Merrill's Financing Strategy: Harming Shareholders

A bank issues a bond, which has a maturity date. When the bond matures, the bank needs to essentially roll over the debt: it issues a new bond for the same amount of money, at (these days) a higher interest rate. The bank's shareholders aren't particularly happy about the extra interest that the bank has to pay, but ultimately the bank's cost of capital is already built in to its share price.

Or, you could be Merrill Lynch.

Andrew Clavell has found some eyebrow-raising news: Merrill has an outstanding convertible bond, known as a LYON, which is puttable back to the bank on March 13. Now in a normal world the bondholders would either tender their bonds or they wouldn't. If they didn't, fine; if they did, Merrill would presumably have to borrow that money elsewhere.

But Merrill didn't want to risk anybody tendering their bonds. So it unilaterally increased the value of holding on to the bonds, by announcing that they would convert not into 14.1 shares, as originally contracted, but rather into 16.5 shares. Oh, and it gave bondholders another couple of redemption dates, too, in 2010 and 2014, which increases the optionality of the bond and therefore makes it more valuable. Says Clavell:

Merrill get nothing in return for these new features, they are unilaterally value-enhancing for the bonds. The bonds' fair value is now presumably sufficiently above next week's put strike that bondholders will not exercise. Cash call avoided, at least for now.

Which is all well and good for the bondholders, but it's pretty bad for the shareholders: Merrill stock sank 7% today, wiping $3.35 billion off the bank's market capitalization. Which is more than the entire value of the LYONs that Merrill modified. Jeffrey Harte of Sandler O'Neill was mystified, according to the AP:

The increased conversion rate means there will be more shares of Merrill Lynch common stock, diluting outstanding stock.
The reset is part of Merrill Lynch's normal financing strategy, and is considered an attractive financing option in today's market, a spokeswoman for the bank said. Merrill Lynch also has ample liquidity to handle any securities put back to it by investors, the spokeswoman added.
Harte said the reset conversion rate "appears to be dilutive, but not enough" to warrant a slide in the share price.

But it's not really future dilution which causes stocks to fall whenever a bank issues (or modifies) convertible bonds. Rather, it's the simple dynamics of the convertible market: these things (especially highly complex instruments like LYONs, which, Clavell explains, are "variable rate accreting nominal, puttable, callable zero coupon convertible bonds") are bought and held overwhelmingly by hedge funds playing the ancient game of convertible arbitrage. Which involves going long converts and short the stock. Clavell:

The increase in conversion ratio increases the delta of the bond, or its sensitivity to the stock price. The required additional delta hedging by convert arb funds, plus the inconvenient reminder that Merrill's balance sheet is stretched, is plenty enough to prompt the share price falls seen today.

I'm quite sure that fiddling around with conversion rates "is considered an attractive financing option" within Merrill: it costs the bank nothing, at least in the short term. But I'm equally sure that it's not considered a very attractive financing option if looked at from the point of view of shareholders, who have now seen their stock sink to levels not seen since 2003.

On the other hand, riddle me this: according to Yahoo Finance, Merrill Lynch is now trading on a price-to-book ratio of 1.68. Meanwhile, Goldman Sachs is trading on a price-to-book ratio of 1.62. Are those numbers correct? If so, can anyone explain what's going on?