Saturday, October 1, 2011

BAC write up from Mike Gattnar of Pavis Capital

Read a good BAC write up from Mike Gattnar of Pavis Capital on SumZero.  With his permission I have posted it below. The format came out a bit funky, but there is a link to a Google PDF of the write up.

Thanks Mike for letting me post this.

Note. The share prices for refers to the 11/29/10 share price which is the date of the write-up. We believe the thesis continues to hold at current prices. Please refer to the attached write-up for the charts.


BAC represents an opportunity to invest in one of the most diversified and attractive U.S. financial services franchises at an attractive valuation: BAC trades at <10x 2010 EPS (2000 – 2009 average of 11.7x) and .85x tangible book value per share (2000 – 2009 average of 3.0x). The Company provides a broad array of financial services products including: retail banking, mortgages and insurance, credit cards, commercial banking, global investment banking, and wealth management. Additionally, BAC has either a #1 or #2 market position in almost every capability offered to its customers. We believe the strength of the BAC franchise should position the company to generate >15% returns on tangible equity over the long term. Improvements in the credit quality of the loan portfolio should support EPS growth from a 2010E of $1.28 to $2.23 by 2013 (a +20% CAGR). We value BAC at 12x our 2013 EPS estimate of $2.23 per share or $27 (a 137% return or 33% annualized CAGR vs. 11/29/10 closing price of $11.31). We note that an investment in BAC is likely to require a long term investment horizon as it will take years to resolve many of the issues creating an overhang on the stock (e.g. mortgage reps/warranty liability) and for credit quality to stabilize.

Bank of America Corporation provides a diversified range of banking and nonbanking financial services and products through six business segments: Deposits, Global Card Services, Home Loans & Insurance, Global Commercial Banking, Global Banking & Markets (GBAM) and Global Wealth & Investment Management (GWIM), with the remaining operations recorded in All Other. At September 30, 2010, the Company had $2.3 trillion in assets and approximately 284,000 full−time equivalent employees. On January 1, 2009, the Company acquired Merrill Lynch & Co., Inc. (Merrill Lynch) and, as a result, has one of the largest wealth management businesses in the world with over 16,700 financial and wealth advisors, an additional 3,000 client−facing professionals and more than $2.1 trillion in net client assets. Additionally, BAC is a global leader in corporate and investment banking and trading across a broad range of asset classes serving corporations, governments, institutions and individuals around the world.
As of September 30, 2010, BAC operates in all 50 states, the District of Columbia and more than 40 foreign countries. The Company’s retail banking footprint covers approximately 80 percent of the U.S. population and in the U.S., serves approximately 57 million consumer and small business relationships with 5,900 banking centers, 18,000 ATMs, nationwide call centers, and leading online and mobile banking platforms. The Company has banking centers in 13 of the 15 fastest growing states and has leadership positions in seven of those states. The Company offers industry−leading support to approximately four million small business owners.

Investment Highlights
Substantial EPS Expansion: BAC is positioned to grow EPS from a 2010E of $1.28 to $2.23 by 2013 (a +20%). We expect declining bad debt provision expense to contribute $1.24 of incremental EPS during the period as provision expense declines to a more normalized level of 1.7% of average loans vs. 3.0% in 2010. We believe BAC has realized over 80% of the losses on its mortgage loans. BAC’s high exposure to early cycle credit (~18% of loans are credit card), low exposure to late cycle credit (7% of loans are CRE) should allow BAC’s credit losses to decline faster than median large cap banks in 2011 and into 2012. Additionally, we believe there are a number of earnings growth levers which we have not factored into our model:

 Expense reduction: We believe there is meaningful room for BAC to reduce expenses as the credit environment normalizes and it completes the integration of legacy acquisitions (MER). For example the Company believes it can reduce headcount by ~20,000 (this area alone could represent >$1.0B of cost reductions) over the long term and has guided to $7.0B of cost reduction associated with the Merrill Lynch acquisition (3.3B run rate realized at year end 2009). We also believe litigation related expenses are substantial and likely at inflated levels. Management has guided to a 2013 – 2014 expense ratio of 55% - 60% vs. 60% in our model at 2013. A 55% expense ratio would represent an incremental $4B in annualized earnings power, $0.45 in incremental EPS, or an additional $5.43 of shareholder value at a 12x earnings multiple. We note that expenses may actually increase in the short term due to heightened litigation associated with rep/warranty put-backs and the “robo-signing” scandal.
 Loss mitigation efforts: Our model assumes a $3B adverse impact as a result of regulation E and the Durbin amendment. While we have assumed the full impact of these adverse regulatory changes, a number of analysts believe BAC’s loss mitigation efforts (revised customer pricing, new products, cost reduction) may mitigate as much as 50% of the lost revenue.
 Long term debt reduction: BAC inherited a substantial amount of high cost long term debt as a result of legacy transactions. Management believes it can reduce long term debt by $150B - $200B over the next several years or an incremental $75B - $150B vs. what is reflected in our models. This could add an incremental $0.08 - $0.15 to 2013 EPS or $0.92 - $1.83 of value.
 Reduced level of mortgage put-backs: Our model assume $3.5B - $4.1B of mortgage put-back expense in the 2010 – 2013 period. We would expect mortgage put-back liability to substantially diminish over the long term. The ongoing cost of mortgage put-backs represents $0.31 of earnings power in 2013 or $3.74 of potential incremental value per share.
EPS Bridge: Refer to chart
Premiere Franchise:  BAC is a market leader across almost every service it provides to its customers. While the breadth of BAC’s service offerings provides a level of diversification to the earnings stream, the strength of the Company’s competitive position is an asset we believe will enable to Company to capture share going forward and generate excess returns over the long term. For example, BAC’s retail banking franchise (~50% of revenues) has the largest deposit base; lowest rate paid and lowest expenses per dollar deposit. We believe this is a source of competitive advantage as it implies a lower marginal cost than competitors on the bank’s core source of funds and provides a level of stability to BAC’s balance sheet. We expect BAC will emerge from the credit cycle a dominant player in domestic retail broking and international investment banking, and mortgage servicing. The slides below summarize the scope and competitive position of BAC across its major customer facing segments.
Refer to write-up for charts
Improving Credit Quality and De-risked Balance Sheet: BAC has de-risked its balance sheet with over 80% of estimated lifetime losses realized thus far through the cycle. This compares favorably vs. analyst estimates of only ~55% for the banking sector and ~70% for the large-caps banking sector. The Company has also seen reduced delinquencies across all major credit categories and is experiencing declining charge-offs across both consumer and commercial segments (refer to charts below).

Cross Selling Efforts/Opportunity: BAC has recently been investing in sales resources and realigning compensation structures in an effort to drive synergies from the Merrill Lynch acquisition and cross sell the full breadth of its services to its client base. We believe this represents an opportunity for the Company to drive incremental revenue across high net-worth clients, institutional clients, and international clients. We note that the Company has already begun to experience share gains across each of these client segments. While we believe these efforts will drive incremental share gain and accelerate growth, we have not (and believe the analyst community has not) given BAC any meaningful credit for these initiatives.

REFER to Write-up for charts

Regulatory Risk:
 Regulation E: In November 2009, the Federal Reserve issued amendments to Regulation E, which implements the Electronic Fund Transfer Act (Regulation E). The new rules have been in place since July 1, 2010. These amendments change the way banks may charge overdraft fees; by limiting the ability to charge an overdraft fee for ATM and one- time debit card transactions that overdraw a consumer's account, unless the consumer affirmatively consents to the bank's payment of overdrafts for those transactions. Under previously announced plans, BAC does not offer customers the opportunity to opt−in to overdraft services related to non−recurring debit card transactions. However, customers are able to opt−in on a withdrawal−by−withdrawal basis to access cash through the Bank of America ATM network where the bank is able to alert customers that the transaction may overdraw their account and result in a fee if they choose to proceed. The impact of Regulation E in the third quarter was a reduction in service charges of approximately $375 million pre−tax. The 2010 full−year decrease in revenue related to the implementation of Regulation E and the impact of overdraft policy changes is expected to be approximately $1 billion after−tax. The Company has implemented a number of changes in an attempt to mitigate the impact of Regulation E including pricing accounts on an activity basis and offering customers new products.

 Debit Interchange Rules: On July 21, 2010, the Financial Reform Act was signed into law. The legislation, which provides the Federal Reserve with authority over interchange fees received or charged by a card issuer, requires that fees must be “reasonable and proportional” to the costs of processing such transactions. In addition, the legislation prohibits card issuers and networks from entering into exclusive arrangements requiring that debit card transactions be processed on a single network or only two affiliated networks, and allows merchants to determine transaction routing. Management believes that debit card revenue will be adversely impacted beginning in the third quarter of 2011 and estimates approximately $2.0 billion of lost annual revenue as a result of the legislation. The Company recorded a non−tax deductible goodwill impairment charge for Global Card Services in the three months ended September 30, 2010 of $10.4 billion. The Company expects to implement a number of actions that will mitigate a good portion of the impact when the laws and regulations become effective.
 Card Act: On May 22, 2009, the Credit Card Accountability Responsibility and Disclosure Act of 2009 (CARD Act) was signed into law. The majority of the CARD Act provisions became effective in February 2010. The CARD Act legislation contains comprehensive credit card reform related to credit card industry practices including significantly restricting banks' ability to change interest rates and assess fees to reflect individual consumer risk, changing the way payments are applied and requiring changes to consumer credit card disclosures. Under the CARD Act, banks must give customers 45 days notice prior to a change in terms on their account and the grace period for credit card payments changes from 14 days to 21 days. The CARD Act also requires banks to review any accounts that were repriced since January 1, 2009 for a possible rate reduction. The provisions in the CARD Act have negatively impacted BAC’s net interest income, due to the restrictions on BAC’s ability to re-price credit cards based on risk, and card income due to restrictions imposed on certain fees. The bill went into effect on February 22, 2010, nine months after it was enacted. The 2010 full−year decrease in revenue is expected to be approximately $1 billion after−tax.
Mortgage Putbacks: The most significant and difficult to quantify risk facing BAC relates to the representations and warranties made with regards to loans it previously sold to investors. Generally, an investor can put back the mortgage to the originator for the full unpaid principal amount if the loan contains an origination defect or was fraudulently originated and has defaulted. We estimate cumulative put-back losses of $15.8B (~$3B of losses have already been incurred). For modeling purposes we use the Morgan Stanley cumulative loss estimates of $21B. For context BAC has a reserve of $4.4B and estimates $750M of quarterly expenses over the next few years (our model assumes $3.5B - $4.1B through the 2014 period per Morgan Stanley estimates). Below we provide our estimates and Morgan Stanley’s estimates of the mortgage put-back liability, a summary of the different classes of investors that have an ability to put back loans to BAC, and BAC’s experience with each class of investor.

 GSE: From 2004-08, BAC/CFC sold $1.2 trillion of loans to the GSEs. It has received $18.0 billion of repurchase claims on lot (1.5%). It resolved $11.4 billion (63%) of these claims with a loss experience of 22% ($2.5B). The areas of greatest dispute have been the reasonableness of stated income, occupancy, and undisclosed liabilities. Its reserve for the GSE reps and warranties exposure is computed to cover both the existing pipeline of claims and a projection of future claims it might receive on loans that have already defaulted, and on future defaults predicted by its loss-forecast models. It expects its repurchase rate with the GSEs to increase. Still, it estimates repurchase claims to date represent more than two-thirds of claims expected on 2004-2008 vintages.
 Private Label Securities wrapped by monolines – Note, each of these reps and warranties counterparties has different contractual rights. It sold $160 billion of loans into monoline-wrapped transactions. This included $73 billion of first mortgages (33% paid off) and $87 billion in second-lien (60% paid off). In addition, of the first liens sold, it estimates $38 billion were sold as whole loans to other institutions, which subsequently included these loans with those of other originators in private label securitization deals in which the monolines typically insured one or more tranches. To date, $4.8 billion (3%) of repurchase claims have been received. Of this, $4.2 billion remain outstanding of which $2.7 billion has already reviewed and declined to repurchase and $550 million approved for repurchase (looking at the remaining $1.5B). It noted it has had limited engagement with most of the monoline insurers in its repurchase process, which has constrained its ability to resolve the open claims. Also, certain monoline insurers have instituted litigation against CFC and BAC, which further constrains a normal business relationship. Without this engagement, it believes it is not possible to reasonably estimate its future repurchase experience and the liability that may exist in connection with these securitizations. Still, it has used its experience with a subset of the monoline universe as a basis for computing a reserve on existing and future claims.
 Whole Loan Investors/Private Label Securitizations - BAC stated that it believes many of the losses observed in these deals have been, and continue to be, driven by external factors, like the substantial depreciation in home prices, persistently high unemployment and other economic trends, diminishing the likelihood that any loan defect, assuming one exists at all, was the cause of the loan’s default. From 2004 through 2008, BAC and CFC (mostly) sold $750 billion of loans (40% paid off). It has received $3.9 billion of repurchase claims ($2.9B resolved), with $1.0 billion remaining outstanding (of which $0.5B already reviewed and declined to repurchase). Approximately $1.0 billion has been approved for repurchase. Many of the claims that it has received so far are from whole loan investors. With respect to claims from private securitizations it acknowledged if it were to get a better sense of its experience with this segment, it would increase its reserve. We estimate that BAC is about >20% through its cleanup. We have baked in our estimates for remaining reps and warranties costs of $16 billion in our model through 2014 . This is higher than what management is guiding towards (it suggests an average of $750 mill per quarter, with volatility likely between $500mill to $1bill, no time frame given on end dates).
o BAC’s shares fell 10% relative to the S&P 500 the week of 10/19/10 based on news a group of investors including Blackrock, PIMCO, Wamco and the NY Fed among others publicly announced a notice of Non-Performance to Countrywide Servicing (Master Servicer) for its servicing of 115 loan pools totaling $47 billion. We discuss this group of private loan investors separately given the size of the litigation and the recent impact on the share price. The investors claim that Countrywide Servicing is not adhering to its PSA agreements. Assuming our estimates for the non-agency assets (30% of the portfolio delinquent or in loss, a weighted average success rate of 33% and a 60% weighted average severity), we estimate BAC's cost of remedying losses on these pools would be $2.8 billion (implies an incremental $2B liability vs. our $6.7B estimate for private loan losses). If we use BAC's experience to date, this implies $176 million. If we double BAC's severity and defect rate, then we get to $353 million. We believe the potential for loss against these pools is captured in the Morgan Stanley $16 billion future reps and warranties costs factored into our model through December 2014.
While it is difficult to estimate the liability associated with the private and mono-line securities, we expect the ultimate liability will be relatively lower than that experienced with the GSE’s. Private label investors do not benefit from the same degree of protection as the GSE’s and there are a number of logistical impediments to centralizing claims which should mitigate losses relative to what has been experienced with the GSE securities:
 GSEs benefit from direct access to the loan files, while private label investors must aggregate claims to request a review of loan files (typically must own 25% of the MBS).
 GSE’s need to demonstrate only that a breach in their conforming underwriting standards has occurred. Private investors must also demonstrate that the breach drove the loss.
 GSE’s have lower hurdles for breaches of representations and warranties due to the relatively higher standard of loan underwriting.
 Private label securities often lack some of the representations and warranties common to agency securities.
 The length of time a loan performs prior to a default is an important consideration as well. The longer a loan performs, the less likely an underwriting reps and warranties breach would have had a material impact on the loan’s performance or that a breach even exist.
 The terms of private securitizations vary substantially. Thus, we would expect not broad classes of settlements but a multi year- long, loan-by-loan fights in court, suggesting a lower request rate and success rate than was the case with the GSEs.
 The rate of put-backs associated with private loans should increase as lawyers and investors become better organized and are better able to coordinate access to loan files.
 The credit quality of the private loans is lower than the GSE loans implying a greater incentive on the part of private investors to put back the loans and greater severity in the case of successful put-backs relative to the GSE experience.
 Issues related to the robo-singing scandal or foreclosure moratorium could embolden private investors to pursue litigation (although we do not expect judges to view private investors in the same light as homeowners) or increase the success rate of putbacks.
However, even if the liability is higher than the estimates we have modeled, it is likely to be spread over a number of years and therefore the likely impact on capital will be muted as a result. During these years we expect BAC to continue to churn out $10+ billion in earnings and as net charge-offs continue to trail down the losses from those claims should be absorbed.
Foreclosure Moratorium: In 23 US states, the foreclosure process requires the lender to prepare an affidavit containing details of the mortgage, and to sign it in front of a notary who has to verify the information in the affidavit. The bank then has to approach a court and obtain an order before it can commence foreclosure proceedings. However, a number of instances have been discovered recently where a large number of affidavits were notarized by a small set of individuals (the so-called robo-signers) without any verification of documents. As a result, a number of banks have halted foreclosure proceedings in these states where a court order has to be obtained in order to strengthen their operations on this front. On October 1, 2010, BAC voluntarily stopped taking foreclosure proceedings to judgment in states where foreclosure requires a court order following a legal proceeding. On October 8, 2010, BAC stopped foreclosure sales in all states in order to complete an assessment of the related business processes. These actions did not affect the initiation and processing of foreclosures prior to judgment or sale of real estate owned properties.
BAC has completed its assessment of the foreclosure affidavit process in the 23 states where foreclosure requires a court order following a legal proceeding. As a result of that review, BAC has identified and is implementing processes to ensure that affidavits are prepared in compliance with state law and have begun a rolling process of preparing and resubmitting, affidavits of indebtedness in pending foreclosure proceedings in order to resume the process of taking these foreclosure proceedings to judgment in those states. BAC estimates this process of resubmitting affidavits will take at least several weeks and could involve as many as 102,000 foreclosure proceedings that were pending as of October 1, 2010. Once these affidavits are resubmitted, there may be prolonged adversary proceedings that delay certain foreclosure sales. BAC continues to assess its processes in the other 27 states and intends to implement enhancements as appropriate. Subsequent to its announcements that it was temporarily suspending foreclosure sales, law enforcement authorities in all 50 states and the United States Department of Justice and other federal agencies have stated they are investigating whether mortgage servicers have had irregularities in their foreclosure practices. Pavis Capital believes the “robo-signing” scandal could trigger a number of risks/challenges for BAC and the banking sector as a whole:
 Bad precedence in some court cases could drive foreclosure buyers away and they could potentially start demanding a much higher discount for distressed sales to counter the remote risk of losing the home at some later time.
 The issue has exacerbated political risk. There have been calls to impose nationwide foreclosure moratoria and some concern that attorneys general could use this investigation to try and accelerate loan modifications by banks.
 The investigations by the attorneys general could result in fines, penalties or other remedies and result in significant legal costs.
 Additional borrower and non−borrower litigation and governmental and incremental regulatory scrutiny (for example class action suits related to foreclosures or increases in mortgage put-back related litigation).
 Costs will likely increase in the fourth quarter of 2010 and will continue into 2011 as a result of the additional resources necessary to perform the foreclosure process assessment, revise affidavit filings and implement operational changes. A congressional oversight panel estimated that incremental costs from foreclosure irregularities could range from $1.5 to $10.0 billion for the entire industry.

 The time to complete foreclosure sales may increase temporarily, which may result in an increase in nonperforming loans and servicing advances and may impact the collectability of such advances and the value of BAC mortgage servicing rights asset.
The worst-case scenario is considerably grimmer. In this view, which has been articulated by academics and homeowner advocates, the “robo-signing” of affidavits served to cover up the fact that loan servicers cannot demonstrate the facts required to conduct a lawful foreclosure. In essence, banks may be unable to prove that they own the mortgage loans they claim to own. The risk stems from the possibility that the rapid growth of mortgage securitization outpaced the ability of the legal and financial system to track mortgage loan ownership.
At this stage the “robo-signing” scandal appears to be a largely a procedural matter since there is little debate around whether the borrowers of these properties have actually defaulted. We believe that the number of loans with documentation issues is not likely to be material and the impact is likely to be limited to a delay in the charge-off process and potentially higher operational costs to strengthen the back end. BAC anticipates over the course of this pause, less than 30,000 foreclosure sales will have been delayed. As was the case for its judicial state review, BAC’s initial assessment findings show the basis for our foreclosure decisions is accurate. It noted in 3Q10, 80% of borrowers had not made a mortgage payment for more than one year (on average borrowers were 18 months delinquent), and 33% of properties were vacant.

Impact of Long Term Bond Compression: Compression in long bond is a challenge for all banks as it drives lower reinvestment rates in RMBS and, in first lien mortgages (less because there are more adjustables in bank-owned portfolios). These are the two asset classes most impacted by the decline in the long bond. BAC’s residential mortgage exposure is only slightly above the median for both of these metrics for large cap banks. BAC has been able to drive a fairly consistent NIM in volatile rate environments over the past several years. Part of this is changing business models and part of this is changing investment strategy. We expect that BAC will be able to mitigate some of the yield curve pressure on its NIM through reducing higher cost of funds as its balance sheet shrinks (management has suggested a $95 billion long-term debt decline at 3% swap adjusted cost of funds yielding a 15bp positive impact, all other things equal, or 5% to NIM over next 10 quarters), reinvesting in AFS loans rather than securities, reinvesting liquidity as Basel 3 rules known. However, the mitigants are over time, while the sharp decline in the long bond will negatively impact NII over the next few quarters. We are modeling a q/q decline in NII of 1% and 4% in 4Q10 and 1Q11 driving a $150 million and $500 million sequential decline in NII. We note that BAC has been able to maintain fairly stable NIM levels over a long time series across multiple cycles.

Trading Profits: BAC generates ~15% of pre-provision revenue from trading related activities. We believe there are a number of risks associated with this revenue stream: 1) lost revenue as a result of the implementation of the Volcker rule, 2) we believe the revenue associated with proprietary trading activities to be of lower quality, 3) both proprietary trading and client driven revenue are volatile and difficult to predict. While we have not been able to disaggregate the proportion of revenue that is driven by proprietary trading vs. client revenue, commentary by management and press releases on trading related headcount suggests that the vast majority of trading related profits is linked to the facilitation of client trades. Additionally, any reduction of proprietary trading revenue associated with the Volcker rule would likely be mitigated via the sale or redeployment of the balance sheet capital used to drive the revenue. While we believe the client driven portion trading revenues is also volatile and difficult to predict, we believe the risk/reward associated with the BAC investment more than compensates us for that volatility. The 2 charts below summarize the risk associated with BAC trading business:
The histogram below is a graphic depiction of trading volatility and illustrates the daily level of trading−related revenue for the three months ended September 30, 2010 as compared with the three months ended June 30, 2010 and March 31, 2010. During the three months ended September 30, 2010, positive trading−related revenue was recorded for 100 percent of the trading days of which 89 percent were daily trading gains of over $25 million. This can be compared to the three months ended June 30, 2010, where positive trading−related revenue was recorded for 81 percent of the trading days of which 59 percent were daily trading gains of over $25 million, eight percent of the trading days had losses greater than $25 million and the largest loss was $102 million. For the three months ended March 31, 2010, positive trading−related revenue was recorded for 100 percent of the trading days of which 95 percent were daily trading gains of over $25 million.

The chart below highlights daily trading revenue and VAR over the 9/30/09 - 9/30/10 period:

Mortgage/Credit Demand Contraction: The decline in the housing market and pace of home sales coupled with a broader trend of deleveraging by corporations and consumers has led to an environment of relatively weak loan demand. Lower mortgage/credit demand ultimately limits BAC’s ability to redeploy the capital on its balance sheet into higher yielding assets and could ultimately lead to lower net interest income growth or even a decline in net interest income. We believe both Pavis Capital and equity analysts projections for loan growth are relatively muted and adequately incorporate and more muted loan demand environment.
Litigation: We believe BAC faces heightened litigation risk. While we have discussed the impact of foreclosure and mortgage put-back litigation, we would also like to highlight that BAC is facing a number of class action lawsuits in Florida and California related to overdraft litigation. We will continue to monitor this litigation given the political and media attention that has been paid to overdraft fees and the potential for lawsuits to generate substantial liability.

 Implementation of dividends and share buybacks – Management has stated a desire to begin paying out dividends and buying back shares “as soon as possible. The company has targeted 30% dividend payout and is not planning on making acquisitions going forward. Excess capital would likely be used to repurchase shares.
 Improving credit quality and reduced put-back liability.
 Resolution of mortgage put-back liability overhang.
 Negative short term catalysts include: higher costs associated with robo-signing related issues, short term analyst estimates may be aggressive (particularly in the other revenue category), wiki-leaks release.

Variant View
We generally agree with the sell side consensus. Key issues that could adversely affect the thesis include: mortgage put-back liability being substantially higher than estimated, compression in long term bond yields, and continued stress in the housing market. Please refer to the attache write-up for a more detailed discussion of risk factors.

No comments:

Post a Comment