Excess control, agency costs and the probability of going private in France

Abstract

The current study investigates the determinants of going private (GP) in France. It contrasts a sample of 161 firms that went private between 1997 and 2009 with a propensity-score-matched sample of firms that remained public during the same period. The results indicate that, unlike for firms that remain public, the largest controlling shareholders (LCSs) of GP firms control their firms using an incommensurately small fraction of ultimate cash flow rights. This is consistent with the view that agency problems between large and minority shareholders make public firms less attractive to investors, which reduces the benefits of staying public and encourages the LCSs to take their firms private or accept takeover offers. Additional results show that GP firms have more undervalued stock prices and higher free cash flows than non-GP firms. Expected interest tax shields, low growth opportunities, and pre-GP takeover interest do not seem to affect the probability of GP.

Introduction

Going private (GP) is a transaction in which a privately held entity buys out a public firm's shareholders and delists it from the stock market. Several explanations are offered for such decisions, but very little for firms outside the UK and the US. UK and US firms are characterized by widely held ownership structures, which makes them more vulnerable to manager–shareholder conflicts of interests and unfriendly takeovers. Widely held firms, however, do not prevail outside these two countries. Claessens, Djankov, and Lang (2000) and Faccio and Lang (2002) show, respectively, that most East Asian and Continental European firms are closely held and maintained through substantial separation between control and cash flow rights. These differences raise questions about the generalizability of the Anglo-American explanatory factors of GP to other contexts, especially as related to corporate governance.

The current study aims to bridge this gap by examining the determinants of GP in France, where the ownership structure of firms is concentrated. In particular, we examine the relation between the probability of GP and the pre-transaction agency costs of the largest controlling shareholders (LCSs). Our perspective is that the decision to go private is affected by the separation between ultimate control and cash flow rights of these shareholders. Such a separation reduces a firm's attractiveness to outside investors, which increases listing costs. When these costs outweigh the benefits of remaining public, the LCSs may accept their firms GP. Based on this reasoning, we focus on the relation between the probability of GP and the pre-buyout separation between the ownership and control of the LCSs.

Our study contributes to the extant literature in several ways. First, it adds to the GP literature by justifying the existence of a relationship between the agency costs due to the separation of control from cash flow rights and the probability of GP. Most previous studies focus mainly on the agency problems between managers and shareholders (e.g., Lehn and Poulsen, 1989, Weir et al., 2005a, Weir et al., 2005b); however, La Porta, Lopez-de-Silanes, Shleifer, and Vishny (1999), Claessens et al. (2000) and Faccio and Lang (2002), among others, argue that the main agency problem outside the UK and the US is the one that may arise between large and minority shareholders. Second, we augment empirical research on GP by showing that the separation between control and cash flow rights affects the decision to go private. Finally, we complement relevant studies by examining the GP decision in the French stock market. French publicly listed firms differ from their Anglo-American counterparts in that they feature a concentrated ownership structure. In this respect, Faccio and Lang (2002) use a sample of 607 publicly listed French firms as of 1996 and find that only 14% of them are widely held (at the 20% threshold). Large shareholders in France enhance their control beyond their ownership stakes through various means, including pyramiding and double voting and non-voting shares (e.g., Boubaker, 2007). France has also been characterized by an active GP market—in terms of both number of deals and their total value—especially during the last decade. For instance, from 2000 to 2009, France was the largest market in Continental Europe in terms of the number of public-to-private transactions and the second largest in terms of the total value of deals (CMBOR, 2009).¹

Recent years have also witnessed a growing awareness of the role of corporate governance in protecting investors, resulting in a wave of regulations, such as the Sarbanes–Oxley Act of 2002 in the US and the Code of Best Practice in the UK. In France, the 2003 passage of the Financial Security Law, which includes a set of legal provisions related to corporate governance, has also increased listing costs. Analysts state that the resulting stock exchange reforms and related costs of compliance can be a catalyst for delistings in France (e.g., Les Echos, July 29, 2004).² Moreover, in the past decade, French GP transactions have been in the media spotlight and generated enormous interest among practitioners, making the term “going private” a household phrase. Additionally, the growing number of reports on GP in France corroborates the importance of this phenomenon.³ These facts justify our choice of this country as a framework for testing our claims.

We construct a sample of 161 French listed firms that went private from January 1997 to December 2009 and a propensity-score-matched sample of 161 firms that remained public during the same period. We adopt a logistic regression specification to analyze the factors affecting the decision to go private. Our empirical results reveal that GP firms are more likely to be controlled through a substantial separation between control and cash flow rights than non-GP firms. This finding supports our prediction that the LCSs of firms with a greater control–ownership wedge are more likely to accept GP transactions in response to reduced listing benefits.

We also document a negative relationship between pre-buyout stock price performance and the likelihood of GP. When insiders perceive that their firm's market value is underpriced, they may decide to take it private to buy out shareholders at an unfair price. We also find a positive and significant relationship between the likelihood of GP and the pre-buyout free cash flow (FCF), consistent with Jensen's (1986) hypothesis that GP mitigates the agency costs of FCF. The existence of low growth opportunities, tax savings resulting from leveraged GP transactions and pre-GP takeover interest does not seem to influence the decision to go private in France.

The remainder of the paper is organized as follows. Section 2 discusses the literature and develops the hypotheses. Section 3 describes the sample construction, the data, and the empirical design. Section 4 presents the empirical results. Section 5 concludes the paper.