Cross-post: Are companies complying with the new SEC compensation clawback rule?
Deep Quarry does a deep data dive and sees 103 companies have noted that financial statements were corrected. Only one company said an exec comp clawback was needed.
This is a cross-post with Deep Quarry, the newsletter I occasionally collaborate on with Olga Usvyatsky. If you value our work, please subscribe!
When SEC Rule 10D-1, which requires companies to claw back erroneously paid compensation after accounting corrections, went into effect on December 1, 2023, I predicted that we might see a post-implementation diversity in practice.
My preliminary analysis of roughly 3,500 10-K filings filed between January 1, 2024, and March 15, 2024, finds inconsistencies in companies’ decisions to provide compensation recovery analysis and points to possible diversity in practice in disclosure of footnotes-only corrections.
Regulatory requirements
Rule 10D-1 of the Exchange Act of 1934 requires issuers to clawback excess compensation paid based on accounting metrics that were subsequently revised or restated to correct accounting errors.
The rule requires companies to adopt a clawback policy by December 1, 2023, and attach such a policy as an exhibit to the annual report. The rule also requires companies to disclose on the cover page of the annual report whether financial statements in the report are corrected, whether a recovery analysis is needed, and - if recovery analysis is performed - the conclusions of such an analysis (emphasis added):
...the Commission’s disclosure rules require all issuers listed on any exchange to file their written compensation recovery policy as an exhibit to their annual reports, to indicate by check boxes on their annual reports whether the financial statements of the registrant included in the filing reflect a correction of an error to previously issued financial statements and whether any such corrections are restatements that required a recovery analysis, and to disclose any actions an issuer has taken pursuant to such recovery policy.
Summary statistics
Number of 10-K filings analyzed - 3,561;
103 companies checked the cover page box of the 10-K filing, indicating that financial statements were corrected;
18 companies checked the box indicating that error corrections are restatements that require a recovery analysis of incentive-based compensation received by the registrant's executive officers;
Of the 18 companies with the recovery analysis box checked, 15 attached the policy as an exhibit to the annual report, 4 included a recovery analysis in the 10-K filing, and one company noted that a clawback is required under its clawback policy.
The companies in the population are primarily large accelerated and accelerated filers because non-accelerated filers have an April 1st filing deadline for their annual reports.
Also, many of the companies in my population have not yet filed their proxy filings. While I couldn't locate a recovery analysis for 14 out of 18 companies with the recovery analysis box checked, these companies might include it in their proxy filings.
Only one company in my sample, NCR VOYIX Corp (VYX) disclosed that the Company may need to invoke its clawback policy following the revision of the financial statements:
Based on our preliminary analysis, there is approximately between $1 million and $2 million (based on stock prices as of the date of this report) short-term incentive compensation and long-term incentive compensation that is required to be clawed back in the aggregate from 14 current and former executives in accordance with our compensation clawback policy as a result of the revision of financial statements for interim 2023 periods referenced above. See Note 20, “Revised 2023 Quarterly Financial Information (Unaudited)”, in the Notes to Consolidated Financial Statements in Item 8 of Part II of this Report, for additional information related to these revisions.
A low number of clawbacks is hardly a surprise because, based on Audit Analytic’s restatement report, most of the error corrections are immaterial “little r” revisions.
Diversity in practice – recovery analysis
My interpretation of the rule is that companies are required to perform the recovery analysis if corrected accounting numbers affect performance metrics used to determine executive officers' compensation. If the errors do not affect the achievement of the performance targets, no clawback is required. (Note that I am not a lawyer; this is a personal opinion and not a legal conclusion).
Several companies in my sample provided a recovery analysis aligned with my interpretation.
Let’s look at an excerpt from TranUnion’s recovery analysis that describes the accounting errors and their impact on performance metrics:
As discussed in Note 1, “Significant Accounting and Reporting Policies,” the Company corrected an immaterial error related to an over accrual of expenses, net of the related income tax effect, during the twelve months ended December 31, 2021, that had previously been corrected out of period during the twelve months ended December 31, 2022 (the “revision”). As this correction impacted financial measures upon which the vesting of performance share units granted to the Company’s executive officers on February 19, 2021 (“2021 PSUs”) under the Company’s Amended and Restated 2015 Omnibus Incentive Plan was based, in accordance with the Company’s Policy for Recovery of Erroneously Awarded Compensation (the “Clawback Policy”), the Compensation Committee of the Board analyzed the impact of the revision on the Company’s achievement of performance metrics for the performance period ending on December 31, 2023. The Compensation Committee determined that the revision had no impact, and accordingly, no recovery is required under the Clawback Policy. Following such analysis, the 2021 PSUs vested and were settled on February 16, 2024.
I did not perform a comprehensive analysis to determine whether revised numbers were used to derive compensation performance metrics. However, at least two companies in my sample corrected errors that reduced net income for fiscal 2022 but did not check the recovery analysis box. Hypothetically, these revisions could have affected income-based performance targets – such as EPS or Net Income.
Diversity in practice – footnote-only error corrections
The question of whether footnote-only revisions are corrections that require cover page disclosure was raised during a Q&A section of the AICPA conference in December 2023. In my view, this is a grey area, and footnote errors are not created equal. Checking the cover page box for truly immaterial errors—such as immaterial typos in the description of an accounting transaction —may introduce noise and would not be helpful to investors.
In my previous post, I discussed the controversy of Archer-Daniels-Midland Co. (ADM) intersegment accounting that led to a correction of intersegment sales. From the disclosure perspective, ADM treated the correction as an immaterial revision, indicated on the cover page of the annual report that recovery analysis is not required, and revealed during the conference call that compensation was unaffected.
Archer-Daniels-Midland Co. (ADM) accounting investigation - the errors are immaterial, but SEC and DOJ investigations continue
·
MAR 19
However, even in less controversial cases, several companies treated errors that did not affect the consolidated results as restatements for the purpose of Rule 10D-1.
The disclosure below is from Zebra Technologies Corp’s (ZBRA) annual report. The Company corrected a misclassification of inventory between raw materials, work-in-process, and finished goods categories in the inventory footnote – with no impact on the high-level consolidated financial statements:
Categories of inventories for the period ended December 31, 2022 include a change to correct an immaterial misclassification without impact to Inventories, net as presented on the Consolidated Balance Sheets.
Importantly, I caution against treating the disclosure of footnote-only corrections as a primary sign of lower accounting quality compared to peers. I am not aware of any academic or non-academic empirical analysis that would support this conclusion, although if such an analysis exists I would love to see it.
Browsing Calcbench's company detail pages reveals hundreds, if not thousands, of revised XBRL-tagged numbers. Could some of these revisions be errors that were not transparently disclosed? I am still trying to figure out the answer.
For questions and data inquiries, please contact olga@deepquarry.com