Disclosure Procedure

Each year, U.S. public companies spend millions of people-hours producing the securities disclosures that undergird public capital markets. But relatively little is known about how firms produce such consequential information, including whether they are spending too much, too little, or just enough on disclosure procedures. Failures of these procedures – from the inclusion of outright falsehoods to inartful drafting – can render a firm’s disclosures misleading, potentially causing investor losses. Of course, those failures can also lead to substantial costs for firms themselves in the form of securities litigation or government investigations. All equal, higher-quality procedures would be expected to reduce or even prevent such errors – and the harms that can flow from them – through controls that force identification and full, accurate disclosure of material information. In other words, procedural disclosure quality matters to substantive disclosure quality.

The optimal design of disclosure procedure is thus of critical concern for investors and firms. This subject is missing, however, from the legal and (to a great extent) accounting literatures, a notable gap considering the substantial body of research on the narrower subject of financial-reporting control. This gap is worth addressing because greater study of disclosure procedure could provide an empirical foundation for designing higher-quality procedures. If adopted by firms, those higher-quality procedures would be expected to produce higher-quality disclosures. That is especially important as investors increasingly demand, and firms increasingly produce, environmental, social, and governance (ESG) disclosures. ESG disclosures, after all, do not fit neatly into existing systems that support GAAP compliance, and so they do not benefit from the long-standing regulatory, practitioner, and scholarly focus on financial reporting.

As a first step, I conducted a study designed to produce a descriptive, generalizable account of disclosure procedure. The study also demonstrates the potential for the kind of systematic comparison of firm-level procedures that would allow for identifying higher-quality procedures. I began the study by interviewing public-company in-house attorneys and accountants on how disclosure is produced within their firms, including what authorities they follow, who participates, and what technologies they use. I then used interview data to construct a survey directed at S&P 1500 Composite Index (S&P 1500) general counsels about their firms’ disclosure practices. The S&P 1500 includes the large-capitalization S&P 500 index, the S&P MidCap 400 index, and the S&P SmallCap 600 index, which together represent approximately 90 percent of U.S. equities. This sample offered the added advantage of allowing me to segment survey respondents by the three component indices, with a firm’s inclusion in a large-cap, mid-cap, or small-cap index serving as a proxy for its organizational complexity.

In the qualitative part of the study, I identified four core procedures common to all public firms: the production of periodic (e.g., 10-Ks/10-Qs), earnings (e.g., earnings releases/talking points), episodic (e.g., 8-Ks/press releases), and governance (e.g., proxy and ESG reports) disclosures. Those procedures can in turn be understood along four factors: authority (whether a procedure relies more on formal or more on informal or ad hoc rules); participation (whether a procedure has a broad or narrow group of participants who draft, review, and revise disclosure); review (whether a procedure is more iterative or linear in terms of drafting, review, and revision); and objectivity (whether a procedure focuses only on producing mandated information or whether it is intended to achieve non-mandatory communications, or even marketing, objectives).

The quantitative part of the study focused on areas of convergence and divergence in firms’ procedures. Findings include that periodic and earnings procedures were generally similar to each other and dissimilar from episodic and governance procedures, a not-surprising result given that the first two happen around the same time and draw from the same sources of information. Meanwhile, although there were few statistically significant differences between the S&P 500, 400, and 600 respondents, there was some evidence that S&P 600 firms economize on their procedures and tend to produce only mandatory disclosures. Perhaps most surprisingly, I found no statistically significant differences when I stratified the survey responses into high- and low-regulation industries. This result was surprising in part because interviewees, at both types of companies, anticipated differences in disclosure procedures between high- and low-regulation firms.

More empirical work is needed before disclosure procedures can be classified as higher- or lower-quality. The survey used in my paper was anonymous, for instance. That helped with obtaining responses for an initial study. But non-anonymous data would be needed to match firm-level disclosure procedures with potential proxies of disclosure quality, such as incidence of accounting restatements or revisions, securities litigation, or government investigations. My sample size of 107 firms was in line with similar surveys conducted by others. But it was opt-in, not random. Companies that were motivated to take a survey about disclosure procedure might, for example, have been more conscientious about their procedures than the average firm, skewing the results. Perhaps datasets on disclosure procedure would not need to include every U.S. equity issuer, but a sample closer to the 90 percent of market capitalization represented by the S&P 1500 would be preferable.

So where would these data come from? I propose that a standard set of disclosures about disclosure procedure – or meta-disclosure – is needed. Firms already must include limited meta-disclosure in their annual and quarterly reports. That is a start. These meta-disclosures follow from mandates to maintain disclosure controls and procedures (DCPs) “designed to ensure that information required to be disclosed [in] reports [filed or submitted] under the [Exchange Act] is recorded, processed, summarized and reported, within the time periods specified in the [SEC’s] rules and forms.”[1] CEOs and CFOs must evaluate the effectiveness of their DCPs, with their conclusions disclosed in forms 10-K and 10-Q (Item 307 disclosures)[2]. The disclosure of this evaluation, although binary between “effective” and “not effective,” gives investors some indication of procedure-related risk. After all, a firm without effective DCP would be less likely to achieve high disclosure quality.

It is possible that the SEC would want to build on this existing, but one-dimensional, Item 307 meta-disclosure by requiring more granular information about disclosure procedure. That information might look similar to the survey instrument used in this paper. This approach, however, could create incremental legal risk for firms and could lead to procedural ossification due to reluctance to change a procedure after it has already been disclosed. Firms might also volunteer more meta-disclosure, but that approach would be non-standardized and would thus make systematic comparison difficult. As a middle ground, I propose that information intermediaries such as credit-rating agencies or proxy advisors might collect and publish standardized meta-disclosures. Scholars, practitioners, and regulators could use that information to identify higher-quality disclosures procedures. Issuers, it would be hoped, would then adopt higher-quality procedures (on a risk-adjusted basis), resulting in higher-quality disclosure and the avoidance of costs associated with disclosure errors.

ENDNOTES

[1] 17 C.F.R. § 240.13a-15(a), (e); 17 C.F.R. § 240.15d-15(a), (e).

[2] Regulation S-K, Item 307, 17 C.F.R. § 229.307.

This post comes to us from Professor Andrew Jennings at Brooklyn Law School. It is based on his recent paper, “Disclosure Procedure,” available here.

Leave a Reply

Your email address will not be published. Required fields are marked *