In the good old days, you recognized the assets you owned. For the reasons why things eventually became a lot more complicated than that, there is plenty of blame to go around. Blame the banks and their minions of financial engineers for arcane securitization transactions that transferred legal ownership of those assets, but not much else, to work the accounting wonders that have now become closely associated with the most recent financial crisis. Blame Enron for trying the same scheme with power plants and such. Blame lots of companies like Enron that were able to keep a lower profile.
From the perspective of the rampant abuses that occurred, I suppose you can't find too much fault with the FASB and IASB for taking another tack for asset recognition/derecognition. But, the fact remains that "control" (notwithstanding the difficulty the boards have had in pinning down an abuse-proof definition) as a recognition criteria has problems of its own. Take, for example, lease accounting, one of the more high profile issues of the day. As in my previous post, I'm going to use some recent remarks on the topic attributed to Jim Leisenring by Simon Brown as a jump off point. Before we get to that, however, I am once more compelled to provide a few caveats.
First, I incorporate by reference the caveats as to Simon Brown's reporting of Jim Leisenring's remarks concerning "high quality accounting standards" (HQAS) . Second, I'm really glad I issued that caveat when I did. Soon after I sent that post hurtling into cyberspace, I received a credible email claiming that Brown's reports on Leisenring's remarks as regards to HQAS were inaccurate and misleading. My source claimed that Leisenring had not even referred to any discussions of the characteristics of high quality accounting standards amongst the IASB members.
Although I regret any inaccuracies that may have found their way into my commentary, I only used the information from Brown to establish the existence of an active discussion among IASB members as to what "high quality accounting standards" meant. I do not believe that any mischaracterizations of Leisenring's remarks, although highly regrettable, should call for alterations to my own personal comments.
As to the current post, I shall with newfound trepidation once again rely on Brown's reporting of comments made by Leisenring – this time, however, concerning lease accounting. I have inquired from my sources as to whether Brown's reporting on this topic was substantially inaccurate; and so far, I have received neither a confirmation nor a denial. As is always my policy, if I should subsequently discover any inaccuracies, I will make disclosures to that effect.
What Brown Said Leisenring Said About Lease Accounting
[Leisenring said,] The boards "have no idea what they're doing collectively, including me," …. [he] said the boards have been arguing about the issue for two months. Some members have argued that lessors should derecognize the portion of the asset they gave up by right of use, which he called "extraordinary accounting." He said the [sic, their?] derecognition model would suggest that every time a hotel rents a room, for example, it should account for the loss of the part of its building that it gave up, then reverse it the next day when a guest checks out.
I Can Fix That Problem!
The hotel example is clearly what one gets by taking a "control" criteria for recognition/derecognition to its logical extreme. I can think of no other way to fix the problem than to restore "ownership" to its rightful place as the principal criteria for asset recognition; to wit, only legal entitlement to an economic resource may trigger recognition of an (as opposed to "the") asset.
By "legal entitlement," I am referring to one of the types of circumstances: (1) ownership of an asset; (2) a right (conditional or unconditional) to receive an asset; or (3) a right (conditional or unconditional) to use an asset. But unlike any other "traditional" accounting models, I would also need to define a liability symmetrically as an obligation to a holder of (2) and (3), above.
This revised asset/liability criteria makes lease accounting very straightforward, although I want to take this opportunity to remind you that I am severely critical of the measurement aspects of its proposals as well — they are truly terrible. Anyway, as regards to recognition only, the lessee (i.e., hotel guest) should recognize an asset for its right to use the property being leased (i.e., hotel room); and an obligation to pay the lessor (i.e., hotel owner) in exchange. The lessor, for its part, recognizes an asset for the right to receive payments from the lessee and an obligation to provide access to the hotel room. Thus, the accounting for the hotel property itself is unaffected by the lease arrangement as no change in ownership occurred.
Does My Fix to the Leasing Problem Create Other Problems?
You may already be thinking of potential shortcomings to my "legal entitlement" approach. So far, I have thought of two – but I can easily live with them. In fact, I vastly prefer them to the results that come out of the control approach.
First, legal entitlement could put all executory contracts on the balance sheet, not just leases, the pioneer of on-balance-sheet recognition of an executory contract. That's because it was rightly seen by many as an essential exception to close one of the most popular off-balance sheet loopholes. Years later, a forward contract meeting the restrictive definition of a "derivative" became another form of executory contract that was put on the balance sheet (SFAS 133).
So, I suppose, that standard setters can continue to pick and choose which executory contracts to scope in or out of its rules no matter whether "control" or "legal entitlement" is the asset recognition criterion. As for my own preferences, except for limited practicability and materiality considerations, the ancient prohibition against recognition of executory contracts has become obsolete.
Second, the "legal entitlement" criterion for liability recognition as I have stated it would not result in recognizing "contingent liabilities" such as pending lawsuits (unless, of course, the case had been adjudicated). It would also not comprehend constructive liabilities. With respect to contingent liabilities, all sorts of regulators have been trying for ages to come up with consistent accounting for contingent liabilities and the results, to be kind, have been inconsistent if not just plain ineffectual. With respect to constructive liabilities, they are merely tools for earnings management.
I'm most happy to get both contingent and constructive liabilities off the balance sheet, although disclosures of contingent and constructive liabilities have their place, and arguably should be strengthened. There is nothing wrong with expansive narrative and quantitative disclosures in notes or in MD&A.
I should also note that my criterion would comprehend contractual warranty obligations and any other similar obligation to stand ready to perform in the future.
Control as a Presentation Criteria
Finally, I should point out that the notion of "control" can still be useful, even though it has done little more than to bollix things up. For another recent example, see the infamous Repo 105 mess.) Its role would be limited to determining the presentation of financial assets. If legal entitlements over financial asset(s) convey the power to control another entity, then consolidated presentation would be appropriate. Other circumstances could also indicate that consolidated presentation is appropriate, too, such as whether an entity is a "primary beneficiary." If the investment(s) confer "joint control," then proportionate consolidation would be appropriate; but, that's certainly fodder for another blog post.