2.5.1 The Definition of the Stake that Triggers Disclosure

In its most straightforward form, a disclosure obligation is triggered upon acquiring a certain number of voting rights attached to shares. But as ownership disclosure mies become more sophisticated, the disclosure obligation will no Jonger be triggered only as the result of holding voting rights. Rather, it will be triggered as the result of obtaining access to voting rights, whether directly or indirectly. This is key to ensuring that the ultimate owner (i.e. the beneficia) owner) can be identified. Indeed, in the absence of ownership disclosure mies with a sufficiently broad definition of the stake that triggers disclosure, it can be quite challenging to identify the ultimate owner. This could be an explanation for why a 1997 study on transparency of ownership structures of EU listed firms found that "[t]he disclosed identity of the agent who has ultimate control over a significant voting block is not entirely reliable in any of the Member States that were surveyed."1

An example of an expanded definition is when a disclosure obligation is also triggered as a consequence of "acting in concert" with others who hold voting rights. The ambiguity of the term "acting in concert", and the resulting controversies, serve as a powerful reminder that while it is important to have a definition that is sufficiently broad, it is important to avoid a definition that is overly broad. Otherwise, ownership disclosure mies risk stiffening communication between shareholders whose aim is merely to coordinate their monitoring efforts, not to gain control over the firm.2 The same applies to mandatory bid mies, which typically can also be triggered by acting in concert.3

Another example of an expanded definition is when a disclosure obligation is also triggered upon acquisition of equity derivatives that grant a right to acquire voting rights, such as physically settled options.4 As mentioned in the Introduction, the FSA has recently taken this one step further. Under UK rules, a disclosure obligation is now also triggered upon acquisition of cash-settled equity derivatives such as contracts for difference (Cfd), merely because these imply the possibility of obtaining access to voting rights.5 Access can potentially be obtained by instructing the Cfd-writer how to exercise the voting rights attached to the underlying shares held by the Cfd-writer as a hedge, or acquiring these shares once the contract expires and the Cfd-writer needs to unwind his position.6 The European Securities Markets Expert Group has recently recommended that the Transparency Directive be amended along the same lines as the UK rules, and the European Commission is now contemplating such amendment.7 Similarly, the SEC is contemplating amendment of the US ownership disclosure rules as part of an overhaul of the proxy rules.8

The measure of ownership disclosure that we have used for our analysis has focused solely on the threshold percentage for disclosure. To measure the stringency of ownership disclosure rules more precisely, future empirical research will also have to take into account the breadth of the definition of the stake that triggers disclosure. The study by La Porta et al. of disclosures to be made when a firm goes public, referred to earlier, has made a first attempt at this. The study uses a variable that equals 1 if a country's ownership disclosure rules require disclosure by shareholders who directly or indirectly control 10% of the shares, and that equals 0.5 "if reporting requirements for the Issuer's 10% shareholders do not include indirect ownership or if only their aggregate ownership needs to be disclosed".9 Despite the reference to indirect share ownership, the variable appears insufficiently specific to capture relevant differences between countries in this respect. Ownership disclosure obligations in virtually every country are triggered by at least some form of indirect ownership. This becomes clear from the fact that not one of the 49 countries surveyed by La Porta et al. is attributed a score of one-half (and the vast majority are attributed a score of one).10 It also becomes clear from the CBR dataset. Although only information on threshold percentages was requested, most country reporters have provided additional information on the relevant ownership disclosure rules. In nearly every instance where they have done so, the information refers to "direct or indirect" ownership or to the "beneficial owner".11 The real question, therefore, is not whether ownership disclosure rules are also triggered by indirect ownership, but how effective these rules are at identifying the ultimate owner. The challenge for future empirical studies will be to define a variable that adequately measures this.