SEC Semi-Annual Reporting — 2018 Comment Letter Tracker (S7-26-18)

89 public comment letters · Release Nos. 33-10588; 34-84842 · comments filed Dec 2018 – Oct 2020 · snapshot 2026-06-04

In December 2018 the SEC issued a Request for Comment on Earnings Releases and Quarterly Reports — asking whether public companies should be able to report less often than quarterly. A Request for Comment is exploratory: it sits one step earlier in the rulemaking process than a proposed rule, gauging whether the SEC should develop a proposal at all rather than putting specific rule text out for public comment. (The later S7-2026-15 rulemaking — tracked here — is an actual proposed rule, one step further along.) The 2018 docket drew mostly institutions — trade associations, accounting firms, issuers, and investors — and split three ways. This page classifies every letter by stance, commenter type, and argument, using the same three-rater method as the 2026 tracker (with a 2018-specific extension for the guidance-vs-reporting argument).

With thanks to Michelle Leder, whose work refreshed my memory of the 2018 docket.

Where commenters landed

Of 89 letters, 83 took a position; 6 stated none. The split is genuinely three-way.

Letters

89

83 took a position

Oppose

36

43% of positioned

Conditional

36

43% of positioned

Support

11

13% of positioned

A further 6 letters stated no position on reporting frequency (procedural or scope-only comments).

Who commented

The 2018 docket is institution-led — trade associations and advocacy groups are the single largest block.

Stance by commenter type

Where each kind of commenter landed (letters that took a position).

Which arguments appeared

How often each argument family was invoked. GU (guidance-vs-reporting — "fix earnings guidance, not reporting frequency") is the 2018-distinctive code and the third most common argument here.

Anti-proposalPro-proposal Alternative / conditionalGU (guidance vs reporting) Legal/proceduralNo rationale

How the conditional letters split

Among the 36 Conditional letters, the basis of the condition divides cleanly.

Frequency-conditional: 21 — would accept less-frequent reporting only for some issuers or some cadence (e.g. semiannual for smaller companies, tri-annual).  ·  Content-conditional: 15 — would accept change only if paired with content reforms (streamlined 10-Q, earnings-release/10-Q integration, guidance reform).

All 89 letters

#DateCommenterTypeStanceRationalesAgree
Methodology — three-rater classification

Each letter was classified independently three times for stance, commenter type, and argument set, reusing the 2026 tracker's rubrics. The headline value is the majority of three. The 2018 corpus added one rationale code — GU (guidance vs reporting) — and an entity bucket for trade associations / advocacy organizations; Conditional letters additionally carry a basis (frequency vs content).

Stance agreement (N=89): unanimous 33 (37.1%) · 2-of-3 55 (61.8%) · split 1 (1.1%) · Fleiss' κ 0.273.

Commenter-type agreement (N=89): unanimous 76 (85.4%) · 2-of-3 13 (14.6%) · Fleiss' κ 0.883.

Argument-set agreement (N=89, multi-label): identical sets 6 (6.7%) · 2-of-3 8 (9.0%) · split 75 (84.3%). Rationale sets diverge more than stance — expected for an inferential multi-label task on long institutional letters.

Note: law firms and bar associations (Cleary Gottlieb, Davis Polk, Sullivan & Cromwell, ABA, NYC Bar) are classified under a dedicated Legal practitioner bucket. The 2018 corpus was not cross-validated against a second model family, so no cross-model figures are shown.

← 2026 tracker (S7-2026-15) · Built by Professor Tzachi Zach, OSU Fisher College of Business, with the help of Claude.