One of the points that Stephen made early in his remarks was that while the objectives of financial reporting were fairly non-controversial, they didn’t give one any traction in deciding what sort of accounting we should have.

As I listened to the rest of his remarks, I found myself having a similar sentiment about the “income statement view,” “forward-looking earnings,” and the notion of keeping speculation out of accounting.  While all of those concepts sound relatively non-controversial, I think that they likewise give one little traction to figure out how to account for things.

As I noted in the text chat during the roundtable, I find it difficult to imagine how one would go about measuring income, determining revenue recognition and expense timing without a reference to the rights and obligations that underlie transactions.  Perhaps it is habit of thought or lack of creativity, but as I see it (and as the conceptual framework, both current and proposed explain), definitions of assets and liabilities are primary, and changes in those items are how one determines income.  This does not put the income statement second; rather it imposes discipline over income measurement that is crucial to having income be an informative number.

I’d readily agree with Stephen that certain measurements of assets and liabilities lead to “better” or “lesser” measures of income as a periodic performance indicator.  To that extent, I believe that the debate about measurement of assets and liabilities is one worth taking seriously.  However, to go the next step and attempt to define income and performance without reference to assets and liabilities has a “you know it when you see it” sort of subjectivity that I think is not operational for determining accounting principles.

As I noted during the discussion, estimation (“speculation”) is an inherent part of accounting regardless of a philosophical view about what income statements should do.  A bad debts expense is based upon speculation about the future behavior of customers and the values of their promises to pay, which are quite similar conceptually to fair value measurements.  Allocation of PPE acquisition prices across time requires speculation as to useful lives and patterns of related revenue generation.  In this way, advocating “historical cost” approaches to accounting potentially misleads one into thinking that it is an escape from subjectivity, forecasting, and speculation, but it seems clear to me that it is not.

Viewed through this lens, a focus on performance measurement instead of anchoring on assets and liabilities represents a tradeoff of two kinds of speculation.  I think it could be useful to debate which kind of speculation managers do better, which sorts of speculation enhance versus diminish the usefulness of earnings as a performance measure, and other such questions.  And perhaps that debate would be more fruitful than the religious war over fair value.

I welcome opposing and clarifying views to help me sort this out in my own mind.