The End of Accounting? Not So Sure
A month ago, with great enthusiasm, I attended the 2017 Institute of Chartered Accountants of England and Wales (ICAEW): Information for Better Markets conference. The thematic question was whether corporate reporting is moving in the right direction. Having been involved in financial reporting policy work for a number of years, I am reasonably aware of numerous shortcomings of existing accounting information, and I am always intrigued by out-of-the-box thinking that can motivate the enhancement of related accounting standards.
Kicking off the ICAEW conference was a thought-provoking presentation made by Professor Baruch Lev, coauthor with Feng Gu of The End of Accounting and the Path Forward for Investors and Managers (2016). Lev’s presentation focused on propounding evidence supporting his overarching observation that the information within primary financial statements has had a sustained and troubling decline in its relevance.
Some of the headline evidence Lev presented while arguing his case included the following:
- Time series trend data spanning more than a 60-year period (1950–2013) show a decreasing correlation between GAAP information (book value of equity, net income, etc.) and stock prices.
- A widening dispersion of analyst estimates reveals increasing earnings estimation uncertainty. According to Lev, this widening dispersion cannot simply be explained away with an argument that businesses have become riskier. He presented evidence showing that revenue volatility hasn’t increased but earnings volatility has increased.
- “Do FASB Standards Add Shareholder Value?” a recently published journal paper by Urooj Khan, Shivaran Rajgopal, and Mohan Venkatachalam (2017) reports that 104 of the 138 standards issued over 1973–2009 are associated with no change in shareholder value. Thirty-four standards are associated with significant abnormal returns. Of these, 19 (15) decreased (increased) shareholder value. Thus, a mere 11% of the standards improved shareholder value.
Lev then elucidated on what he considered to be the key culprits contributing to the inferred, growing irrelevance of earnings and other line items within financial statements. His principal explanations revolved around the following:
- Failure to adequately update accounting and disclosure standards to cater to the proliferation and pervasive existence of intangible assets within corporations. The rate of corporate investment in physical capital (tangible assets) fell by 35% over the 1977–2012 period, whereas the rate of investment in intangibles increased by 60% during the same period.
- Lack of matched recognition of revenue and related expenses.
- Increased measurement uncertainty, judgments, and application of fair value accounting across line items.
After his tour de force in presenting evidence and arguments deconstructing the relevance of accounting, Baruch mainly argued for the capitalization and the amortization or impairment of intangible assets regardless of whether they are internally generated or acquired (i.e., amortized cost recognition and not fair value balance sheet recognition). Furthermore, in The End of Accounting, Lev also proposed the enhanced disclosure of strategic assets. Following are my main takeaways from Lev’s session.
The Compelling Case for Enhanced Intangible Reporting
The sum total of the evidence and arguments made by Lev makes a compelling case for the necessity to enhance the reporting of intangible assets, including his proposal to develop a framework for reporting strategic assets. It is self-evident that balance sheets are incomplete because of the limited recognition of intangible assets and the necessary high threshold for recognizing such assets as a result of measurement unreliability. Incomplete balance sheets have contributed to the gulf between accounting book value and market-based valuation of reporting entities. Any effort to better inform investors on entity-specific intangibles is overdue.
The Evidence of Declined Relevance May Not Be Conclusive
The mosaic of empirical evidence that Lev presented, regardless of how compelling it was at a headline level, seemed to draw perhaps premature inferences about the usefulness of accounting information. For starters, it was hard for a nonacademic audience to readily discern whether the observed correlation patterns had not been inappropriately taken as synonymous with causation. Was it simply a case of getting the right and comprehensive answer to the wrong question? It is beyond the scope of this blog piece to critically evaluate the research settings, robustness tests, and inferences that can be made from the array of empirical evidence presented by an accomplished academic expert whose work has been published in leading peer-reviewed journals. Nevertheless, as pointed out in a 2001 Journal of Accounting and Economics article by Robert W. Holthausen and Ross L. Watts, “The Relevance of the Value-Relevance Literature for Financial Accounting Standard Setting,” associations between accounting numbers and common equity valuations have limited implications for standard setting; they are a mere association.
I do not question the need for studies that focus on the role of accounting in valuation. I fully support evidence-based standard setting and the need for an increased two-way engagement between academics and practitioners. Mary E. Barth, William H. Beaver, and Wayne R. Landsman’s 2001 rebuttal to Holthausen and Watts, “The Relevance of Value Relevance Literature for Financial Accounting Standard Setting: Another View,” which shows that value-relevant studies are useful, makes some sensible arguments. That said, a healthy dose of skepticism on these inferences made by academic studies and critical and cautious consumption by practitioners of such studies is warranted.
I was left yearning for a balanced and perhaps broader analysis, including one that could identify, evaluate, and sufficiently refute alternative, plausible explanations for the observed declining correlation between accounting information and stock prices. Of course, investor dependence on other information beyond accounting information (i.e., substitution) has increased because of the growing availability of timely, forward-looking information as well as the concurrent inherent constraints in conveying this forward-looking accounting information, which is issued only after economic events or corporate actions have occurred and have been reflected in stock prices. Such substitution, however, does not negate the value of accounting information. This information still has confirmatory value and enables investors to have governance of other timelier capital market information, such as press releases and 8-K disclosures (i.e., investors are able to assess the reliability and consistency across the varied management communication platforms).
With the bold claims of its growing irrelevance, I wanted to see an analysis of whether and how other important roles of accounting information may have changed over time. Beyond being an input for earnings predictions, accounting information assesses stewardship effectiveness and has confirmatory value that is useful for the capital markets system.
In effect, rather than simply assessing the extent to which companies are churning out persistent (and easily forecasted) relatively matched net earnings and whether these earnings are strongly correlated to stock prices, a broader analysis of what is useful financial statement information for investors is required.
Accounting Needs to Be Updated for Increased Business Complexity
I was also intrigued by Baruch’s assertion, while attributing the widening dispersion of earnings to accounting information shortfalls, that businesses haven’t gotten any riskier over the years. Although such an assertion probably makes sense in regard to the economic environment risk profiles (e.g., the global economic environment in the 1970s was highly risky), it is also observable that idiosyncratic entity-relevant risk types (e.g. operational, legal and counterparty risk) have increased over time, driven by increased business model complexity and sophistication in technology, transactions, and type of contracts. Certain aspects of these business models have gotten fiendishly complex over time, including the application of sophisticated financial instruments by companies for their financial engineering and risk management objectives. Accounting standards have enhanced accounting approaches, including requiring fair value reporting and enhanced disclosures. Such changes should not be cast as diminishing the relevance of financial statements.
Increasingly, Nonpersistent Earnings Are Overstated as a Problem
Lev also seemed to emphasize the persistence of earnings (i.e., more easily forecasted) as a gauge of their usefulness. One cannot dispute that investors need to be able to identify and understand the more persistent components of earnings. In part, this is why non-GAAP reporting is so popular (see “Investor Uses, Expectations, and Concerns on Non-GAAP Financial Measures,” CFA Institute 2016).
Yet, a more fundamental problem with earnings goes beyond the noted secular decline in its persistence attributes. As shown by other academic studies, components of earnings can lack persistence but have economic information content (e.g., they can be relevant for assessing risk and a company’s effective stewardship). Furthermore, net income as a summary performance measure is an amalgamation of dissimilar items, including results from core operations, one-off special items, realized gains and losses (including recycled items), short-term unrealized gains or losses, and adjustments to estimates—and this amalgamation without correspondingly sufficient useful disaggregation and classification of items is exactly where the problem arises. To redress this problem, accounting standard setters should significantly enhance the presentation requirements for income statements and cash flow statements so that these components are better disaggregated and the classification of economically similar items is more meaningful.
Endgame Strengthening Required
As Lev shared his view for the way forward, it felt like his answer to the perceived irrelevance of accounting was to roll back to pre-1950 income statement recognition approaches, pare down the level of accounting estimates, make adjustments and fair value remeasurements, and allow the amortized cost recognition of intangible assets on the balance sheet. Incidentally, it sounded like he would have no problem if the balance sheet were viewed as a residual statement rather than one that at an aggregate level is designed to be informative on a company’s financial condition.
It is hard to objectively gauge and I find myself doubting whether such an approach that mainly gives heavy weight to the matching principle and accords primacy to visibility of core operations performance but also introduces its own set of conceptually unjustified practical expedients (e.g., not being concerned about enhancing overall balance sheet reporting) can, at an aggregated level, be superior or even comparable to existing accounting requirements. I find myself agreeing with Arthur J. Rading and Thomas Selling’s 2016 CPA Journal book critique, which observes that Lev and Gu, after their forceful and sometimes persuasive criticisms of accounting information, somewhat disappoint when it comes to crafting solutions (i.e., produced a molehill rather than a mountain). Similarly, at the end of Lev’s presentation, I felt like when it came to the endgame of a thought-provoking and fascinating argument for change, I was left with more questions than answers.
If you liked this post, consider subscribing to Market Integrity Insights.
Photo Credit: ©Getty Images/Caiaimage/Chris Cross