FASB Proposed Modifications to Hedge Accounting: Good Thing, Bad Thing, or Just a Thing?*
If we don’t destroy ourselves first, we will someday discover intelligent life on another planet. But when we do, the chances are about one in a billion that we’ll find hedge accounting standards more complex than our own.
Now would also be as good a time as any to peel the onion on hedge accounting since the FASB has recently reached a consensus on a revisions to rules that have been in place since the issuance of SFAS 133 in 1998.
The Basics
At the risk of oversimplifying, the FASB addressed three problems in SFAS 133:
First, there was the problem of accounting for derivatives, which without additional guidance would be measured at historic cost. Historic cost accounting is always suboptimal, but it is especially problematic when it comes to derivatives. Consider, for example, a financial institution with $9 billion in liabilities covered by $10 billion of assets. Next, assume that said financial institution enters into a (near) cashless interest rate swap with a notional amount of $10 billion — or a credit default swap, or a commodities future contract. Basically, it enters into any kind of financial derivative contract, I don’t care which.
All accounting measurement conventions applied to this derivative would produce a net value of (near) zero at inception because the present value of the contract’s receivable leg would be (nearly) equal to its payable leg. But, should the “underlying” of the contract (e.g., an interest rate, a commodity price, a credit rating) change even a tiny bit, there will be a large change in the fair value of the derivative contract — owing to its relatively large national amount .
You don’t need to be a derivatives expert to figure out what’s going on here: derivative contracts are the soft underbelly of historic cost accounting. Failure to recognize the economic effects of the market risks from being a party to a derivative contract renders the entire endeavor of accounting for entities like this hypothetical financial institution an utter sham. Consequently, the FASB correctly decided that interests in derivative contracts must be, without exception, measured at fair value.
First problem solved. But, it creates two additional and related problems, which I will call Problems 2a and 2b:
Problem 2a is how to deal with the irony that if a company were to enter into a derivative contract reduce a source of risk — i.e., reducing the volatility of future enterprise value — then marking a derivative to market through net income could be expected to increase the volatility of future net income. This could be the case if GAAP requires that the item creating the risk in the first place (e.g., a commodity held as inventory or a fixed-rate mortgage loan) is measured at historic cost.
Problem 2a was addressed in SFAS 133 by the so-called “fair value hedge accounting” treatment if the source of the risk is the change in the fair value of a recognized asset, liability, or “firm commitment.” The issuer may elect to offset the gain/loss recognized in income on the derivative with an offsetting change to the hedged item.
Fair value hedging might seem like a reasonable accounting treatment, but there are a number questionable aspects to it. Two of these are:
- Inconsistencies in measurement — Assets, liabilities and firm commitments that happen to be linked with a derivative in fair value hedge accounting are measured one way, and unlinked items are measured another way. Moreover, an added source of inconsistency exists since “special” hedge accounting is optional. For example, both Kellogg and General Mills report that they hedge their commodities positions with derivatives. But Kellogg uses hedge accounting and General Mills doesn’t. Obviously, this is not helpful when trying to analyze the differences in their gross margins.
- Arbitrary measurement — The measurement of the assets and liabilities in the hedging relationship are neither historic cost nor fair value. They are something in between — what former FASB member Tom Linsmeier dubbed “mutt accounting.” This is not much different than the insane numbers generated by the FASB’s treatment of foreign subsidiaries set forth in SFAS 52, and which I described in a recent post as one of the worst and most divisive accounting standards ever written. One of the reasons I was particularly harsh in my assessment of SFAS 52 is because I don’t think that the SFAS 133 fair value hedge accounting provisions would have been at all palatable (or even considered) if SFAS 52 had not opened up a Pandora’s box of arbitrary accounting measurements. (And, as we will see later in this post, it also legitimized the concept of dirty surplus — euphemistically termed “other comprehensive income).
Problem 2b is that if a company were to enter into a derivative contract for the purposes of risk reduction, but the risk was not a recognized asset, liability or firm commitment, then marking the derivative to market through net income would again increase the volatility of future net income. However, fair value hedging would not be an effective solution since there is no recognized hedged item on which the offsetting changes could be lumped into.
The solution to Problem 2b that the FASB came up with is known as “cash flow hedging.” It temporarily parks the portions of the gains/losses on marking the derivative to market that are actually “effective” (more on that term later) as a hedge in Accumulated Other Comprehensive Income (AOCI). When the risk being hedged actually hits the income statement, the appropriate offsetting amounts in AOCI are transferred to net income.
Are you with me so far? These are just the first layers of the onion. I still have to tell you about additional provisions that can make hedge accounting very difficult to pull off in practice. Many of these details could be seen as anti-abuse provisions, and the FASB is, after many years of hearing issuer complaints, finally reconsidering whether all of these details are necessary.
For example, SFAS 133 prohibits arbitrary pairing of a derivative with just any old asset, liability or off-balance sheet risk. The mechanism for this prohibition is known as “hedge effectiveness.” Stated as plainly as I can, if the value of the hedged item decreases by $100, you must be able to document a reasonable expectation that the derivative will more or less (say +/- 20%) cover the loss in the value of the hedged item. FAS 133 requires an assessment of hedge effectiveness at the outset of the hedge accounting and at least once per quarter thereafter.
There is also the question of what to do about ineffectiveness, i.e., the fact that the gain or loss on the derivative could be as much as 20% different than the potential income effect of the hedged risk. Should all changes in the fair value of a derivative designated as part of a hedging relationship be treated the same, or should the portion of the change in the fair value that is effective as a hedge be treated differently than the ineffective portion?
Highlights of the Proposed Changes
So, ready or not, here is what the FASB has in store for hedge accounting:
Hedging a component of a non-financial asset — Say you hold an inventory of coated copper wire. Even though we all know that changes to the value of coated copper wire should be highly correlated with changes to copper prices, the extant rules prohibit hedging of a component of a non-financial item. The reasoning was that the correlation is a matter of speculation; hence, measures of hedge effectiveness were not reliable. But now, the FASB would permit designation of the copper component of the wire as the hedged risk. (Mutt accounting for inventory, here we come!)
Recognition and presentation — Changes in the fair value of the derivative would, with one technical exception, no longer have to be split between effective and ineffective portions. Also, for the first time, authoritative guidance would specify where changes to derivatives would be presented on the income statement.
Interest rate risk — Again, without getting into detail, interest rate risk can be more broadly specified, and the use of the so-called ’shortcut method’ of demonstrating hedge effectiveness has been liberalized.
Determination of hedge effectiveness — You would only have to assess hedge effectiveness at the inception of a hedge — instead of every quarter — unless facts and circumstances change significantly.
Disclosures — Information regarding hedging activities and their effect on the financial statements would be expanded.
Where I Stand
I am supportive of the changes the FASB will soon propose. Hedge accounting is too costly, and trying to get the details right can frustrate managers who are legitimately motivated to use derivatives for creating shareholder value.
But there is a caveat to my support. To start with, I would have promulgated very different hedge accounting rules. Regarding hedging of risks associated with recognized assets and liabilities, in practice these are generally either commodity inventories and financial instruments (e.g., fixed rate loans). If GAAP would require fair value measurement of these assets, then fair value hedge accounting would not even be necessary. (Note that I would not permit fair value hedging of firm commitments. For reasons that I might explain in a follow-up post, it is a bridge too far.)
And, yes, I am aware of the reliability concerns with fair valuing commodities and loans. For commodities, differences in quality and location will add to estimation error. For loans, I have become weary of responding to the steady drumbeat of excuses from the too-big-to-fail financial institutions.
My stock answer to all of these reliability concerns — whether expressed in good faith, or out of fear of losing the ability to manage earnings — is that “perfection is the enemy of the good.” Current market-based information, even if it does require a modicum of extrapolation to derive reasonable measures for particular assets beats historic cost measurement any day of the week.
Regarding cash flow hedging, if you have been reading my blog over the years, you might not be surprised to learn that I am actually pretty indifferent to the rules in FAS 133. Mainly, I rue the unnecessary costs of applying them.
In general, I support accounting rules that only put real assets and liabilities on the balance sheet, measured consistently and reflecting an attribute that is relevant to investors’ evaluation of the performance of an entity’s management. While I snigger about the complex rules of deferring gains and losses on derivatives in AOCI, I really don’t care much about that. I would only like to see better disclosure. Cash flow hedge accounting should be transparent enough for an analyst to easily unwind it — i.e., to be able to precisely determine what reported net income, and each of its components, would have been if cash flow hedge accounting had not been elected.
So, before it publishes its proposal to make incremental improvements to hedge accounting, I hope the board members will ask whether, with added disclosure requirements, analysts will be able to do that one little thing.
* * * * *
Criticizing the FASB’s extant hedge accounting rules is a lot like shooting fish who, entirely of their own accord, jumped into a barrel. It’s not a fair fight. But I’m going to do it anyway.
Hedge accounting is basically a Faustian bargain: granting management an option to attribute gains and losses to what it believes is the “right” period — in exchange for making the balance sheet into a diabolical admixture of asset/liability measurements and OCI deferrals. Because of that, the impending proposals won’t actually improve the quality of financial reporting very much, if at all, but it will at least reduce some of the burden of sanctioned income smoothing, and perhaps add a modicum of transparency.
Even if the FASB is doing something for investors, it’s not worth bragging about.
——————–
*Self-promotion — Speaking of hedge accounting, on December 6, 2016, I will be presenting “Derivatives and Hedge Accounting: The Essentials” in Atlanta for the Georgia Society of CPAs. And one day earlier, I will present “The Essentials of SEC Reporting.” Eight hours of CPE are available for each course.