprisingly, these groups had dramatically different interests and time horizons. Polish financial firms and the links they promoted between firms led the network structure of ties to be much more horizontal. As a result, firms were locked into a web of mutual interest with other firms by debt and equity holdings. These allowed information to flow and bound firms into ongoing relationships. They also generated a distributed interest in the performance of each firm. Bulgarian firms had no such ties, and individuals in charge of both state and private property had only their immediate self-interest in view in a highly uncertain political context that offered few incentives to invest in the long term.

To give more texture to this argument that owner types and their networks influence management style, the following sections discuss in more detail the development of the network of firms in each country.

Poland’s path to privatization: banks promote broad networks

In Poland, two features were distinctive in the first period of the transformation. First, banks were a critical part of the post-1989 capitalist development, emerging at the center of clusters of firms. Banks became involved in a series of debt/equity swaps that created broad networks of cross-ownership because of a government preference for workouts. Under the workout procedure, banks could negotiate workout agreements with problem debtors and force them on creditors. This put banks in a position to focus on long-term value and the development of a broad network of firms connected by ownership ties. Second, the facilitation of debt/equity swaps was part of a distinct policy that avoided or delayed the foreign purchase of many attractive firms. This policy delayed the transition of nearly a half of all ownership shares in financial firms until after 2000 and caused the majority of shares in industrial firms to remain in domestic hands. As a result, in the second stage, private entrepreneurs found it difficult to dominate national economic – and hence political – activity. The new Polish magnates were certainly important in Polish politics, but they were as much dependent on political parties as the parties were dependent on them. When asked why his firm had such extensive ties to parties and the state, one business leader commented, “Because the political sphere is so present – in ownership transformation, in regulation. It’s like in business: in the early phase, you look for a partner with whom you can do something together.” The level of political competition raised the stakes for both sets of elites. As a result, a broad horizontal network of ownership emerged in Poland that was distinct from the other countries under comparison.

The initial step in the distribution of ownership stakes was the process of privatization. The most common methods of privatization were: (1) the restitution of property to former owners (this applies only to property existing before nationalization); (2) direct sales of state property, either to domestic or foreign owners; (3) MEBOs; (4) free distribution through a voucher system; or (5) a combination of these strategies (Andreff Reference Andreff2005). Poland largely chose direct sales, with some voucher privatization to legitimate the process (Stark and Bruszt Reference Stark and Bruszt1998), the creation of national investment funds to inject capital (Błaszczyk et al. Reference Błaszczyk, Hashi, Radygin and Woodward2003), and some MEBOs (Svejnar Reference Svejnar2002).

Moreover, privatization proceeded slowly. Indeed, despite being hailed as an example of the success of neoliberal reform, Poland actually proceeded more slowly with regard to the overall reduction of the state sector than did countries such as the Czech Republic (Stark and Bruszt Reference Stark and Bruszt1998). It even lagged behind Hungary, often cited as an example of gradualism, as late as 2000. Progress was halting and largely marred by political infighting and the consciousness that privatized property would be valuable in future political contests. Privatization in Poland, as elsewhere, was also marked by political manipulation. Great lengths were taken by Polish state officials to create politicized spheres of property: firms were privatized to domestic business groups in exchange for future benefits in the form of campaign contributions to political parties. As a former minister of privatization, Janusz Lewandowski, said, “The government has frequently used words like ‘national’ or ‘Polish’ in consolidating state assets in the sugar, power or shipbuilding industries. In the end, however, ‘national’ often turns out to mean ‘partisan’” (Polish News Bulletin 2004). Businesspeople also spoke of the need to shield national entrepreneurs from foreign pressure – a discourse that resonated with broad nationalist sentiments already in the first decade of transition. For example, one business leader, asked to identify priorities for economic policy, stated, “The government needs a system of supporting business – a system of financing, a system of supporting Polish owners [emphasis added].”

Bank restructurings in particular were subject to politicization because of the emphasis on workouts rather than liquidation, as described by McDermott (Reference McDermott2007), and the rejection of bank sale by distributing vouchers to the general public (Balcerowicz and Bratkowski Reference Balcerowicz and Bratkowski2001). At the same time, large privatizations were used as a means of plugging holes in the budget and reducing liabilities for the state. Addressing budgetary shortfalls is a common element of most privatization processes, but the combination of workouts, domestic favoritism, gradual pace, and political competition created ample opportunities for the entrance of politics into the process in Poland.

The system of corporate governance further enforced these trends. Poland adopted a two-level system of corporate governance. This structure was chosen because supervisory boards would act as state agents, doing what was known in Poland as “sanitizing” (bringing to a healthy state) firms without political interference. In practice, this process was heavily politicized (Jarosz Reference Jarosz2001: 47), with the supervisory boards often staffed by political insiders loyal to the then minister of the Treasury (Grzeszak et al. Reference Grzeszak, Markiewicz, Dziadul, Wilczak, Urbanek, Mojkowski and Pokojska1999).

Finally, Poland’s early governments were reluctant to accept foreign ownership, to the point that one early privatization minister stated that he had created as many obstacles as possible to foreign purchases of Polish companies in order to encourage a class of domestic owners (Stark and Bruszt Reference Stark and Bruszt1998; Schoenman Reference Schoenman2005). This last position reinforced the politicization of Polish business.

Altogether, the Polish government’s policy of economic transformation amounted to an attempt to manage the slow combination of private market forces with state intervention in order to retain both political and policy influence. After 1997 economic pressures led governments to look increasingly to foreign investors for injections of capital, but these were still managed and cautious attempts to attract investment without relinquishing control over economic development policy. Throughout, these moves increased the breadth of the network of ownership.

Romania’s path to privatization: a redistribution of assets to political clients

The Romanian case offers a second path. Here, ownership networks were narrow, and uncertainty was low. Firms had no way to coordinate, as they did not have the broad links that Polish firms could use to share information and mobilize for cooperation. Early post-1989 governments in Romania delayed reform questions, opting in favor of a gradualist approach that was preferred by managers and in the interest of the new political class. Although they began to put the infrastructure of privatization in place by 1991 with the passage of the first mass privatization law, only 260 companies were privatized by 1993, and 92 percent of these were small firms (those with fewer than fifty employees). Only two of the 708 larger companies on a 1990 list of firms to be privatized were actually sold by the end of 1993 (Roper Reference Roper2000: 95). This was a very small number compared to the overall task of privatization. More importantly, the National Salvation Front – a broad coalition of caretakers dominated by transformed members of the Communist Party elite emerging from the second circle around the former dictator, Nicolae Ceausescu – was firmly in control of government.

Hence, privatization took on the form of redistribution to constituents and insiders. Highlighting the flaws of the process in an interview, one official concluded, “It was a fake process of wealth creation by spoiling state property.” The majority of firms that were privatized were actually transformed into joint-stock companies by distributing 70 percent of their capital to state ownership funds and the remaining 30 percent to the private ownership funds that were, formally, owned by the public (Roper Reference Roper2000: 91; Earle and Telegdy Reference Earle and Telegdy2002: 661; Grahovac Reference Grahovac2004: 28, 68). The public received certificates that could be used to purchase the tranche owned by the private ownership funds, thus linking private shareholders with state ownership. This is important in terms of the creation of the network of ownership, because it diluted ownership, granting public participation to a limited extent but without bringing about privatization in any meaningful way (Grahovac Reference Grahovac2004: 28). Further, according to Earle and Telegdy (Reference Earle and Telegdy2002: 661), the private ownership funds remained state-governed, with government-appointed directors who were approved by parliament, and effectively no way for the public to exercise control.

Moreover, all other state-owned enterprises (SOEs) were turned into so-called regies autonomes (RAs), which remained under state ownership and were overseen by the Ministry of Finance. The 450 companies that became RAs were utilities, natural monopolies, or other companies of sufficient interest or importance that they could be deemed “strategic.” The precise motives for placing companies into this group is unclear (Earle and Telegdy Reference Earle and Telegdy1998a). With 450 companies in this group, however, it seems likely that the strategic nature of some firms was largely a function of the number of jobs they provided or the income they generated for political interests. Thus, the strategic discourse became an excuse to shield companies from privatization until an unspecified later date. Within this context, the privatization program began under a dark cloud. The state ownership funds rarely offered shares in the best companies, and those that represented politically powerful constituencies were sheltered from privatization and allowed to continue operating under a deficit (Ahrend and Oliveira Martins Reference Ahrend and Oliveira Martins2003: 333).

By 1992, after numerous splits within the NSF that brought a faction known as the Party of Social Democracy of Romania to power, the government recognized that the economic situation required a reorientation in economic policy and enacted a strategy of price liberalization, wage stabilization, and austerity combined with increased export concessions. The hope of the government of Theodor Stolojan, who replaced Petre Roman as prime minister, was that industries would be able to increase their exports and reduce the large trade deficit, offsetting the difficulties associated with austerity measures. The companies that were encouraged to export did not produce goods that could be sold on the world market, however, and required increasing energy imports, which the government subsidized (Roper Reference Roper2000: 92; Ahrend and Oliveira Martins Reference Ahrend and Oliveira Martins2003: 337). The move was also meant to redirect trade from former Council for Mutual Economic Assistance countries toward the European Union. Although EU trade increased, imports from Europe also rose drastically. According to Daniel Daianu, former minister of the economy, the weak results of the election that brought Stolojan to power over factional rivals prevented the government from pursuing real export-led growth tactics (Roper Reference Roper2000: 93).

Reform continued but took on a stop-go character. In 1995 new measures were introduced to limit inflation, then rescinded after several months, largely as a result of complaints from political constituencies. Government subsidies to the industrial sector persisted. By 1995 only 25 percent of commercial companies identified in 1990, and only 8 percent of large SOEs, had been privatized (Economist Intelligence Unit [EIU] 2001).

Serious economic problems ensued, leading to the creation of another mass privatization program (MPP) in mid-1995. Vouchers were again distributed to the public, but it took until May 1996 for 93 percent of the vouchers to be exchanged for shares. Moreover, this method introduced no new capital into the industries, and none of the important RAs were part of the second MPP. This move also made the network even more disconnected (Earle and Telegdy Reference Earle and Telegdy1998b).

The victory of the opposition leader Victor Ciorbea, head of the Christian Democratic National Peasants’ Party, in 1996 – an isolated instance of protest voting against the poor economic policies of Stolojan – created the chance for a new era of Romanian economic reform. The government passed a set of reforms that gained the confidence of the International Monetary Fund (IMF) and secured a new loan package. The other members of the new Romanian Democratic Convention (CDR) coalition, however, the Union of Social Democrats (USD) and the Democratic Party (PD), opposed many of the reforms promised in the IMF agreements. For example, the USD and PD were closely allied with business interests that rejected the elimination of subsidies to industry (Roper Reference Roper2006). It would also have been difficult to interrupt the supply of credits to industry by banks, a sector that had not begun to privatize. The entrenched interests in the banks and ministries worked to prevent the government from making any progress on the restructuring of firms (Roper Reference Roper2000: 92).

In the end, the Ciorbea government was unable to maintain the pace of reform. Many SOEs were placed on the closure list and then never closed (Roper Reference Roper2000: 102). The same was true of companies slated for privatization. Disagreements continued between Ciorbea and Roman’s USD. The narrowly constructed economic networks allied with the opposition helped prevent any progress in the construction of a new economic framework during this brief window of opportunity. The failure of this potential new force in Romanian politics paved the way for an end to Ciorbea’s short and ineffective disruption of one-party dominance.

The conflicts that had existed during the Ciorbea government continued with the successor government of Radu Vasile, organized under the umbrella of a political alternative. Roman and his political allies from the previous era had little interest in promoting economic reform because of pressure from interest groups allied with the USD. The miners, a potent force that could be brought in to riot in Bucharest, also continued to play a role in Romanian politics. After several violent outbreaks, Vasile’s government came to an undisclosed agreement with the miners’ union, which allegedly postponed or revoked the closure of two mines and brought financial benefits to the miners. Even in such devastated sectors, reform seemed nearly impossible. Ultimately, Vasile’s government was short-lived. Ion Iliescu, the leader of the NSF/PDSR (now renamed the PSD), returned to power by capitalizing on the failures of the Ciorbea government – failures that were largely a consequence of PDSR obstruction.

Bulgaria’s path to privatization: the state struggles to control firms

In Bulgaria, ownership networks developed quite differently from how they did in Poland and Romania. First, because of political infighting, there was an inability to commit to a policy of privatization. In this context of political chaos, and because of it, groups of insiders became institutionalized as economic stakeholders and gained an inordinate amount of influence. These insiders even attempted to coordinate so as to prevent the entrance of foreign business. Although banks were politically prominent initially, the spectacular bank failures during the financial crisis of the late 1990s – itself caused by insider dealings – removed them as a viable organizing pole. The network of firms that developed was ultimately narrow in structure.

Privatization in Bulgaria took place under a mix of direct privatizations and voucher privatization. The latter, like most voucher programs, was intended to speed up the process and attract local participation (Prohaski Reference Prohaski1998). The Bulgarian state lost control of the process of economic transformation soon after Decree 56, passed in 1989, allowed for the limited creation of private firms and gave SOEs some autonomy. These were the basic preconditions for the emergence of a private sector that had parasitic relations with the now more autonomous SOEs, given the political context. Radical decentralization of the state banking sector, in combination with liberal licensing policies for private banks, assisted the emergence of large conglomerates of questionable origin.

Efforts at limiting these trends produced weak results. Attempts in 1991 by the then finance minister, Ivan Kostov, failed to reel in the more influential economic agents, who continued to enjoy access to credit. The Privatization Act was passed by the Bulgarian parliament in 1992, allowing privatization of state-owned enterprises to move forward. After February 1993, when the first SOE was privatized, privatization gradually gained momentum.

Proponents of privatization faced serious opposition from entrenched groups, such as SOE managers, government officials, and large banks. State officials, not surprisingly, were generally unwilling to give up their ability to reap personal rewards from the situation. Thus, until 1996 only small privatizations were successful, and practically no progress was made on the privatization of large industry. The situation of enlarged bad debt finally led to massive bank failures in 1996 and stricter conditionality from the IMF, to the point that it became the main factor in the collapse of Zhan Videnov’s socialist government in 1997. A currency board was subsequently put in place, effectively eliminating the possibility of central bank lending.

Bulgaria was also the first country in which the socialists were re-elected to power, in 1994, and it was not until 1997 that a full mandate was given to nonsocialists. The re-election of the Bulgarian Socialist Party (BSP) was taken as a U-turn away from the transition. This gives a sense of the political and ideological tension that emerged in Bulgaria after 1989. In addition, there were widespread allegations of clientele-favoring behavior against the center-right Union of Democratic Forces (UDF) (Stanchev Reference Stanchev1999). Even during the center-right UDF government of Kostov (1997–2001), often assessed as the first government committed to making difficult but unpopular policy choices associated with economic reform, preferential privatizations continued to take place (Stanchev Reference Stanchev1999). An official noted in an interview that this was the product of a broader vision and tolerance for political wealth creation: “There was a popular slogan at the beginning among UDF members: we want to create the blue [right-wing] bourgeoisie, not the red [left-wing] bourgeoisie. But when this happens only by corruption, it creates a corrupt bourgeoisie. It’s very much the official policy of the party and it was defended as appropriate because the Socialists also created their bourgeoisie.”

Throughout the transition period, Bulgarian politics has been marked by the strong influence of economic interests – a dynamic that did not fail to attract the ire of the electorate. Bulgarian frustration with the lack of state capacity and the influence of powerful figures in the economy created a window of opportunity for the return of the former tsar Simeon II, who had been in exile since 1946. Having left Bulgaria as a young boy and worked in finance in Spain, Simeon was seen as a political outsider who would, it was hoped, be able to bring together a group of young professionals with foreign work experience and few ties to the existing networks.

Even Simeon’s government had difficulty sidestepping existing social structures, however. According to political scientist Ognyan Minchev, speaking on Radio Free Europe/Radio Liberty on April 3, 2002, the government of the former tsar (2001–2005) and the new party he created for the Reference Ikstens, Smilov and Walecki2001 election, the National Movement Simeon II (SNM), faced the problem of legislators who did not “work in the interest of the state, but act[ed] as lobbyists for business interests or even on behalf of business groups linked to organized crime, which flourishes under a fragile, powerless government.” In an interview, an official observed, “Big business is happy under SNM because, even though they started as thieves, the new government does not interfere. We cannot get rid of the criminals.”

Networks

The strong point of network analysis is not just in identifying the most connected members of a group but also in considering the structural position of a particular member of the network in relation to other members. Although the early discussion in this chapter makes it clear that three very different alliances emerged to support marketization in Poland, Romania, and Bulgaria, the tools of network analysis can add a great deal of depth to this picture. The previous discussion of the development of state–business ties in each country shows the radically different forms of this relationship that can sustain the transition to markets. This background allows us to return to a macro-view of each of these variants.

The most striking feature in the transformation of ownership networks in Poland is consistent with the development of private firms discussed above. In 1995, despite the beginnings of privatization, Poland’s ownership network retained some of the features it had had in 1989. Most notably, the state still was the main owner in the economy. As a result of partial privatizations and the generally slow pace of change, more than 50 percent of the largest firms were still wholly or partly state-owned at this time. The resulting network had a star shape, created by firms owned mostly or wholly by the state, as seen in Figure 3.1. In other words, partial privatization accentuated the central role of the state by linking private firms to the state through jointly owned state and private businesses.

Figure 3.1 Firm networks in Poland, 1995

Note: Arrows point to owners.

By 2000 more of such shared ties with private owners had developed, as seen in Figure 3.2. This was the result of the privatization strategy discussed above, which used debt/equity swaps and created opportunities for private investors to purchase partial shares of state-owned firms. Such hybrid forms of property ownership – neither public nor private – limited the ability of policy makers to take autonomous decisions by aligning the economic interests of firms with the state. It thus added to the security of firms (Perotti Reference Perotti1995; Stark Reference Stark1996)

Figure 3.2 Firm networks in Poland, 2000

Note: Arrows point to owners.

A more significant shift occurred in these five years than the increase in number of such ties, however. In 1995 there were few connections linking the private actors among themselves, as the initially desirable move for new private firms was to create joint ventures with the state by taking ownership in state-owned firms that were being partially privatized. Between 1995 and 2000 the ownership position of the state in the network did not become much less significant, contrary to what one might expect given the prominence and international pressure for economic reform during that period. A new feature, however, was that the ownership network changed from one highly centralized around the state to one in which firms also held numerous ownership ties in other firms. These ties between firms also served to cement alliances between firms. Such swapping of ownership ties is a common practice in Japanese capitalism (Gerlach Reference Gerlach1992).

The sheer number of ties that resulted from this ownership transformation was a distinctive feature of the Polish economy. The number of ties maintained by an average firm when compared with other countries can be seen in Table 3.2. The table shows the mean degree – the number of network ties maintained by an average member – and a normalized value that can be compared across networks of different sizes.⁵

Table 3.2 Mean degree and normalized mean degree, 1995, 2000, 2005

In 1995 Poland did not have the highest level of connection among countries. Instead, the Bulgarian network of firms was the most connected, reflecting the persistently high level of state ownership. As a result of the slow pace of privatization, few firms had disconnected from the key owner, the state. Hence, the mean degree was high, by virtue of firms’ ownership tie to the state. Nevertheless, the Polish network was only slightly less connected.

Over time, the density of network ties dropped in both Bulgaria and Poland, as privatization took place, more firms entered the network as owners, and firms became disconnected from one another as they lost their tie to the state. A falling mean degree is to be expected as the network grows. This change happened much more slowly in Poland, however, as firms replaced their ties to the state with direct ties to other firms and indirect ties to the state through banks. Although the average Polish firm maintained a decreasing number of ties over time, the network of firms was much more connected through the first decade, and Poland remained the most connected of the three countries by 2005.

As the early parts of this chapter began to show, a distinguishing characteristic of the Polish case was not just that the network was more tightly connected throughout the first fifteen years of transformation but that certain types of firms held the most ties in the network. Which were the most connected firms in the Polish ownership network? Was a specific type of actor privileged in the Polish case, subsequently influencing the way that the Polish economy and institutions developed? These can be identified by looking at the most “central” firms at each point in time – those that are most connected by ownership ties and are the centers of larger groups of less connected actors. These firms were leading the broader reorganization of the economy and placing themselves at the center of a wide web of ownership relations, because their ownership rights granted control rights over many other firms. Thus, more connected firms are frequently able to impose their policy preferences in the governance of firms in which they hold a significant stake.

By calculating the degree of each firm, the number of connections it shares with other firms, one can identify these “most linked” firms, as shown in Table 3.3 for 2005. In 1995 the state was the largest owner, with direct connections to seventy-five other firms among the sample used here. Four foreign financial institutions (the EBRD as an international lender, ING, Barings, and Creditanstalt) and three foreign financial groups (two Finnish and one US investment fund) also appeared in the top twenty. The remaining top owners were nine Polish banks, one pension fund, an import-export company, and the electrical parts company Elektrim. Thus, a ranking of the top twenty most connected firms reflects the policy decision to push banks and financial firms forward as principal owners of other firms in the process of restructuring.

Table 3.3 Top twenty most connected firms, 2005⁶

In 2000, after a decade of reform, the same trend can be observed. Apart from the state, which then had 104 direct connections to other firms among the top 200, two foreign banks were present (Bank of Austria and Bank of New York) and the investment firm Franklin Resources. The remaining owners were fourteen Polish privately managed investment funds of state capital (NFIs), banks and financial services companies, and the oil and gas giant PKN Orlen.

That largely “bank groups” and a few “industrial groups” dominated the ranking of powerful firms by 1995 and in 2000 highlights a particular development process under way in Poland. In the Polish development strategy, groups that were organized primarily around domestic financial services companies came to occupy strategic positions within the network of successful firms. The owners in the process of acquiring strategic positions in the economy were largely Polish banks and, in the latter half of the 1990s, domestic investment funds. These banks and investment funds stood between the state and the long chains of connected firms and were the centers of groups of firms. In this way, they were able to channel capital to subsidiary firms for the initial period of reform. A second key part of the strategy, however, involved the integration of foreign capital. By 2005 the situation had changed dramatically, in that Polish firms had switched places with foreign firms among the key financial companies. The list of top twenty most connected firms in 2005 had only three Polish financial firms and sixteen foreign financial groups, showcasing a virtual “Who’s who?” of the financial world, headed by the likes of Unicredito, ING, and Credit Suisse. The state still dominated the list, however, with 121 connections. Comparison against the other two cases explored in Table 3.3 points to the distinctiveness of this strategy.

The comparison between the three countries, even just in 2005, displays fundamentally different types of entity as the most connected owners and shows how the top firms in Poland maintained dramatically more ties to other firms in absolute terms than firms in both Romania and Bulgaria (the top twenty firm owners in Poland averaged 40.55 connections, while in Romania they averaged 5.2 and in Bulgaria 4.7). As noted earlier, the most connected firms in Poland were almost exclusively foreign financial firms by 2005, while in both Bulgaria and Romania they remained mostly industrial firms.

Firm networks in Poland

When compared to Romania and Bulgaria, the novelty of the Polish strategy is noticeable. The decline of state ownership was slower in Poland than in the other two cases. In 1995 the state was still the largest owner by far, but already many partial privatizations were visible (see Figure 3.1).

This trend continued in the latter half of the 1990s, so that many firms were partially state-owned and partially privately owned. When compared to the other network graphs, the Polish network in 2000 (Figure 3.2) is distinct for the number of firms that were partially in the private sector and partially in the public sector. Even more striking is the presence of state-owned firms with shares in other firms. Thus, there are horizontal ties within both the state sector and the public sector. As discussed above, these ties often led to banks, which came to be the focal points of ownership in the economy.

Turning to Figure 3.3, we see that, by 2005, state–firm ownership networks were still extensive, particularly when compared to the other two countries, where state ownership was reduced dramatically by fiscal burdens and international pressure that drove those countries to speed up privatization. The trend that began early in Poland – the emergence of banks already in 1995 as key owners of other firms – was fully cemented by 2005. Although the state continued to play an important role, the state–bank alliance of the early 1990s was replaced by an alliance between industrial and financial firms, as seen in Table 3.1. While the majority of industrial owners were domestic, the vast majority of the financial capital by 2005 was foreign. As discussed above, however, a key component of the Polish path was to delay this move as political and economic elites both sought ways to retain a domestic presence in industry and finance alike and hold off foreign competition. The network graph here also shows that, despite its decline as a shareholder, the state was structurally still vital.

Figure 3.3 Firm networks in Poland, 2005

Note: Arrows point to owners.

Bank centrality – the quality of banks as firms with particularly many ties to other firms – can be seen more clearly by examining the network ties of a single bank (Figure 3.4). Bank Handlowy in 2000 was representative of the network ties maintained by banks in Poland. The bank was by then partially owned by Citibank and by the state, and had ownership stakes in a number of other industrial firms, including some firms that were also partially owned by the state Treasury. The state maintained its position, and in fact grew in relevance, in contrast with Romania and Bulgaria.

Figure 3.4 Ownership network of Bank Handlowy, 2000

Note: Arrows point to owners.

This mixed nature of financial institutions so late in the transition process reflected a central choice of policy makers to mix state intervention and nonmarket forms of interaction with market signals. Although in all the cases discussed above we can find some co-mingling of state and private ownership, the mixed nature of the economy here was not just a case of partial privatization. The financial sector in Poland took on forms of governance that reflected a decision to retain nonmarket forms of activity and the persistence of practices, such as the maintenance of ties to other firms and the state, that promoted the development of alliances between the industrial and financial sectors (Williamson Reference Williamson1985: 20).

Firm networks in Romania

The situation in Romania was quite different. Looking at the whole network in 1995 in Figure 3.5, we see again a sparsely connected network of inter-firm ownership. The star at the bottom shows the group of firms connected to the state. There are a number of chains of firms extending away from the state, representing the early privatization deals struck by that time. There is a second ring of owners around the state, representing shares held by employees and manager groups, and not shares sold via privatization to other firms.

Figure 3.5 Romanian ownership network, 1995

Note: Arrows point to owners.

In Figure 3.6, we can see how the network had developed by 2000. Keeping in mind that the second ring of owners around the state (cluster on right) are actually employees and managers, inter-firm ownership ties played a much less significant role than in the other cases under study. Romanian firms did not seek to create bonds of ownership with other firms as actively as their counterparts in Poland and Bulgaria.

Figure 3.6 Romanian ownership network, 2000

Note: Arrows point to owners.

As in Bulgaria, once privatization had advanced and state ownership (the state is the cluster at the bottom center) had receded, the main cohesive element linking firms to each other disappeared, with nothing new coming in its place. By 2005 the Romanian ownership network was made up of narrowly connected firms and their owners (see Figure 3.7). The identities of these owners was similar to that in Bulgaria: Romanian capitalism is underpinned by an alliance of domestic individual owners, many of them closely tied to the Social Democratic Party, and mostly industrial firms.

Figure 3.7 Romanian ownership network, 2005

Note: Arrows point to owners.

Firm networks in Bulgaria

The firm network in Bulgaria looks strikingly different from the initial discussion above of the Polish network. In 1995 (Figure 3.8) it consisted of a core of state-owned firms (cluster at the right) that were directly linked to the state (with only three partial privatizations), as well as a non-state sector made up of many small business groups and firms with many small owners.

Figure 3.8 Bulgarian ownership network, 1995

Note: Arrows point to owners.

By 2000 the network structure had changed quite radically, as seen in Figure 3.9. The most connected firms were industrial companies in the chemical and gas industries. Among the owners, new key actors had emerged since 1995, but these were also of a strikingly different character from those in Poland. Three main types of industrial owners were present. The first were holding companies that integrated firms within a single industrial sector, such as the tobacco and pharmaceutical industries. These were holding companies established by state actors under a plan to consolidate control over firms in a single sector under one corporate structure in preparation for future privatization. A second type were holding companies established by private industrial owners. Third were privatization funds, such as the Neftochim fund and the Petrol fund, set up by the managements of stable large companies that were also strategically important. This last form was conceived as a substitute for management privatization. They were also seen as a viable way of transferring ownership of these companies in the absence of investors and in light of the difficulties of financing privatizations with credits from Bulgarian banks (Prohaska Reference Prohaska2002).

Figure 3.9 Bulgarian ownership network, 2000

Note: Arrows point to owners.

In addition to these industrial holding companies, a second distinguishing feature in Bulgaria was the prominence of individual owners in possession of ownership shares greater than 5 percent. From early in the transition, individual owners were a key component of the ownership transformation. By 2000 individuals held nearly the same share of the ownership of the largest corporations as industrial firms, and by 2005 individuals exceeded the share held by the industry.

Apart from the state (the cluster at the bottom), the most connected firms were companies such as Bulgartabac Holding, or private actors trying to gain a hold on large industrial firms. These new groups (such as Albena Invest Holding and Aktioner Holding) emerged alongside already powerful holding groups such as MG Elite (the renamed Multigroup) that were well established in the early 1990s.

Some horizontal ties among firms are visible. Across the network, however, the average number of ties that a particular owner maintained was much lower in 2000, as is confirmed by the data in Table 3.2. In other words, although there were many actors, each of these had few ties to other actors. This suggests a much narrower type of embeddedness of owners: firms were connected to the state but not so much to one another. Another key feature adds to the distinctiveness of the Bulgarian model: the broad use of holding companies put firms at an arm’s length from the state while nevertheless providing opportunities for interference, as the example of Bulgartabac above illustrated. Consider the contrast to the Polish approach, in which the government used swaps to broadly restructure the economy while retaining political access to the firms. In Bulgaria, state officials found themselves struggling to control and restructure assets that they had themselves distanced but that were still a burden on the state budget.

By 2005 the Bulgarian path of market building was solidly founded on the alliance between individual stakeholders and largely domestic industrial firms, toward which it had been trending for the previous fifteen years. As can be seen in Figure 3.10, networks between firms had become much sparser with the retreat of the state (the cluster at the bottom right). This removed the state as a source of joint ownership ties and thus a possible source of cohesion and alliance among firms. Firms themselves did not, and perhaps were not able to, create networks with other firms on their own. The overall picture, therefore, is one of individual firms connected to owners but with few ties to other firms through those owners. In other words, the majority of firms are detached from other firms except their direct owners.

Figure 3.10 Bulgarian ownership network, 2005

Note: Arrows point to owners.

Conclusion

The mixed nature of the economy in Poland – the combination of market forms with the deliberate persistence of state ownership – also brought with it hybrid institutional forms. These mixed ownership forms in turn supported hybrid forms of market behavior. Activities such as bank lending were not fully marketized, but neither did they take place fully on the basis of social networks. Instead, market signals operated within and were supported by network structures. This distinguishes the Polish path from that of the other two countries examined, in which network forms distorted market outcomes in a way that altered their macroeconomic trajectories.

Business networks developed in a broad fashion in Poland, organized around banks that exchanged debt for equity and developed a broad network of stakeholders. This breadth also empowered and integrated domestic private investors and created opportunities for foreign firms. By contrast, networks of ownership were much more hierarchically organized in Bulgaria. They also developed around a different set of actors: holding companies formed by the state, which often created ownership ties with private holding companies and business groups. By the end of the period under study, the Bulgarian economy was dominated by a mix of individual investors (many with unscrupulous business methods) and ungainly industrial firms. Finally, in Romania, networks retained strong hierarchal features and linked a vast number of passive shareholders, state ownership funds, and large industrial firms – a structure that ultimately empowered a small group of state insiders. Although the effects were not as dramatic as those in Bulgaria, Romania was hardly a robust example of economic development in post-communist Europe.

The question posed at the outset of this chapter was: what role did these different network configurations have in setting the broader political path that each country has followed? With the evidence above, I have argued that both the composition and the structure of business networks shaped the pattern of state–economy interactions. Thus, the prominence of banks with state participation and broad, horizontal business networks in Poland created a context in which much of the leading business sector was part of a network of joint obligation, distributed risk and access to credit, information, and political action.

Bulgaria’s ownership network linked a much narrower group of firms, and from an early stage put an emphasis on linking firms along sectoral lines. This move was intended partly to facilitate political control of assets, but, ironically, it failed both to retain strict hierarchical control and to establish a set of distributed stakeholders across the economy. Thus, owners had a different impact from the one they had in the Polish case. As the examples discussed above suggest, this network structure facilitated the parasitic relations that took hold in Bulgaria.

Finally, the strong hierarchical nature of ownership ties in Romania reinforced the position of insiders against the weak participation of largely passive outsiders, and explains much of the battle of interests against legislative and regulatory reform.