By now, most academics with an interest in standard setting are aware that the FASB and IASB view assets and liabilities as ‘primary’ elements of financial statements, from which income is derived.  However, it is all too easy to see primacy as indicating importance, which I suspect Kothari, Ramanna and Skinner have done in their JAE conference paper.

Upon reading a paper by William Bratton, a legal theorist and next week’s Roundtable speaker, I think a better understanding is that the primacy of assets and liabilities is driven by the position of FASB as an administrative agency, rather than a sense that the balance sheet is more important than the income statement. (Please bear in mind that these are simply my own views, and I am simply working off of published documents, not any ‘insider’ info.)

Bratton argues that the Conceptual Framework (which lays out the primacy of assets and liabilities) allowed the FASB to position themselves as experts hewing to a coherent model as they pursue an overriding goal (decision usefulness), rather than simply a political body bending to the will of varied constituencies. Thus, it was essential for the FASB to have a conceptual framework that was logically sound and defensible. The JAE Conference really brought this point home to me, for reasons I will describe below.

But first, a little history. Those who developed the CF realized that they would be on firmer logical ground by starting with assets and liabilities, which they could define without referring to income, and work from there. You can see this clearly in FASB issued Statements of Financial Accounting Concepts #3 in 1980. This statement was superseded, but I am interested in its historical value. Paragraph 11 captures the Board’s view of the primacy of assets:

Economic resources or assets and changes in them are central to the existence and operations of an individual business enterprise. Business enterprises are in essence resource or asset processors, and a resource’s capacity to be exchanged for cash or other resources or to be combined with other resources to produce exchangeable goods or services gives it utility and value (future economic benefit) to an enterprise. Since resources or assets confer their benefits on an enterprise by being exchanged, used, or otherwise invested, changes in resources or assets are the purpose, the means, and the result of an enterprise’s operations, and a business enterprise exists primarily to acquire, use, produce, and distribute resources. Through those activities it both provides goods or services to society and obtains cash and other assets with which it compensates those who provide it with resources, including its owners.

After defining assets, the statement moves to liabilities (obligations to transfer assets or perform services), and from there to equity (assets minus liabilities). It then defines investments and distributions to owners, so that it can create a definition of comprehensive income. From the ‘highlights’ page:

Comprehensive income is the change in equity (net assets) of an entity during a period from transactions and other events and circumstances from nonowner sources. It includes all changes in equity during a period except those resulting from investments by owners and distributions to owners.

Now, here is what I find interesting: after defining revenues, expenses, gains and losses, the statement stops short of defining earnings. Again from the highlights:

The Statement does not define the term earnings, which is reserved for possible use to designate a significant intermediate measure or component that is part of comprehensive income.

Here is where I think we academics tend to get confused. We assume that because earnings are not even yet defined at this point in the CF, the FASB doesn’t care about them. But the political perspective yields a different conclusion: the FASB couldn’t defend a definition of earnings without further conceptual grounding.

Discussions at the JAE conference really drove home the difficulties of talking about earnings at all, even after we understand what assets and liabilities are. The conference opened with a presentation by Cathy Schrand of her paper on Earnings Quality (with Dechow and Ge). I am not sure I have seen a better demonstration of how difficult it is to define quality of earnings. Ideally, those studying earnings quality would be able to define an underlying construct (the paper refers to ‘the fundamental earnings process’), and then view reported earnings as a proxy that unobservable variable; the more tightly those are tied, the higher the quality of the reporting process. But researchers are forced to do something that is one step removed from that: they identify proxies for quality of the reporting process itself, but looking at things like earnings response coefficients, SEC enforcement actions, restatements, timeliness of loss recognition, persistence of reported income and the like.  this is a very useful approach for conducting positive research, but it ducks the issue in a way that the FASB cannot:  they must construct a proxy for the fundamental earnings process (whatever that is).  To the extent they deviate from a strong theoretical grounding, the definition will be driven by political pressures from preparers who want standards that serve their interests (e.g., reporting smoothable rosy numbers), rather than being driven by expert understanding of the nature of firms’ performance.

In fact, one might argue that standard setters have made assets and liabilities primary because it is so important to get earnings ‘right.’

p.s. I can’t find it online, but I strongly recommend reading Storey, Reed K., and Sylvia Storey. The Framework of Financial Accounting Standards and Concepts. Norwalk, Conn.: FASB, January 1998.