“When you talk, you are only repeating what you already know. But if you listen, you may learn something new.”                                                                                     – Dalai Lama XIV

Disclosure regulation is a cornerstone of modern securities markets. Its economic consequences have been extensively studied and heavily debated. A widely recognized benefit of mandatory disclosure is that it levels the playing field by publicly disclosing to everyone what is known only to sophisticated investors. This leveling reduces trading costs and consequently reduces the firm’s cost of capital.

In our recent paper, available here, we show that this reduced informational advantage of sophisticated investors is not unambiguously desirable. In particular, when sophisticated investors stop trading in a firm’s stock, there is a reduction in the ability of firm managers to glean decision-relevant information from the stock price. In other words, mandatory disclosure impedes the feedback effect of stock prices on managerial decisions, which in turn could harm investment efficiency. In documenting such an effect, we demonstrate that the economic consequences of mandatory disclosure are broader than often thought and, under certain conditions, mandatory disclosure can crowd out informed traders’ ability to acquire information, resulting in less managerial learning from the stock price. This in turn lowers the efficiency of the manager’s investment decisions.

The starting point of our conceptual framework is the direction of information flows between the manager and outsiders. Prior studies on mandatory disclosure often characterize information flows as occurring purely from the firm to outsiders. In contrast, we explore the role of mandatory disclosure from a different angle by allowing for the possibility that information flows in the reverse direction, from the market to the firm. Recent theories posit a role for prices in guiding managerial actions when market participants collectively know more than managers about the implications of those actions. This information can be communicated to the manager only through the price-formation process (Hayek (1945)), thereby creating price-based feedback to managerial actions (Bond, Edmans, and Goldstein (2012)). We call this the Learning perspective.

This Learning perspective introduces a new dimension to the role of disclosure—by including information that is known to the manager in the price, disclosure could crowd out information that is unknown to the manager (and beneficial for the Learning perspective). The overall impact of disclosure now becomes ambiguous as the benefits of less information asymmetry must be traded off against the potential costs of less learning. We study this tradeoff in the context of firm investment, a major corporate decision. We use the traditional investment-q-sensitivity framework (where future investment is regressed on current q) and incorporate disclosure within this framework.

We emphasize that the crowding-out effect does not manifest unconditionally. Rather, its occurrence depends on how the two sources of information (i.e., what the manager knows and what she wishes to learn) interact. If these two sources of information are positively correlated (as in Gao and Liang (2013)), then more disclosure about the known factor results in less informed trading about both the known and unknown factors. Thus, even if the total information included in prices increases, the manager places less emphasis on the price in guiding her future investment decisions, resulting in lower investment-q-sensitivity. On the other hand, if the two sources of information are substitutes (as in Goldstein and Yang (2017)), such that more disclosure about the known factor results in more informed trading about the unknown factor, then mandatory disclosure could lead to more managerial learning, resulting in greater investment-q-sensitivity. Since it is difficult to predict in advance which theoretical assumption is more valid, we leave this to the data.

We select the mandatory change in segment disclosure enacted under SFAS 131 in the U.S. as our setting. In 1997, after extensive lobbying by analyst groups, the Financial Accounting Standards Board (FASB) issued SFAS 131. The new standard’s management approach required that disaggregated segment information be presented to outsiders based on how management internally evaluates the operating performance of its business units. We verify that mandatory segment reporting did indeed reduce information asymmetry, as reflected in increased stock liquidity and less informed trading. We believe SFAS 131 lends itself to the detection of a potential crowding-out effect of mandatory disclosure. The Learning models generally assume two dimensions of uncertainty, viz., assets-in-place (which reflect the firm’s existing production technology about which managers have relatively more information) and growth opportunities (which incorporate factors such as market demand and competition about which managers know relatively less and have greater scope to learn). The availability of disaggregated sales information under SFAS 131 is key here because sales have a natural link to external market demand and the larger competitive environment, information about which the manager may not have and is likely to learn from outsiders. If informed traders acquire private information about these two sources collectively, then the manager’s disclosure about realized sales generated from assets-in-place can crowd out what informed traders themselves learn about growth opportunities.

We find a more pronounced decrease in investment-q sensitivity for affected firms (i.e., firms reporting more segments) as compared with unaffected firms after regulation. This evidence appears to support the crowding-out effect, where more disclosure about what the manager knows dissuades sophisticated traders from incorporating into prices certain information that might be useful to the manager. It also suggests that a (presumably unintended) consequence of leveling the playing field is to decrease the ability of prices to guide investment. In terms of economic significance, affected firms’ reliance on the price to guide future investment fell 18 percent more than did unaffected firms’ reliance during this period.

Our study offers two main contributions. It is the first to document the negative real effects of mandatory disclosure in the U.S. through the Learning perspective. In doing so, it reveals additional economic consequences of mandatory disclosure. Our evidence suggests that disclosure could impose real costs on firms while also providing benefits – something that has not been demonstrated empirically. We stress that our results should not be interpreted as a blanket recommendation against disclosure regulation, and that one needs to be careful in extrapolating our inferences to other types of disclosure regulations in the U.S. or to similar rules globally.

Second, our results caution against attributing a negative connotation to information asymmetry. Most prior studies on the economic consequences of disclosure document decreases in information asymmetry as an argument for regulation. Our evidence indicates that informational efficiency and real efficiency need not go hand-in-hand, and that the former is neither a necessary nor sufficient condition for the latter – something long-postulated by economists (e.g., Hirshleifer (1971)) but not yet documented in the literature.

This post comes to us from professors Sudarshan Jayaraman and Joanna S. Wu at the University of Rochester. It is based on their recent paper, “Is Silence Golden? Real Effects of Mandatory Disclosure,” available here.