Page 330 of our 2002 book, Quality Financial Reporting, includes this comment: "No one should accuse accountants of being glacial; after all, glaciers move."Contrary to our jab, the accounting glacier did actually move in May 2008 when FASB issued Staff Position APB 14-1, ponderously titled Accounting for Convertible Debt Instruments That May Be Settled in Cash Upon Conversion (Including Partial Cash Settlement). This FSP is something of a sleeper because no one's said much about it, even though it kicked in for fiscal years and first quarters beginning after Dec. 15, 2008.
Its impact promises to be huge, however, in part because it requires retrospective application, but more so because it throws off the shackles of a decades-old political compromise. Ironically, though, all the staff position accomplishes is taking GAAP back to where it was 42 years ago!
We urge you not to get attached to this pronouncement, because it is only a stop-gap until FASB issues the comprehensive liabilities and equity standard that is currently in the works.
SOME HISTORY
The FSP's history dates back to 1966, when the Accounting Principles Board issued Opinion 10 that required issuers to account for convertible bonds by recognizing a liability equal to the apparent fair value of the instruments' debt portion and equity for the remaining proceeds. The idea was that convertibles are two different instruments that cannot be usefully reported as just one.
However, the board issued Opinion 12 only a year later, in order to "temporarily" suspend Opinion 10, after the board members changed their minds and decided to study the issue further.
In 1969, Opinion 14 required convertibles to be accounted for as if they were plain bonds because the "inseparability" of the debt and equity portions prevented them from being accounted for individually. The glacier then ground to a halt and stood still for four decades.
This political compromise on a controversial conundrum remained intact until FASB and its staff decided to end the widespread game-playing spurred on by this bad standard.
THE CHANGE
As we indicated, the FASB staff position reverts U.S. GAAP back to Opinion 10, but only for convertibles that can be settled in full or in part with cash payments to the investors. This variety of convertibles didn't exist, the staff observed, when APB 14 was issued, so technically they didn't fall under its thumb. Furthermore, it made sense to update practice because so many of this variety have been issued since 2000.
Specifically, when these convertibles are issued, the issuer must estimate the debt component's initial fair value by discounting the contractual cash flows to their present value using the issuer's market interest rate. The difference between the full proceeds and the amount allocated to the liability is recognized as equity.
Subsequently, interest is accrued on the liability at the initial market rate, which means that the reported debt amount grows as the initial discount is amortized over the bonds' life. The equity balance remains fixed until conversion or settlement. Any shares that are issued through conversion are booked at the sum of the retired debt's fair value plus the pro rata share of the initial value of the conversion feature equity. This results in a gain or loss on the income statement for the difference between the debt's book and fair value.
Despite their ancient vintage, these practices will have profound effects on financial statements. Legions of managers have used the APB 14 loophole to report interest expense on convertibles at low nominal rates (even down to zero). When retrospective treatment kicks in, the obvious difference will be a smaller reported debt than under present practice, decreasing the reported debt/equity ratio.
The less obvious but still noticeable disclosure will be a new large chunk of equity that threatens to dilute existing shareholders' percentage ownership. As time passes, another big change will be lower reported earnings, because amortization of the initial discount will create a much larger reported interest expense.
AKAMAI'S COCOS
For example, the Internet company Akamai recently issued contingent convertible - or COCO - bonds with a face value of $200 million and a 1 percent nominal rate. Under old GAAP, it reported only $2 million of annual interest expense. Over the bonds' 30-year life, total reported interest expense would be only $60 million. The small print in the 10-K shows that the bonds created a potential dilution of 13 million shares, or about 7.5 percent.
We've tried to figure out what Akamai will report going forward. Assuming the company's market interest rate is 6 percent, the present value of the COCO bonds' contractual future cash flows was about $62 million when issued, not the $200 million face value. When this number is used instead of the face, reported debt drops and reported equity increases by $138 million.
However, that good news is offset by the fact that the first year's reported interest expense would jump to $3.7 million when the 6 percent rate is applied to the debt's $62 million book value. Further, unlike old GAAP's constant (and low) annual interest expense, the new accounting puts ever-increasing deductions on future income statements. Eventually, the total interest for 30 years would equal the cash paid plus the initial discount, or $198 million instead of $60 million.
These accounting changes are significant, and will be for a great many other managers who issued convertibles to take advantage (or so they thought) of bad GAAP to produce good-looking but totally specious numbers. This new pronouncement's better accounting means that bad-looking but somewhat more truthful numbers will soon be reported.
WHAT ELSE IS COMING?
As we said, don't get attached to these practices, because FASB is working on a major project that will replace them. Whereas the FASB staff position didn't fall too far from the ancient GAAP tree, a 2007 preliminary views document (Financial Instruments with Characteristics of Equity) describes radically different but totally rational and useful new practices.
Specifically, the so-called equity portion of convertible bonds would be moved above the line and reported as the derivative liability it really is. Because its value would vary according to the common stock's value, it would be marked to market, thus providing a continuous reading on the pending dilution and wealth transfer from existing shareholders to the convertibles' holders. Because the debt portion can be settled by issuing shares to the holders, it would also be marked to market.
The outcome would be like a reversion to the original APB 14 practice of reporting both the bond and conversion feature portions as debt, but the components would be quite different in nature and amount compared to the equity's stable book value and the debt's systematically growing book value under FSP APB 14-1.
WAIT, THERE'S MORE
We're eager to see this broader project get completed, because other potential changes will improve financial statement quality. For example, consider the new preferred stock being issued under the Troubled Asset Relief Program and non-governmental bailouts. Based on the preliminary views, this preferred stock will be reported as debt, with its dividends deducted on the income statement.
Those perennially pesky employee options that are now reported at minimal amounts within equity will be clearly identified and marked to market in the liability section, with changes in their value recognized in earnings. This transparency will end the currently invisible compensation created when option values skyrocket after the grant date and subsequent option exercise transactions transfer huge amounts of wealth to managers without reporting any additional expense.
Overall, the new treatment will put substantially more debt on companies' balance sheets and more expenses on their income statements. Furthermore, this new practice will reveal presently concealed volatility.
We expect the fight to be ugly as progress toward truth is resisted by managers who want to maintain the status quo and vainly keep capital markets from knowing what they're up to. We shudder that they don't understand that the markets are trying to estimate these numbers and bake them into stock prices. The new standard's contribution will be more precise and use verified measures that will diminish uncertainty and increase stock prices.
ONE MORE CONSEQUENCE
As a result of all these changes, we expect managers to stop issuing new convertibles and to retire the old ones. With the recent global decline in share values, they might even fritter away more shareholders' wealth by inducing conversion. And that doesn't make much sense, does it?
C-R-A-A-A-C-K!
What was that sound? Near as we can tell, the accounting glacier is actually moving ... and that's a good thing.
Paul B. W. Miller is a professor at the University of Colorado at Colorado Springs and Paul R. Bahnson is a professor at Boise State University. The authors' views are not necessarily those of their institutions. Reach them at paulandpaul@qfr.biz.
