The long case for Bank of America is not novel and was last formulated here by sag301 in June when the stock traded $10.60. Since then, BAC is down ~50% without a fundamental change to the thesis.

The upside is well understood and was advertised by Paulson, Berkowitz to be $3/share earnings power and a $30 stock price at 10x earnings, when earnings recover. With the implementation of various pernicious pieces of legislation (Durbin, Dodd-Frank, etc) and minor dilution from the Buffett deal and the recent preferred exchanges, the per-share earnings power has certainly decreased and been pushed out. With the fed’s rejection of BAC’s dividend request, returns of capital are no longer a short-term catalyst but, I would argue, do not change the intrinsic value of the franchise. A more realistic upside today would be closer to $1.50-2.00 EPS without increased interest rates, which translates to a $15-$20 stock if/when uncertainty over capital and legacy losses.

The downside comes from the prospect of highly dilutive 08-style capital raises forced upon Bank of America by regulators at the most inopportune time. Unlike the salad oil crisis at American Express, the downside is highly unlikely to be worse than $0 in the common. This is an extreme case, as even in extreme 2008 conditions, having a thin common equity cushion, BAC equity was not completely wiped out. To be intellectually honest, however, any ”believer” must admit that there are scenarios under which current BAC common is nearly wiped out a-la AIG.

If the upside is a $20 stock and the downside is highly dilutive capital raises at the most inopportune time, reward is 4x and risk is -100%. For purposes of analysis BAC can be thought of as a binary situation that boils down to whether the bank needs to raise expensive equity capital at very bad prices. In reality, of course, between the upside and downside cases lies a probability distribution of outcomes for equity holders. Everyone saw this with the Buffett preferred, which will increase the outstanding share count beyond the 10.1bn shares assumed by Paulson, but which clearly still makes BAC a home run if the final share count is 10.8bn after Warren exercises his warrant.

To tilt the odds further in our favor and take advantage of the asymmetry of the bet, we are buying a LEAP call spread between $7 and $12, expiring January 2014. The spread can currently be bought for under $1.00 per share  - the ask on $7 calls is $1.38 and bid on $12 calls is $0.41, so worst case cost of $0.97. If the stock trades $12 or above, the spread will mature at $5 in two years, earning a 5x return.  If the stock is below $7, the loss is 100%, and the break-even is $8.00.

This bet is not without historical precedent. In November 1992, Joel Greenblatt read an article in the OID where Bruce Berkowitz (then at Lehman) made a passionate case for his thesis on Wells Fargo (that article is here: http://www.fairholmefunds.com/pdf/oid1992.pdf). The abridged math for Wells at the time was: stock at $61, normalized pretax preprovision earnings of $36 per share, $30 normalized pretax, $18 normalized EPS, $180+ stock after recovery.

After seeing the article, Greenblatt had the insight to make the bet even more asymmetric.  He explained his options bet in “You Can Be A Stock Market Genius”, excerpt at http://www.scribd.com/doc/67041669/Excerpt-WFC-leaps. Simply put, he recognized that there *was* a possibility that Berkowitz and Buffett were wrong on Wells and that a confluence of a more severe California recession and company-specific issues could tank Wells, saying “but still—a bank is a funny animal. You never really know exactly what makes up its loan portfolio.”  Although he ascribed a low probability to the downside case, he structured his investment through 2-year LEAPSs struck at $80 per share, paying $14 in premium with the stock at the time trading at $77.

I am reluctant to conflate the thesis for 1992 WFC with 2012 BAC.  The problems at BAC today are different than those at Wells in the early 90s.  Unlike 1992 Wells, which gave details of real estate occupancy and cash flow, Bank of America does not provide the loan book transparency to allow investors to get supremely comfortable with the economics of the underlying collateral. Global regulators weren’t trying to overshoot on capital requirements. Wells was not facing the uncertainty of reps & warranties litigation, mortgage putbacks, etc.

At the same time, Bank of America has several very positive attributes today that make this bet very positive in expected value.  After all, we need a 20% probability of a 5x return for it to start to make sense and think the odds are significantly better than that.

The math at BAC is actually better in some respects than the math of the 1992 Wells thesis. With the stock at $5.60 and current pretax preprovision earnings over $3 per share (Moynihan cited $8.9bn as the quarterly number), investors pay under 2x PTPP.  Normalized earnings are higher, as the company is spending over $2bn a quarter cleaning up the mortgage mess, which translates to an additional $0.80 pretax earnings.  At over $13 in tangible book and a Merrill Lynch business that is less capital intensive, we can easily see BAC trading north of $12 in two years, which is all we need for our bet to be a home run.

The bears would argue that the company needs more capital and will be forced to raise it (without putting forth specific math for why they think more capital is needed).  This static argument ignores the fact that Bank of America has options at its disposal, such as a bankruptcy of Countrywide, the issuance of tracking stock for MER, accretive sales of business units, and the further shrinking of its risk weighted assets to meet Basel III on an accelerated schedule, if push came to shove. 

Importantly, despite repeated assurance that significant equity dilution is not necessary, BAC could issue a lot of equity and *still* do well for common holders.  As a mathematical exercise, it is instructive to run the math on a doubling of the share count at current prices.  If BAC issued 10 billion shares at $5.50, it would raise $55bn of equity and have 20 billion shares outstanding. Its tangible book per share would be diluted to about $9.30 per share, which would grow above $10 over 2 years through retained earnings. This would make BAC the best capitalized large bank in the country and it could still earn over $1 per share in a normalized environment.  There is, of course, the nightmare scenario that a capital raise would be forced with the stock at $1, in which case the math is very unpleasant (and why we prefer the options bet to the common).

It is also interesting to look at the sum of the parts of the various businesses BAC acquired over the years.  For example, Merrill through the 90s and before the beginning of the housing boom earned $2-5bn pretax (depending on the year) and is likely worth more under BofA. Our conversations with individual wealth advisors suggest anecdotally that the potential book of business is much bigger today through cross-selling opportunities to existing BofA clients. Within BofA, the Merrill businesses are allocated $40bn of equity and earn mid-teens returns on equity, which is roughly worth today’s entire BAC market cap. Similarly, MBNA earned over $2bn pretax before being acquired for $35bn. For those interested, BAC in its 2011 investor day lay out the normalized metrics on the cards business, and the returns are very high even after including Durbin impacts.

For our bet to be a home run, we need a stock price of $12 by January of 2014. 2 years is a long time for a number of catalysts to play out, including less legacy mortgage issue uncertainty, investor confidence that banks will have a phased-in period of meeting Basel III + SIFI, improved economy, etc.  Meanwhile, BAC organically generates a significant amount of capital (estimates of over $2 per share in combined 2012+2013 earnings). Simply put, if investors figure out that the “worst is over” and BAC has over $14 in tangible book, with earnings power of $1.50-$2 per share, we have a hard time seeing the stock below $8 (our break-even) and can easily get to north of $12.