Let’s say an analyst who is reporting on a computer retailer wants to calculate a basic turnover ratio: sales/average inventory. Typically, the analyst must utilize sales and inventory amounts from company financial statements to calculate this ratio. But the numerator, sales, presents a major problem because, due to U.S. GAAP requirements, revenues do not necessarily equal cash; indeed, they rarely do.
The current revenue recognition rules and the new rules starting at the end of this year for public entities require, respectively, that revenue be recognized when multiple elements are delivered or when performance obligations have been met. So getting a clear picture of when a customer gets billed and pays is next to impossible since, under either standard, billing and payment do not, by themselves, allow a company to recognize revenue.
One logical solution is for the retailer to provide a non-GAAP metric that simply shows sales revenues as if recognized when the customer is billed. Unfortunately, the answer to Question 100.04 in the SEC’s “Compliance & Disclosure Interpretations for Non-GAAP Measures” flatly prevents any such disclosure — even outside SEC filings, such as statements on a company’s website. The SEC cites its rule prohibiting non-GAAP measures that may be misleading as justification for their outright prohibition.
As a forensic accountant, I can understand the SEC’s concern about misleading revenue data. Fraud studies over the last several decades point to revenue recognition as the financial measure that fraudsters most frequently manipulated. Giving weight to these findings, U.S. GAAS (generally accepted auditing standards) requires auditors to test a company’s revenues no matter how much confidence the auditor has in that company’s internal control structure.
No other item on the income statement or balance sheet gets this level of scrutiny under GAAS. Just revenues.
All that said, I find it difficult to understand why the SEC believes that presenting sales-as-if-billed will mislead investors, much less lead to fraud, if the proper GAAP measure of revenues is used in the financial statements. The non-GAAP metric is not misleading in that it merely reflects the fact that a bill was sent to a customer on a certain date, and it has a social benefit in that it provides a useful and comparable tool to analysts and other users of financial statements.
Moreover, the sales-as-if-billed metric would actually help us forensic accountants look for fraud. One early warning sign of revenue fraud is a spike in revenues accompanied by a spike in receivables. The underlying rationale is simple: if a fraudster is fabricating revenues, it is very unlikely that anyone is paying the fabricated invoices. Therefore, we look for simultaneous spikes in both revenues and receivables.
GAAP revenue recognition rules cloud our visibility into revenues, however, because the rules push recognition of revenues out over time while the receivables are usually booked when a bill is sent. This cash-accrual mismatch throws off our analytics, but a non-GAAP measure based on billings would provide forensic accountants (and academicians who study accounting fraud) with better tools to monitor and analyze fraud.
The sales-as-billed prohibition is just one example of SEC overreach. If the SEC’s FAQs were subjected to the normal rule-making process, issues like these could surface in comment letters and the SEC staff could respond, with the commissioners having the final word. I think the time has come for these the compliance and disclosure interpretations to be withdrawn and republished as proposed rules so that all interested parties can weigh in.
Charles Lundelius is managing director of the capital markets group at Berkeley Research Group, a global strategic advisory firm. The views and opinions expressed in this article are those of the author and do not necessarily reflect the opinions, position, or policy of Berkeley Research Group, LLC or its other employees and affiliates.