The accounting for convertible bonds is complicated, but it provides insight into how accounting loopholes emerge. First consider a plain convertible bond that is settled, or repaid, using only the issuer’s stock. In practice, that means investors exchange the bond for a set number of company shares, and a profit is made if the converted shares are worth more than the cost of the original bond. That profit is referred to as the conversion spread.
Traditionally, issuers have liked convertible bonds because the interest rate paid to investors is set lower than on a conventional bond, so the company records a lower interest expense on its income statement. Despite the lower coupon rate, the bond still attracts investors because the interest is supplemented by the embedded call option on the issuer’s stock (the conversion option), which gives investors the opportunity to benefit from a rise in the company’s stock price.
The downside for issuers, however, is that convertible bonds dilute a company’s EPS. Under FAS 128, companies are required to apply the “if-converted” method of computing diluted EPS, whereby an issuer assumes that the bond will be repaid using company stock. The shares are then added to the common shares outstanding tally, and earnings are divided by the total to produce the diluted EPS.
The dilutive effect, however, is partially offset because under FAS 128, companies can add their after-tax interest expense back to earnings. The company assumes — for accounting purposes — that the bonds will be converted into stock, so it does not have to pay out the after-tax bond interest amount that accrued for the period.
C-ing Is Believing
In the 1980s, investment bankers developed a net-share-settlement convertible bond that could be settled in cash and shares. Several accounting-rule adjustments followed, and by 1990, a FASB task force issued definitive guidance for the bond that it dubbed “Instrument C.” That new guidance, contained in EITF 90-19, also gave rise to beneficial EPS treatment, in that companies were required to settle the par value of the bond in cash, and the conversion spread in shares.
Because the bond agreement required that the bulk of the bond’s value was settled in cash, issuers had to assume — for EPS purposes — that only the much smaller gain, or conversion spread, would be settled in stock (as opposed to the entire instrument). Consequently, companies no longer got to add the instrument’s after-tax interest expense back to the earnings portion of the EPS calculation, but only the shares issuable to satisfy the conversion spread were factored into the EPS denominator. Companies commonly refer to this as applying the treasury stock method to convertible debt. Because the treasury stock method is typically much less dilutive than the if-converted method, this development was viewed favorably by issuers.
To improve upon what issuers considered “a good thing,” investment bankers introduced Instrument X soon after Instrument C made its debut. By 2003, Comerford, then with the SEC, called the debt instrument “the golden goose of convertible bonds” in a speech before the American Institute of Certified Public Accountants.