This write-up was mostly completed when Barrons wrote a piece this Saturday about the security. Unfortunately for potential buyers, this has caused a pop (temporary or otherwise) in the security from $8 to $10. I still think it is value here, though obviously with 25% less upside. Potential buyers could own some here and increase the position if it fades back to the pre-Barron's attractiveness.
The Bank of America E Pfd, a non-cumulative low-dividend floating rate preferred stock, is a very attractive risk-adjusted reward for US taxpaying investors. Equating the preferred with debt and solving for the same after-tax yield, the BAC E's 10% yield is the equivalent of 13.1% yield for a corporate bond (assuming dividend tax rate of 15% and interest tax rate of 35%). The CDS is currently at 125 bps so one is getting paid roughly 8x the credit spread for taking preferred risk vs CDS risk (with some obvious financing differences in funding the cash). While non-cumulative preferred stock are admittedly a weak structure, the BAC's seem quite attractive relative to other similarly structured preferreds at other large banks (even those with much wider CDS spreads) or the credit of the large banks.
There is obviously a lot of discussion that can go on here with BAC's balance sheet, how the US governnment will intervene on continued weakening, and the appropriate cost of capital in this market. Nevertheless, considering BAC's senior debt credit spreads and its equity market cap cushion, the attractiveness of the BAC E's is undeniable relative to other BAC prefs and the pref securities of other large banks. They have greater upside, greater yield, less downside, and an embedded interest rate option.
BAC 125 bps 5 year CDS
BAC E, L + 35 bps with 4% floor, current yield 10%, trading at 40% of par, non-cumulative
BAC D, 6.204% coupon, current yield 10.66%, trading at 58% of par, non-cumulative
BAC U, 5.875% coupon, current yield 9.04%, trading at 65% of par, trust pref structure (may defer cash coupons for 5 years)
C 205 bps 5 year CDS
C S, 6% coupon, current yield 10.71%, trading at 56% of par, trust pref
C I, 6.5% coupon, current yield 11.82%, trading at 55% of par, non-cumulative, convertible at $33.73/share
GS 300 bps 5 year CDS
GS C, L+ 75bps with 4% floor, current yield 7.46%, trading at 54% of par, non-cumulative
JPM 125 bps 5 year CDS
JPM G, 6.37% coupon, current yield 7.57%, trading at 86% of par (cumulative pref, formerly BSC)
MS 410 bps 5 year CDS
MS A, L + 70 bps with 4% floor, current yield 8.70%, trading at 46% of par, non-cumulative -- directly comparable to BAC E, trading higher, yielding lower with wider CDS
WFC 125 bps CDS
GWF, 5.625% coupon, current yield 7.10%, trading at 79% of par, trust pref
The upside is potentially high -- were BAC E's to tighten to similar levels as the JPM or WFC shown here, the security would appreciate from its current price of $10 to $13.5 or so, a return of 35%, prior to considering the interest rate option (worth an additional $1, conservatively). Were bank preferreds to start trading at double the CDS spread-- which seems justifiable if one assumes that government bailouts would not protect holdco debt holders more than holdco pref holders-- BAC E would trade to $20.
How much is the interest rate optionality provided by the 4% floor worth?
The BAC E's floor coupon of 4% creates a valuable interest rate option. One can think of it as a 4% pref security combined with a series of call options on the 3 month libor struck at 4% (these are called caplets by fixed income people). Interest rate options are extremely expensive in the current environment, as there is much uncertainty about the sustainability of current low rates. Due to the perpetual term of the security, the imbedded BAC credit risk, and the issuer call, there is some complexity in valuing interest rate optionality. Without being more precise than my expertise justifies, I'd point out that 3 month libor has averaged about 5.25% over the last 25 years and was 10% in 1989. So there is real value in the floor coupon optionality, particularly if one believes that Bernanke monetary policy will lead to inflation down the road. I put the value of the option at greater than $1 per share, but I would encourage any interest rate derivative experts to opine.
Trust Prefs vs Cumulative Straight Prefs vs Non-cumulative Straight Prefs
The preferred market can be broken into three types: trust prefs, cumulative straight prefs, and non-cumulative straight prefs. Trust prefs are pref shares issued by a trust and backed by subordinated corporate notes. They are taxed on a look-through basis as notes, so "dividend" payments are taxed as interest income. Trust pref payments are linked to the underlying subordinated notes, which usually have a deferral option that allows the issuer to defer interest payment for up to five years. Cumulative straight prefs are prefs which accrue dividends for all periods, whether declared or not. If the Board deems it imprudent to pay the dividend in cash, dividends accrue in arrears. Non-cumulative straight prefs do not accrue dividends in arrears, so, if the Board chooses not to declare a dividend, that dividend is simply lost forever.
The trust prefs offer the increased rights of being a look-through creditor-- albeit the most junior creditor imaginable. The advantage of being a creditor was demonstrated by the UST-led, creditor-backed GSE coup, as value was gifted from preferred and common equity holders to creditors and homeowners without anything resembling a fair process. Some may speculate that, should the banks ultimately need government aid in excess of what has already been provided, trust prefs may be treated better than straight prefs.
Offsetting the advantages of creditor rights are the substantial tax advantages of owning straight prefs, both as an individual and as a corporation. Individuals are taxed at the 15% dividend rate and corporations only pay tax on 30% of dividends received.
There are three reasons to believe that banks will continue to pay pref dividends in all but the worst of instances. The first, and perhaps least convincing, is that the pref market has been an important source of financing and any issuer failing to pay dividends would shut off the pref market as a source of funding for the foreseeable future (it's worth noting that the pref market seems basically shut to all issuers at the moment). The second reason is that all the big banks have accepted TARP preferred investments, which are pari passu to existing prefs, meaning that an issuer can't short-change pref holders without short-changing Paulson at the same time. The final reason is the common dividend, which would have to be eliminated if pref dividends weren't paid in full.
The non-cumulative pref agreements generally forbid an issuer from declaring cumulative pref dividends without declaring dividends on non-cumulative shares of equal rank. Allowing cumulative and trust pref dividends to accrue in arrears is a sign of substantial weakness. Given the leverage of all the big banks, such a situation is likely to either resolve itself quickly (say, with a bailout) or lead to the demise of the institution. It's important to note that, in the Lehman bankruptcy, trust prefs = cum prefs = non-cum prefs = common shares = 0, which is likely to be the case for any of these extremely leveraged institutions.
Why low coupons are a good thing
In the pref asset class, a high stated coupon seems to perversely garner a premium yield valuation. I will leave it to Psychology hobbeyists to speculate why that might be so. If the market does prefer a high coupon, BAC E's low coupon, a ridiculous L +35bps, makes it vulnerable to cheapening. However, there are two reasons to prefer a low coupon pref to a high coupon pref of the same yield. The most important reason is that the low coupon pref is less burdened by the imbedded issuer call. If credit conditions improve, the upside of high coupon prefs is capped by right of the company to buy back the prefs at par. After an initial non-call period, virtually all prefs are short this call option. BAC E's terribly low coupon makes the company's call at par (or 250% of current market value) essentially worthless. The second reason to prefer low coupon prefs over similarly yielding high coupon prefs is that, in event of a dissolution of the company in which prefs do not recovery full liquidation preference, the lower coupon pref will have less downside in all scenerios except one in which all prefs get zero. This is so because the low coupon pref trades at a lower percentage of its liquidation claim.