Are you comfortable with allowing management to choose whether to recognize unrealized gains or losses based on management’s intent to hold securities to maturity?  How about basing the timing of revenue recognition on the firm’s business model?  Differences in intent and business models might very well alter what information users find relevant.  However, intent and business models are not easily verified, so the audit process seems unlikely to address management’s obvious conflict of interest.

On the other hand, it is hard to see how the proposed revenue model will be implementable without some way of taking account of such ‘soft’ considerations.  To see why, consider one of the cases discussed at the Financial Reporting Issues Conference:

A company that sells a one year policy for professional liability insurance for one premium paid at the beginning of the coverage period.  The customer is covered for incidents occurring during the year, but cash outflows for both the claim and related processing costs are most likely to be incurred three or five or even ten years later.

Under the proposed revenue recognition model, the firm would debit cash for the amount of premium as soon as it is received, and credit an equal liability for a ‘performance obligation’ to the customer.    The firm would recognize revenue as they satisfy the performance obligation (debit PO, credit revenue).  The question is whether the performance obligation to the customer includes performing duties arising after the claims period is complete.  One could argue that the obligation to the customer is simply to provide coverage, and at the end of the coverage year, the insurer should have taken all of the PO into revenue.  There is still a liability, of course, for the expected legally obligated payments, but that is created by a separate entry:  debit coverage expenses, credit a reserve (liability) for the expected payout.

An alternative argument goes like this:  part of the performance obligation is to provides the services necessary to settle the claim.   This may not be cheap, especially if the contract requires the insurer to provide an aggressive legal defense to protect the insured’s reputation.  In this case, the customer is buying the insurer’s settlement service as much as they are buying the agreement to cover expenses.

At heart, the question is how preparers are supposed to distinguish obligations to their customers (which are performance obligations and result in revenue) from other obligations.  One could make the case that the distinction actually depends on the intent of the customer.  Did the customer buy the insurance in order to receive post-coverage services?  If so, delay revenue recognition until the customers’ demands for those services are fulfilled.

If you buy this argument, the ‘management approach’ method for segment reporting might be effective.  That approach entails looking at the reporting structures within the firm — who reports to whom, how do internal documents identify segments and business lines — and using those structures to identify segments that should be reported separately for external financial statements. The analogy for revenue recognition would be to look at how the firm presents its value proposition to customers.  In the insurance case, if the firm does not promote post-coverage services, the obligation to pay for such services would not be not a basis for revenue recognition.

The segment reporting guidelines are successful, in my view, because the firm is unlikely to alter reporting structures within the firm in order to change how they can present segments.  If a firm is unlikely to change its marketing materials and sales pitches in order to get desired revenue recognition, the revenue recognition model might well be reasonable.  Otherwise…well, we might have a problem.