Equity sales in Belgian corporate groups: expropriation of minority shareholders? A clinical study

https://doi.org/10.1016/S0929-1199(02)00047-0Get rights and content

Abstract

In Belgian corporate groups, complex pyramidal structures and interlocking ownership lead to separation of ownership and control. This may generate incentives for the controlling shareholder to divert resources within the group through intragroup equity sales. This in turn could lead to significant private benefits at the expense of the minority shareholders. We test this hypothesis by investigating the stock price reaction to the announcement of equity sales in Belgian groups. Our results suggest that intragroup equity sales create value for minority shareholders. Equity sales between group members and non-group members do not seem to affect the value for minority shareholders in Belgian groups.

Introduction

Gaining insight in the financial mechanisms in Continental European corporate groups is difficult and research on this issue is relatively scarce. The literature primarily focuses on American conglomerates and multidivisional structures, which are very different from European groups. American conglomerates are tightly structured, whereas Continental European groups are often loose organizations with an ill-defined global policy. In Belgium, these groups are characterized by complex pyramidal structures, interlocking ownership and voting pacts. These structures allow the ultimate owner to maintain control over a large group of companies through cascades of holding companies, while owning only a fraction of their cash flow rights.1 This separation of ownership and control generates strong incentives for the controlling shareholder to divert resources between companies in the group.

The shares of most listed Belgian holding companies, which function as intermediates in ownership structures, trade at a significant discount relative to their net asset value. If the price at which these shares are traded on the stock market reflects the value for minority shareholders, this suggests that there are significant private benefits for the controlling shareholder(s) at the expense of the minority shareholders. A minority shareholder who holds a stake of x% in the company's equity does not receive x% of total company value, but only x% of [total company value−private benefits extractable by the controlling shareholder(s)]. The diversion of resources between companies at the expense of minority shareholders has been labeled tunneling (Johnson et al., 2000). It can take many forms, such as selling of inputs or purchase of outputs at nonmarket prices, or high or low interest rate loans. In pyramidal structures, wealth may be transferred from the companies located at the lower levels of the group towards companies at the upper levels, where the ownership of the controlling shareholders, in terms of cash flow rights, is less diluted. Belgium provides a good setting for testing this tunneling hypothesis, as it has the lowest score of all countries on the La Porta et al., 1998, La Porta et al., 2000 investor protection index (0:6). Moreover, unlike other countries like, e.g. Germany, Belgium does not have a ‘group law’ (Konzernrecht) that protects investors of controlled companies against self-dealing in corporate groups.2

In this paper, we first investigate the stock market price reaction to the announcement of equity sales between companies that belong to the same Belgian group (at least one of them being a holding company). While there can be several reasons for equity sales, such as a change in the control structure, transparency reasons, etc., we focus on wealth transfers. If these equity sales lead to a transfer of wealth from companies at the lower levels of the group towards companies at the upper levels of the group, we expect a negative price reaction for the company at a lower level. Moreover, we expect a positive price reaction for the company at the upper level if some of the wealth transferred to this company is not expropriated as private benefits, but is available to all shareholders of that company. The sample we use to test this hypothesis consists of all 15 significant intragroup equity sales that were announced in the leading Belgian financial newspaper ‘De Financieel Economische Tijd’ in the 1996–1999 period. The results show a significant positive price reaction to the announcement of equity sales within Belgian groups for both upper and lower level companies. The hypothesis that these sales are priced so as to transfer wealth to the upper level at the expense of the minority shareholders of the lower level company is therefore not confirmed. Instead, our results suggest that the buying/selling of equity stakes within Belgian corporate groups creates value for the minority shareholders of both companies. An explanation for these results is that the equity sales are part of a strategy of Belgian groups to create a more transparent group structure.

Second, we investigate the share price reaction when a Belgian holding company announces either the sale of an equity stake to a non-group buyer, or the purchase from a non-group seller of an additional equity stake in a company in which the holding company already has a stake. If these transactions are part of a strategy that creates value for minority shareholders, we expect a positive price reaction of the holding's shares. On the other hand, if these transactions destroy value for minority shareholders, a negative price reaction is to be expected. The sample we use to test this hypothesis consists of 26 equity sales to or from a non-group partner in the 1996–1999 period. We find no evidence that the sale of an equity stake to a non-group buyer or the purchase of an additional equity stake from a non-group seller significantly affects the value of minority shareholders of Belgian holding companies.

The remainder of the paper proceeds as follows. Section 2 reviews the features of Belgian corporate groups and holding companies. In Section 3, we outline the research questions. We discuss the sample and the research methodology in Section 4. The results are presented and analyzed in Section 5. We conclude in Section 6.

Section snippets

Corporate groups and holding companies in Belgium

Banerjee et al. (1997) define a holding company as “a professionally managed institution owning a portfolio of stocks in public and private companies with the purpose of influencing them. In realizing this objective, a holding company acts both as a financial intermediary and as an active shareholder.”3

Research objectives and hypotheses

Our research question is whether listed holding companies in Belgian corporate groups create private benefits for the controlling shareholders at the expense of the minority shareholders, through the sale or the purchase of equity stakes. We restrict this study to equity sales involving holding companies, because it is precisely the goal of holding companies to create value through buying, holding and selling equity stakes in other companies.

First, we investigate equity sales within corporate

The database

In order to obtain the relevant events, we first need a clear definition of a holding company. Following previous studies on Belgian holding companies, we use the classification of the Belgian Stock Exchange (BXS), which excludes pure financial holding companies. The BXS defines holding companies as “those companies whose purpose is to invest in other quoted companies, except financial institutions”. We also exclude investment companies, which are organizations that take a pool of investors'

Hypothesis 1

The results of both the parametric and nonparametric tests are summarized in Table 3.8 Fig. 5 shows the average cumulative abnormal return of companies at the upper level over the 60-day period around the announcement date of an intragroup equity sale. The CAAR−30,+30 based on the market-adjusted model is +2.1%, while the

Conclusion

For a 60-day event window as well as for a 10-day event window, we find a positive price reaction to the announcement of equity sales within Belgian groups for both companies located at the upper and the lower level of the group. The hypothesis that these sales are priced so as to transfer wealth to the upper level, at the expense of the minority shareholders of the lower level company, is therefore not confirmed. Instead, our results suggest that the buying/selling of equity stakes within

Acknowledgements

We thank Vladimir Atanasov, Ariane Chapelle, Linda Van de Gucht, Ilse Verschueren, Wim Voordeckers and an anonymous referee for comments and suggestions. We also thank participants at the 2001 meeting of the Flemish Economic Association, the 2001 Belgian Financial Research Forum, the 2002 EFMA/IESE symposium in Barcelona, the 2002 FMA meeting in Copenhagen, and the 2002 EFMA meeting in London for their comments. Financial support by the Fund for Scientific Research-Flanders (Belgium) is

References (22)

  • Bianchi, M., Bianco, M., Enriques, L., 2000. Pyramidal groups and the separation between ownership and control in...
  • Cited by (0)

    View full text