Description

Hudson City Bancorp Inc. / M&T Bank Corporation

Three years ago this month, Hudson City Bancorp (HCBK) agreed to be sold to M&T Bank Corporation (MTB) for a mix of cash and stock with value equivalent to 0.0840 shares of MTB.  The transaction has yet to receive regulatory approval and the merger agreement has been extended on three separate occasions. These delays have led to a high level of investor fatigue and, we believe, a spread that is very wide relative to the underlying risks.  In particular, we believe the market has failed to recognize how little is at risk should this transaction fail to close. 

Brief History

The transaction was announced on August 27, 2012 and was expected to close in the spring of 2013.  Instead, on April 12, 2013 the companies announced that additional time would be required due to concerns by the Federal Reserve regarding M&T’s compliance capabilities related to the Bank Secrecy Act and anti-money-laundering requirements (BSA/AML).  Subsequent disclosures made clear that, in order to address the concerns of the Federal Reserve, complex systems, policies and BSA/AML processes would need to be created and implemented and the transaction would not be considered for approval until sufficient progress had been made in these areas.

Since then, M&T has spent well over $100 million on improvements to their BSA/AML compliance capabilities and has greatly increased headcount in this area.  The primary systems have been operational for many months and a significant amount of data has been collected to verify their effectiveness.

On March 23, 2015 Hudson City notified shareholders that the companies were planning for a possible closing on May 1, 2015. However on April 6, 2015, the companies announced a further delay and on April 17, 2015 they announced that the merger agreement has been extended until October 31, 2015. Importantly, this announcement disclosed that the companies had been advised by the Federal Reserve that it intends to act on the merger application no later than September 30, 2015.

Arbitrage Opportunity

Under the terms of the merger, HCBK shareholders will receive value equal to 0.840 shares of MTB with 60% of the value in shares and 40% of the value in cash with the cash amount based on an averaging period prior to the close.  As mentioned above, there is an election opportunity between the forms of consideration for which the details can be found in the Merger Agreement.  Adjusting current prices for likely borrow costs and one MTB dividend, the arbitrage spread is approximately $0.65 or 6.5%.  Based on an expectation that the Federal Reserve acts on September 30, 2015 and that the election forms are mailed to HCBK shareholders shortly afterward, we believe a mid-November closing date is reasonable.  If correct, the annualized return is approximately 25%.

We believe that this spread is exceptionally attractive, in part, because we think the probability of the transaction closing in the stated timeframe is quite high and, in part, because we believe that, in the unlikely event that the transaction breaks, the arbitrage outcome will not be severe.  This latter point appears to be a non-consensus view.

Probability of Approval

Any experienced arbitrageur will be able to tell you that bank deals close.  While this transaction has been an extraordinary outlier with regard to the length of its regulatory review, ultimately we believe the outcome should not be an exception.  MTB is an experienced acquirer and management has been good stewards of capital for decades.  Furthermore, they are acquiring a company that will benefit from a greatly expanded product mix and the combined company will be better able to serve the communities within its footprint.

Regarding the BSA/AML issues, we believe that sufficient progress has been made for this transaction to be approved.  This belief is based on our understanding of M&T’s quite limited compliance capabilities at the time at which these issues were first flagged as well as the comprehensive nature of the capabilities that have been put in place and have been operating for some time.  Our view is also bolstered by the actions taken this spring.  While signals clearly got crossed with regard to the status of the merger review, we believe that a fundamental misunderstanding related to M&T’s progress related to BSA/AML requirements is highly unlikely to have been the reason.  Rather, we believe the reason is more likely related to standard issues that are addressed during the normal review of bank deals and that the confusion in the spring may have been related to lapses in communication between the NY branch of the Federal Reserve and those in D.C. 

Some market participants have expressed concern about issues related to the Community Reinvestment Act, but such issues have not resulted in prior transactions being blocked, are rather easily addressed and generally can be cured at little incremental expense.

It is worth noting that while no large bank deals had been approved for a very long time following the economic crisis, this has changed recently.  BB&T’s acquisition of Susquehanna Bankshares was the first to break this trend when it was approved on July 7, 2015.  Subsequent to this, CIT received approval to acquire OneWest Bank on July 21, 2015 and BofI Holdings was given approval to buy H&R Block’s banking unit on August 5, 20015.  While all transactions are reviewed based on the unique facts and circumstances of the relevant companies, we take some encouragement that other deals are now being approved.

Deal Break

We believe that the risk of loss in the event that this transaction breaks is quite limited.  Our view is in part based on our belief that MTB is overvalued on a standalone basis and in part (and this is the controversial bit) based on our belief that the HCBK is not exceptionally overvalued at current prices. 

HCBK’s Deal Break Value

Hudson City is widely perceived to be a broken bank with a broken balance sheet.  This view was amplified by an April 6, 2015 report by Morgan Stanley analyst Ken Zerbe which suggested that the stock could be worth as little as $3 in the absence of a deal.  In contrast, we believe that during the three year pendency of this transaction, very little thoughtful analysis has been done on HCBK and that its value on a standalone basis is far closer to its current stock price. Our rational is presented below:

1.       Hudson City’s earnings power and fundamental value are higher than perceived.  HCBK has $12.2 billion in high-cost term borrowing and $11.6B low yielding assets which combine to greatly depress its margins.   The average cost of these liabilities is 4.59% while the average yield on these assets (overnight funds and U. S. Treasuries) is only 0.26%.  The negative spread associated with these positions depresses HCBK’s eps by approximately $0.65 per year. 

To asses HCBK’s current earnings power, we have eliminated the cost of these liabilities and the interest earned from these assets for the first half of 2015.  We have also eliminated gain-on-sale benefits that have occurred during this period as they are non-recurring.  This analysis results in pro forma H1 eps of $0.31 per share or $0.62 when annualized.  This compares to Bloomberg consensus (non-pro forma) estimates of approximately zero.

Furthermore, during Hudson City’s 3 years in limbo, it has been preparing its balance sheet for the merger rather than focusing on profitable growth.  During this period, it has systematically reduced the maturity and yield of its loan portfolio while allowing deposits to decline for lack immediate use for the funds.  In short, it is significantly under earning relative to its commercial potential.  Given this, we believe that it is fair to value the $0.62 of annualized earnings at a multiple of 15x which results in a value of $9.30 or 7.0% below its current price of $10.00.

2.       The cost of restructuring Hudson City’s balance sheet is misunderstood.  At the time that the merger was announced, the pre-tax cost of retiring HCBK’s high-cost borrowings was estimated to be $2.5 billion and the total after-tax cost of restructuring Hudson City’s balance sheet within the contest of the merger was estimated to be $1.3 billion.  This compares to HCBK’s tangible capital at the time of $4.5B (these figures can be found in MTB’s deal presentation from August 27, 2012).  More recently, in HCBK’s 2015 Stress Test Results the pre-tax cost of restructuring the balance sheet was estimated to be $1.33B which was based on market data from September 30, 2014.  

Importantly, the cost of a balance sheet restructuring can be independently estimated and rolled forward to any relevant time period.  The data for doing so appears in HCBK’s most recent 10-Q and is reproduced in the first table below:

Adjacent to this table we have provided current rates for FHLB loans and have calculated the annual cost difference for each maturity, the cumulative difference per year and the total difference.  For purpose of this analysis we have assumed that a restructuring occurs on December 31, 2015 and that the loans that are being retired all expire at the end of the stated year.

As you can see, the undiscounted cash flows related to the debt retirement total approximately $985 million.  We discount these cash flows at the weighted average FHLB rate and add a small cost to account for the fact that some of the loans are puttable by the lenders (we use $5 million although a recent quote on this structure indicated that the likely cost was less than $2.5 million).  This results in a pre-tax cost of retiring the high-cost liabilities of $965 million and an after-tax cost of $627 million. 

3.       Hudson City’s adjusted tangible book and standalone value is higher than perceived.  HCBK has 529.5 million shares outstanding and $8.82 of tangible book per share.  Reducing tangible book by the per-share after-tax cost of the balance sheet restructuring results in an estimated pro forma tangible book per share of $7.63.

M&T’s 2012 bid for HCBK was priced at 1.2x HCBK’s adjusted tangible book at that time.  Today this multiple suggests a value of $9.16 or approximately 9.8% below HCBK’s current price of $10.16.

Furthermore with $0.62 of pro forma earnings, HCBK would earn a return on pro forma tangible book of over 8%.  In the table below we show a scatter plot of 18 similar-sized US banks and a regression of their price / tangible book and return on tangible book.  We also represent where pro forma HCBK would lie on this chart based on its current price, the 1.2x multiple used by MTB and its position if it were to be valued today based on the regression.  While outside of our thesis, it is worth noting that the latter approach suggests a value of $12.84 per share.

4.       Hudson City is highly over-capitalized before and after a balance sheet restructuring.  Based on its call report for June 30, 2015, HCBK is exceptionally overcapitalized with a common equity tier 1 capital ratio of 32.2% vs. a regulatory minimum of 7.0%. 

In the table below we show the relevant items from this call report as well as the adjustments necessary to calculate the pro forma capital ratios. 

Please note that the relevant capital levels are reduced by the estimated pre-tax cost of the restructuring because bank capital rules exclude deferred tax assets whose value is dependent on future earnings.  Also note that the risk-weighted assets are unchanged by the restructuring while total assets are reduced by the amount of the relevant overnight deposits and U.S. treasuries discussed above as these assets have a zero risk weighting.

Having made these adjustments, in the table below we show HCBK’s current and pro forma capital ratios as well as the regulatory minimums.

To put these capital ratios in to context, we have also examined what the pro forma capital ratios would be if $3 per share of excess common equity tier 1 capital were excluded and we find that HCBK is still significantly above the regulatory minimums.

Since $3+ of excess capital per share is not required to generate the $0.62 of annual pro forma earnings, we believe that it is not unreasonable to view this value as incremental to HCBK’s earnings-based valuation described above.  Doing so would suggest a value for HCBK of $12.30 ($0.62 x 15 + 3 = $12.30) which, as it would happen, is not far from the valuation suggested by the peer group regression shown above.

We summarize our different approaches to valuing pro forma HCBK in the table below.

Please note that our pro forma analysis is not particularly speculative.  Hudson City has made quite clear in their quarterly press releases and regulatory filings that they have been pursuing a dual-track strategy and that the alternative to a transaction would be a balance sheet restructuring.  We think it is also worth noting that the pre cursor to executing a restructuring would very likely be a strategic process to consider the sale of the company to another party.  This is because the tax losses related to the restructuring could be utilized by a buyer if the restructuring were to occur after a transaction closed but would be lost to the buyer if the restructuring occurred first.

Also, please note that our valuation methodology is not intended to assess where HCBK is likely to trade immediately following a deal break announcement.  Rather it is intended to assess the likely price of HCBK after the liquidation of arbitrage positions that generally occurs following a deal break.

MTB’s Deal Break Valuation

We believe that MTB’s current valuation incorporates meaningful deal accretion and a premium multiple that would be diminished in the event that this transaction is blocked. While consensus earnings are $9.24 for 2016, we estimate that at least $0.60 of this is due to expected accretion from the transaction. Thus on a standalone basis MTB is trading at approximately 15.0x 2016 standalone earnings of approximately $8.64.  This valuation represents a premium of nearly 25% to peers such as KEY, FITB and RF who are currently valued at approximately 12.1x 2016 consensus earnings.

The table below provides our estimated price targets for MTB in the event of a deal break for a range of multiples.

While MTB has historically traded at a premium to this group*, we believe that this premium would be diminished if they are unable to make significant acquisitions as M&A has been an important component of MTB’s historic growth profile.  With this in mind, we believe that a premium of 1 PE point to the peer group is a fair-to-generous valuation for MTB in the event that this transaction breaks. If correct, MTB is likely to have as great or greater fundamental downside in the event that the transaction breaks than does HCBK.

* For the two years prior to the announcement of the HCBK / MTB transaction, MTB traded at an average forward PE multiple that was 1.8 higher than this peer group.

Conclusion

We believe that the Hudson City / M&T merger arbitrage provides a highly compelling risk reward in which it is highly likely that the transaction will close and arbitrage investors will earn a high IRR.  In the unlikely event that the transaction breaks, we believe that investors face very little risk of a permanent loss of capital and believe that patient investors could quite conceivably earn a profit in either outcome.

Catalyst

Merger closes ... or it doesn't