1.2.1.2 Transparency of Capital Movements

Disclosure of major shareholdings, according to the Transparency Directive, should also enhance "overall market transparency of important capital movements."1 As we will see below, such transparency may improve market efficiency through several mechanisms.

The European Commission's initial proposal for the Directive envisaged that a disclosure obligation would not only be triggered by exceeding a threshold percentage of voting rights, but also by exceeding a threshold percentage of the capital.2Moreover, when filing the notification, not only voting rights but also capital interests (i.e., cash flow rights) would have had to be disclosed. These provisions did not make it into the final version of the Directive.3 Nonetheless, it is instructive to consider the rationale of requiring disclosure of cash flow rights.

According to the Commission, disclosure of cash flow rights would have reflected "not only the actual influence an investor on securities markets may take in a publicly traded company, but more generally its major interest in the company performance, business strategy and eamings."4 Such disclosure, however, was only found necessary in case of deviations from one share-one vote.5 Studies show that European firms frequently deviate from one share-one vote, including by issuing shares with multiple voting rights or non-voting preference shares.6 Apparently, the Commission deemed it desirable that there be transparency of cash flow rights in these firms.

Why do cash flow rights matter? Because they determine the extent to which a controlling shareholder bears the cost of private benefit extraction and the benefit from increased monitoring. If voting rights exceed cash flow rights, this encourages private benefit extraction because a disproportionate share of the costs thereof will be borne by outside investors. Theoretical models show that disproportionate structures can distort the controlling shareholder 's incentives to make efficient decisions with respect to project selection, firm size and roles of control.7 Other models show they can distort the market for corporate contro1.8 Conversely, higher cash flow ownership discourages private benefit extraction by making it costlier. It also provides the controlling shareholder with a greater incentive to monitor management and to encourage it to optimize cash flow through dividends. In sum, cash flow rights determine the extent to which the controlling shareholder's interests are aligned with the interests of outside investors. The case for one share-one vote, therefore, turns primarily on its ability to match economie incentives with voting power.9

Still, it remains controversial whether mandating one share-one vote would be socially beneficial, as illustrated by the hefty debate that has recently taken place in Europe over this issue.10 While it may be true that disproportionate voting rights encourage private benefit extraction, they also provide a cheaper way to monitor management.11 As a result, the effects of shifting to one shareone vote are likely to vary per firm. The main objection against mandating one share-one vote, therefore, is that one size does not fit all.

The case for transparency of disproportionality between voting rights and cash flow rights, however, is much stronger. Transparency signals that the controlling shareholder's incentives are distorted, and thus enables investors to better anticipate agency costs.12 Some scholars even argue that as long as companies make adequate disclosure, there is linie justification to restrict the ability to deviate from one share-one vote.13 Empirical studies confirm that outside investors price in the expected costs and benefits of disproportionality. They tend to positively value the incentive effect of cash flow ownership, while negatively valuing the entrenchment effect of disproportionate voting rights.14 This result is consistent with the notion that disproportionality can impact the firm's future cash flows, and that information on disproportionality is therefore fundamental information.

Transparency of "important capital movements" may also enable the market to understand the interest in the share. As we will see below, disclosure of major transactions can be instrumental in conveying other underlying fundamental information to the market, thereby accelerating the process whereby such information is impounded in share prices.

The starting point of this line of reasoning is that investors may possess fundamental information that is not yet impounded in share prices. Of course, in a perfectly efficient market this would not be possible. But the evidence suggests that equity markets are merely semi-strong form efficient with respect to easily obtained and easily interpreted information.15 This means there is still money to be made by trading on information that, although public, is hard to obtain or interpret. A trader with the resources to gather and analyze such information may conclude that the share is overvalued or undervalued and capitalize on this insight by selling or buying shares, respectively.16

Once the trader starts trading, the fundamental information is impounded in the share price through several mechanisms. First, even in liquid markets major shifts in supply and demand can impact the share price directly, pushing the share price towards a new equilibrium.17 Second, the resulting movement in share price may enable price decoding by other traders who suspect that the trading against the market signals the presence of fundamental information and, therefore, start trading in the same direction.18

Price decoding thus occurs when the attention of other traders is captured by trades that signal the presence of fundamental information because they are conducted against an unusual price. Yet, trades can also signal the presence of fundamental information because of other unusual features. Such features include the volume of the trades, the sequence of trades, the purpose of the trades, the resulting ownership level and last but not least, the identity of the trader — Warren Buffet is but one example of an investor perceived to be well informed.19 If other traders become convinced that the trades are driven by fundamental information, they will start mimicking the informed trader. As a result, the process whereby the fundamental information is impounded in the share price is accelerated, through a mechanism referred to as trade decoding.20

How do uninformed traders become aware of such unusual trades? Potential sources of information are the trading book and the stock exchange's transaction reporting system, but these offer limited insight. Traders are able to conceal the volume of their transaction by conducting a series of smaller transactions over time or placing iceberg orders.21 They are also able to remain anonymous through the use of intermediaries or by trading in so-called dark pools (trading venues that do not publicly display bid and offer quotes).22 Finally, they are not required to disclose their intentions or their resulting ownership level.

This brings us to an alternative means through which uninformed traders are alerted: public disclosure of major transactions. Consider the disclosure by a passive mutual fund manager that it has sold its substantial stake in a portfolio company. The sale may be driven by a need for liquidity or a desire to rebalance the portfolio. But it may also be driven by the possession of fundamental information. Thus, the market may interpret the sale as a signal that the share is overvalued.

From this perspective, whether there is marginal value in mandating disclosure of major transactions depends on how rapidly the fundamental information conveyed by such transactions is impounded in the share price. Clearly, there would be linie point in mandating disclosure if the fundamental information would become fully reflected in the share price even before the disclosure is made. But the evidence suggests that, generally, this is not the case. Empirical studies of announcement effects show abnormal returns on both transaction dates and announcement dates, even if there is no overlap between the two.23 The impact of disclosure is nicely illustrated by the following chart, which shows that abnormal returns surrounding major transactions by hedge funds see a jump of about 2.0% on the filing day and the following day.24

The abnormal returns could, of course, be the mere consequence of the control implications of the transactions concerned. In fact, this is the most likely explanation for the abnormal returns shown in Figure 1, which focuses on filings by activist hedge funds.26 These transactions do not necessarily convey underlying fundamental information; rather, the transactions themselves constitute fundamental information. So we need to take a closer look at the evidence and filter out transactions with control implications. This is challenging, because it is not always clear upon disclosure what the control implications are. Two variables are particularly relevant here: the identity of the trader and the purpose of the transaction.

U.S. disclosure rules provide some insight into the purpose of a transaction, at least at the time of the transaction. Qualified parties who purchase shares without the purpose or effect of changing or influencing the control of the issuer file a statement on Schedule 13G, otherwise on Schedule 13D.27 This has enabled an empirical study that examines the differences between the same blockholder's passive (13G) and active (13D) holdings. The study finds that filings of both passive and active holdings produce abnormal returns, even though the returns from the former cases are smaller.28

By contrast, to draw conclusions from empirical studies with respect to firms listed in Europe, one will often need to rely on the identity of the trader as a proxy for control implications. For example, mutual fund managers may be less likely to monitor than family investors, and more likely to gather and analyze complex information on the fundamental value of the share. But mutual fund managers too may act as monitors, and it therefore remains challenging to determine to what extent announcement effects are driven by control implications and to what extent they are driven by value implications. Empirical studies measuring the announcement effects of transactions by investors who are relatively likely to be perceived as informed traders document abnormal returns, though again, they are modest.29

What matters for present purposes, however, is not the magnitude of the abnormal returns. It is the mere fact that the market responds, at least on average, to the disclosure of transactions that are relatively likely to be driven by fundamental information. This is consistent with the notion that such disclosure can convey underlying fundamental information to the market and thereby accelerate the process whereby such information is impounded in share prices.

One implication of this reasoning is that disclosure of short positions could also contribute to market efficiency. After all, short sales are particularly likely to be driven by fundamental information.30 There is some evidence that disclosure of short sales triggers a significant market response.31 This suggests that disclosure accelerates the rate at which fundamental information is impounded in share prices.32

For this reason, a number of countries require disclosure of short positions to the market.33 Their number has increased significantly over the recent years, as regulators across the globe have responded to the recent financial crisis by tightening disclosure requirements.34 To be sure, these measures are primarily driven by concerns about market abuse,35 which is why some countries only require short positions to be reported to the regulator.36 Nevertheless, informational benefits are also cited as a reason for requiring disclosure.37

Even if disclosure can accelerate the process whereby fundamental information is impounded in share prices, one should be cautious in concluding that mandating disclosure for this reason would necessarily result in markets becoming more efficient. One reason for caution is that, as the behavioral finance literature teaches us, investors may not necessarily respond rationally. The recent financial crisis has given skeptics further reason to doubt the market's ability to correctly estimate fundamental values.38 Thus, the FSA recently wareed that disclosure of short sales may cause herd behavior, triggering excessive sales and price declines.39

Another reason for caution is that by reducing the rewards of trading on fundamental information, disclosure reduces the incentives to search for such information. As Grossman and Stiglitz observed nearly thirty years ago, "[t]here is a fundamental conflict between the efficiency with which markets spread information and the incentives to acquire information."40 Moreover, investors who are reluctant to reveal their trading strategies may limit their trading activity to avoid triggering disclosure, which could adversely affect liquidity.41 Mandating disclosure also entails other costs, as we will see below.

Finally, transparency of "important capital movements" enables the market to estimate the size of the free float, i.e. the part of the total number of outstanding shares that is actually available for trading. In at least one European country this is an explicit objective of the ownership disclosure regime, and perhaps for a good reason: the size of the free float may impact liquidity, which in turn may impact the share price.42

First, consider the link between free float and liquidity, that is, the ability to quickly trade large size at low cost.43 One can imagine this becomes harder as the number of free-floating shares becomes smaller. There is some research suggesting that a decrease in the free float does indeed adversely affect liquidity, but compelling evidence is scarce.44 This is different for the link between liquidity and share price. Several studies have tested and confirmed the hypothesis that the more illiquid the stock, the higher the expected return, and thus the lower the share price.45

The function of disclosure of major shareholdings, then, would be to enable investors to understand the implications of the size of the free float. Are there one or more large shareholders who are likely to hold on to their shares, for example to exercise control, and how does this affect liquidity? Taking into account expected trading costs, what is a particular share worth paying for? Mandatory ownership disclosure may help answer these questions.

Of course, there are more direct ways of assessing liquidity, notably by looking at trading volume. Ownership disclosure therefore would seem particularly useful to the extent it can help the market interpret changes in liquidity. Consider the hypothetical where a reduction in the free float causes a decline in trading volume and the decline in trading volume causes the bid-ask spread to widen. Market participants could interpret this widening as a signal that someone is trading on private information and may become reluctant to trade. If, however, they are enabled to interpret these developments as the mere result of a reduction in the free float, they may be more likely to continue to trade, thus contributing to liquidity and ultimately market efficiency.