A company’s annual income statement should be a transparent disclosure of its revenues and expenses that investors can readily interpret. Most aren’t, largely because income and expenses classified according to generally accepted accounting principles (GAAP) can be difficult to interpret. In fact, many sophisticated investors tell us they have to re-engineer official statements to derive something they’re comfortable using as the starting point for their valuation and assessment of future performance. In response, many companies — including all of the 25 largest U.S.-based non-financial companies — are increasingly reporting some form of non-GAAP earnings, which they use to discuss their performance with investors.
Eliminating that duplicated effort should be simple. A common-sense revision of GAAP-based income statements would divide the report into two parts: recurring operating income in the first, and non-operating income or expenses and nonrecurring items in the second. Such a structure would provide investors with a clearer summary of income and expenses. It would also be consistent with two core principles for financial-statement presentation proposed by a joint project of the Financial Accounting Standards Board (FASB) and the International Accounting Standards Board (IASB) in 2010, which state that financial-statement information should be presented “in a manner that disaggregates information so that it is useful in predicting an entity’s future cash flows” and “portrays a cohesive financial picture of an entity’s activities.”
The trouble with GAAP-based income statements
Strict adherence to the conceptual principles of accounting often leads to confusion and distortions in an income statement. When companies make an acquisition, for example, GAAP requires they allocate part of the difference between the purchase price and current market value to intangible assets. It then requires companies to amortize the value of those assets over some period of time, reducing their future earnings — in the same way they would depreciate physical assets. The calculation is theoretically consistent but provides no insight into future required cash investments. The annual amortization of acquired intangibles is a non-cash expense and, unlike physical assets, companies either don’t replace them or if they do invest in them, those investments show up as expenses, not on the balance sheet.
Not surprisingly, we haven’t seen any investors or companies using the amortization of intangibles for analysis or valuation work. Most sophisticated investors we talk to tell us they add the amortized value of these intangibles back into income when they analyze a company’s performance — as do most of the companies that report non-GAAP numbers.
A bigger problem with GAAP is its emphasis on producing a single number, net income, that is supposed to be useful to the company as well as its investors and creditors. But sophisticated investors don’t care about reported net income. They want to know its components — or, specifically, to be able to distinguish operating items (sales to customers less the costs of those sales) from non-operating items (interest income or interest expense). They also want to know which items are likely to be recurring and which are likely to be nonrecurring (that is, restructuring charges). Finally, they want to know which items are real and which, like the amortization of intangibles, are merely accounting concepts.
A modest proposal to revise GAAP requirements
It would make life easier for everyone if GAAP requirements themselves were adjusted to require what companies and investors already use, after making all their adjustments, instead of making everyone do twice the work. That wouldn’t require big changes. Simply separate operating and non-operating items in a standardized manner and combine acquired intangible assets with goodwill without amortizing them. Such an approach would enable investors to quickly understand a company’s true earnings and operating performance. It would provide them with the detail they need to assess the economic significance of non-operating and nonrecurring items and decide for themselves how to treat them. And it would enable them to notice trends and patterns and compare performance reliably with peers.
The treatment of non-operating items may warrant some additional transparency relative to today’s reporting. Many are obvious and clearly identified in the current income statement, such as interest income, interest expense and goodwill impairments. Others should be treated differently. For example, gains and losses from asset sales should be treated as non-operating items, with detailed explanations in footnotes. Costs related to closing plants or restructuring operations should be highlighted in their own line items, once again with detailed disclosure in footnotes so investors can assess whether they are truly one-time costs or will be recurring. Pension-related items, such as revaluation of liabilities due to changes in interest rates, expected earnings on the portfolio of assets, and interest on the pension liability, should be separated into their operating and non-operating components. The operating component would be what is currently called the current-year service cost. Everything else is related to the performance of the pension portfolio and changes in the value of the pension liability and thus should be classified as non-operating.
Several leading companies have already started to report their non-GAAP results this way, with approval from investors. The effect can be substantial. For example, IBM reports the non-operating component of pension expense (after taxes) ranged from negative $1.2 billion in 2001 to $400 billion in 2012, with both positive and negative effects in between. Before IBM introduced non-GAAP reporting, investors had to hunt through the footnotes to see what the effect of the pension items was. This also made communication about results quite complex. Now the results and communication with investors are much simpler. It would be even easier if GAAP statements reflected this change.
Changing financial-reporting standards is a slow and complex process, of course. At a fundamental level, U.S. reporting depends on a rules-based system with a strong preference for bright-line definitions, whereas what we’re calling for would require some judgment.
Stringent rules on the disclosure of non-GAAP metrics do prevent companies from using them to mislead investors. Yet the issue remains that companies already provide investors with these data — though investors do need to dig for it in financial statements and public filings. If anything, the current practice of spreading out non-operating adjustment information increases the likelihood that something critical will be overlooked and makes it harder for investors to make informed decisions.
Some users of financial statements may also be concerned, on an income statement like the one we propose, that recurring operating income typically would be higher than the current GAAP-reporting equivalent, which might give investors a rosier-than-warranted view of companies. However, if the new profitability metric were more closely related to continuing operations — and it likely would be — then it would still be more useful for valuation purposes than the GAAP equivalent. Furthermore, net income might end up being the same as current GAAP net income, but investors would have more information to work with in a consistent way. And adjustments won’t always work in a company’s favor; operating income can be adjusted down. From 2000 to 2004, and again in 2008, for example, IBM disclosed that its non-GAAP earnings would have been lower than its GAAP earnings due to negative pension-related adjustments. Finally, sophisticated investors armed with more detailed disclosure are unlikely to be misled.
To prevent abuse, the Securities and Exchange Commission and FASB can take additional steps to require more disclosure about items the company classifies as non-operating or nonrecurring expenses. This will also make for easier comparison across companies, as investors would be confident that items classified as a particular expense would be similar across peers.
Changing the way the GAAP income statement is structured will help investors find the information they need for decision-making in one place and in a format that is easy to understand and compare.
Ajay Jagannath is an analyst in McKinsey’s New York office, where Tim Koller is a principal.
This article was originally published on McKinsey.com. Copyright © McKinsey & Company. All rights reserved. Reprinted by permission.