Description

Banca Monte Dei Paschi FRESH Convertible Bonds (L+425bps spread; due 2099) (ISIN: XS0357998268)

SUMMARY THESIS

• The FRESH Convertible Bonds (FRESH Bonds) currently yield 21% to final conversion in 2099 based on coupon payment resumption in April 2018
• The bonds should have significant upside if the bank is able to return to profitability and resume dividend payments.  A reasonable risk-adjusted base case implies a value in the mid-40s for the bonds, which represents in excess of 100% upside to the current price of 18.5.

COMPANY BACKGROUND

Banca Monte Dei Paschi Di Siena (MPS) is the oldest continuously operating bank in the world, tracing its origin to 1472.  Until 1995, the bank operated functionally as a not for profit, using the profits from its banking activities to fund charitable and social projects in the Siena region.  In 1995, the structure was modified such that the bank became a for-profit enterprise which was wholly-owned by a not-for-profit entity called the Fondazione Monte Dei Paschi di Siena (Fondazione).  Other “mount of piety” banks in Italy underwent similar restructurings into for-profit banks and foundations.  In 1999, MPS went public, and over time the Fondazione’s stake was reduced.

In 2007 MPS agreed to purchase Antonveneta for €10.3 billion from Banco Santander – who had only just recently acquired the bank itself as part of the break-up bid for ABN Amro.  MPS partly funded this purchase through a combination of a €4.1 billion equity rights offering and the €1 billion offering of the FRESH convertible bonds.

MPS suffered extensively during the global financial crisis of 2008 and the ensuing European sovereign debt crisis of 2011.  A brief summary of the problems include:

• Poor loan portfolio performance, in particular the portfolio acquired from Antonveneta
• Large exposure to Italian sovereign debt
• Off balance sheet transactions (Santorini and Alessandria) designed to hide losses and shore up capital which were subsequently exposed and created further losses
• The lack of willingness on the part of the Fondazione to raise equity due to its own financial challenges
• The close relationship between the Fondazione, bank management and politicians resulting in both loans to politically connected entities as well as a lack of willingness to cut costs by reducing headcount

MPS’s financial issues lead to a €1.9 billion bail-out in the form of the so-called Tremonti bonds[1] by the Italian government in June 2012.   This was increased to €3.9 billion in December 2012.  Under European State Aid rules, this required approval by the European commission, which gave their approval of a restructuring plan in late 2013. 

Pursuant to the EU-approved restructuring plan, MPS conducted a €5 billion rights offering in June of 2014.  €3.5B of the proceeds were used to redeem the State Aid Tremonti bonds, and the remainder was used to shore up MPS’s capital position ahead of the ECB Asset Quality Review in October 2014.

Due to weak coverage of non-performing loans, MPS showed the largest gap in the ECB’s stress test of any bank, with a capital deficit of €1.9 billion in the stress case.  As a result of the stress test outcome, MPS took two significant actions: 

1. Provisions: Recorded €5.4 billion of loan loss provisions in the 4th quarter of 2014 to better align the loan portfolio’s carrying value with the results of the ECB’s Asset Quality Review; and
2. 2nd Rights Offering: in May 2015 a fully underwritten €3 billion rights offering was launched.   €1.1 billion of the rights issue proceeds was used for repayment of the remaining Tremonti bonds and the remainder was used to bolster MPS’s capital position

Both of these actions are fundamentally positive for the FRESH bonds.  The large loan provisions that have been recorded reduces the necessity for future loan loss provisions and brings forward the date from which MPS will be profitable.  This is important because MPS must be profitable before it can recommence paying coupons on the FRESH bonds (discussed in more detail below).  The large equity raise is positive because it increases the quantum of capital that is “junior” to the FRESH bonds (i.e. increasing MPS’s loss absorbing capacity before the FRESH Bonds would be fundamentally impaired). 

FRESH BONDS – INSTRUMENT OVERVIEW

Some background and in particular certain specifics regarding the FRESH instrument and documentation is important in order to understand why the investment opportunity exists (as well as the risks of the investment / the downside case).  Documentation issues represent both the opportunity and the risk for this investment.    

In summary, the structure of the FRESH Bonds is unusual because these bonds are not directly issued by MPS.  In simple terms the economic substance of the structure results in the FRESH bonds being subordinated junior capital instruments of MPS and convertible into shares of MPS. 

More details are as follows.  The bonds are issued by Bank of New York Luxembourg on a fiduciary basis, meaning that the only recourse of the FRESH bondholders is the specific assets that underlie the bonds.  In this case, the proceeds of the bond offering were invested to make the upfront payment on a swap agreement with JPMorgan. Via this swap agreement the bondholders receive the rights to exchange into the underlying shares which were issued from MPS to JPMorgan in the original 2008 transaction.  These shares are subject to a usufruct agreement which allows Monte Dei Paschi to receive all dividends on these shares and to vote these shares.  In exchange for these rights, MPS agrees to make a floating payment of L+425 on a quarterly basis in periods in which the bank is both profitable AND pays a dividend.  The current accounting treatment for the bonds is that the initial shares issued as the exchange property are included in Core Equity Tier 1 capital for MPS (80% of the principal) and the remaining portion of the principal value (20%) is included in Additional Tier 1. 

Amendment of Terms
It is important to highlight an amendment that was made to the terms of the FRESH bond in March 2009.  The importance of this amendment is twofold.  Firstly, the changes made to the features of the bond are important in their own right.  Secondly, the changes are important because of their relevance to the potential future treatment of these securities.

By way of background, the unusual (and convoluted) structure of the FRESH Bonds was constructed in an attempt to gain Core Equity Tier 1 capital treatment for the underlying shares and Tier 1 treatment for the entire transaction.  This was done at a time where the final terms that regulators would require for Basel III compliant capital were not yet clear.  The original terms of the FRESH Bonds required that the coupon be paid if the bank was profitable OR paid a dividend.  This was similar to the language in many legacy Tier 1 instruments. 

MPS became aware that the Bank of Italy, its regulator, did not believe that the original transaction structure would allow the initial shares to be included in core capital.  This was a result of the way the Bank of Italy guided Unicredito to structure its very similar and larger CASHES transaction.  Specifically, the Bank of Italy required that in order for the instrument to qualify for core capital inclusion it could only pay a coupon is the bank was profitable AND paid a dividend.  MPS approached FRESH bondholders seeking an amendment to the terms of the bonds to make the instrument complaint with the guidance from the Bank of Italy.  In order to encourage bondholder participation MPS claimed to FRESH bondholders that there was a high likelihood of an “Increased Burden Event” if the amendments were not made and this event would lead to an Early Automatic Exchange of the bonds (more details below).  MPS convened a meeting of bondholders to amend the terms of the bonds, in particular the coupon payment terms.

The amendment was adopted by FRESH bondholders, in part because the Fondazione owned a majority (just over 50%) of the issue.  There were dissenting bondholders led by Jabre Capital who argued that there had not been any increased burden event and that therefore the proposed changes were unnecessary.  Furthermore, these bondholders argued that the Fondazione supported this transaction to benefit itself as a shareholder (and prevent the need for additional equity) rather than because there had in fact been an Increased Burden Event.  These bondholders retained legal counsel and engaged with MPS. 

After this vote, JPMorgan and Bank of New York asked MPS to indemnify them for any claims from the dissenting bondholders.  MPS did so. 

There are at least two important takeaways from the modification of the terms and the events surrounding that modification:

• Firstly, despite direction from the regulators that the initial shares could not continue to be counted as core capital, MPS did not declare an Increased Burden Event but instead gave FRESH bondholders the option to modify terms to avoid the (possible) Increased Burden Event.  This is relevant regarding MPS’s potential behaviour in future; 
• Secondly, there was significant doubt on the part of FRESH bondholders (advised by legal counsel) that these circumstances constituted an Increased Burden Event.  The behaviour of JPMorgan and BoNY, who both asked MPS to indemnify them is further evidence of this.  It is likely that any future attempt by MPS to declare an Increased Burden Event as a result of regulatory treatment would be met with extensive and costly litigation from bondholders.

Automatic Exchange
Since the FRESH bonds were issued in 2008 MPS’s stock price is down in excess of 95%.  Additionally, the rights from the two MPS rights offerings completed by MPS in 2014 and 2015 (described above) were distributed to holders of the FRESH bonds (in accordance with the terms of the document).  As such, the underlying shares subject to the usufruct agreement which represent the Exchange Property for the FRESH bonds are now worth close to zero.  This is important because it means that any event which results in an Automatic Exchange of the FRESH bonds will result in the bonds becoming worthless (i.e. by being forcibly exchanged into the underlying shares recovery on the FRESH bonds will be close to zero).

As stated above this documentation issue is the biggest risk to these securities.

The terms under which an Automatic Exchange can happen are covered in section 5 of the Terms & Conditions of the FRESH Bonds.  Running through each in turn:

• 5(a) Automatic Exchange due to Share price: This would happen if parity on the bonds went to 150% of the face of the bonds.  This is functionally impossible and the term is irrelevant.
• 5(b) Early Automatic Exchange following an Event of Default:  This would be triggered if either JPMorgan or BoNY defaulted (irrelevant) or if MPS defaulted.  If MPS defaulted it is unlikely that recovery on the FRESH Bonds would be materially different from other MPS subordinated bonds.  Therefore this term does not represent any additional risk over other subordinated debt (note: a comparison of the FRESH Bonds relative to other MPS bonds is discussed in the valuation section)
• 5 (c) Early Automatic Exchange following a Capital Deficiency Event of the Company:  This would happen if MPS’s total capital ratio on either a consolidated or non-consolidated basis fell below 5%.  Pragmatically, MPS would never reach this level without either raising common equity in a rights offering (as it has already done twice recently, once in 2014 and once in 2015) or being bailed out by the government.  If the bank is able to raise equity capital then this term is irrelevant.  If the bank is bailed out by the government, it is unlikely that recovery on the FRESH Bonds would be materially different from other MPS subordinated bonds.  While less clear than 5(b), I believe the same conclusion applies, namely this term does not represent any additional risk over other subordinated debt
• 5 (d) Early Automatic Exchange on Non-Equity Offer:  This condition needs to be read in accordance with section 9(a).  This is intended to deal with a situation where a cash offer is made for the shares of MPS.  Essentially JPMorgan is required to use the proceeds to purchase other shares (ideally shares of the acquirer) to maintain the interest of bondholders.  However, if this is not possible for some reason and / or the Swap Agreement becomes unenforceable the FRESH bonds would convert early.  This represents the M&A risk that some market participants are concerned about.  

JPMorgan (credit research) has suggested that the bonds could be converted due to any acquirer making a “claim” that the Swap Agreement is not enforceable.  This is clearly not the intention of the term from any reasonable reading of the clause (i.e. that the bonds can be forcibly converted by any acquirer).  The comments from Italian lawyers familiar with this transaction agree with this interpretation: 

As a general consideration, in an M&A scenario, if BMPS SpA is merged into a buyer entity, the “Company” definition would transfer to the buyer’s entity. Merger between companies is to be construed as a “succession” of the merging company into all rights and obligations of the merged company.  Hence, the merger as such should not trigger automatic conversion.  This conclusion seems confirmed also under Condition 14I,(ii) which provides as an event of default of the Bond, inter alia, the fact that BMPS ceases to carry on its business or operation, “except for the purpose of and followed by a … merger or consolidation”.

More specifically, the event specified under conditions 5(d) and 9(a) is substantially triggered if, further to the receipt by JPM of an offer of exchange of BMPS securities, the relevant offer cannot be implemented and/or completed in any of the manners indicated therein. Also, as highlighted in your email, the relevant event is triggered if, as a result of the offer, the swap is no longer enforceable or the person making the offer claims that the swap is no longer enforceable. In the latter case, the “claim” should in any case be a consequence of the offer being made (i.e. “as a result of the offer”), which gives some kind of “materiality” to the event and would probably exclude a simple “claim”.  

Even if the courts ultimately agreed that a simple “claim” of some sort could result in an Early Automatic Exchange, this would only happen after costly litigation (a disincentive for an acquirer to pursue this path).  Also, it would require a healthy, solvent bank deciding to behave extremely aggressively versus subordinated creditors, something we have never seen in Europe and which seems quite unlikely.  It is important to note that any M&A event that did not result in an Automatic Exchange would likely be very credit positive and as such this must be balanced against any potential documentation risk.  On an expected value basis, I believe M&A would be positive

• 5 (e) Early Automatic Exchange following an Increased Burden Event of a Tax Event:  This has historically been the term that has most vexed market participants.  The concern is that a change in the regulatory capital treatment of the instrument would result in an “Increased Burden” for MPS and therefore MPS could force an Automatic Exchange.  There are two main concerns:

1. The terms of the FRESH bond coupon which require payment if MPS is profitable and pays a dividend provide for some circumstances where MPS, is forced to make a coupon payment to FRESH bondholders.  This is more stringent that other Additional Tier 1 capital.  Specifically, subsequent development of the allowed terms of Additional Tier 1 capital do not mandate any circumstances in which a bank can be required to pay a coupon on AT1 instruments.  Therefore, some market participants are concerned that the regulator could mandate a change in capital treatment of the FRESH bonds; and
2. The ECB has recently taken over as the single regulator for all of the major banks within the Eurozone.  Last year’s Asset Quality Review was intended to provide consistency of treatment for assets across the single currency area, but did not address capital treatment.  Although the capital treatment of the FRESH bonds has been settled by the Bank of Italy, the ECB may take a different interpretation. 

As described above in 2008, MPS took the interpretation that the change in regulatory treatment could result in an Increased Burden Event.  However, there are two important factors to consider if this regulatory change circumstance presents.  Firstly, as discussed above in 2008 MPS did encourage FRESH bondholders to amend the terms of the instrument to ensure that the instrument had continued compliance for regulatory capital purposes.  Secondly, there remains substantial doubt regarding whether or not an Increased Burden Event would take place even in the event of a change in regulatory treatment.  Commentary from the same Italian lawyers as above:

A different capital treatment by the ECB of the instrument would in principle fall within the concept of an “increased burden” (e.g. a change in the official interpretation of the laws/regulations by the ECB or a change in the relevant law/regulations). However, the Increased Burden Event would only be triggered upon receipt by the Issuer or JPM of an opinion of a nationally recognized law firm stating that, as a consequence of such change of interpretation or change of law, there is more than an insubstantial risk that the Issuer, the Counterparty (or any of their affiliates) or BMPS is or will be subject to more than a de minimis amount of burden or cost in relation to their relevant obligations under the instruments or the swap agreements. In our experience, it is very unlikely for a law firm to release such type of opinion due the amount of liability connected thereto.  Also, the fact that the amount of burden or cost must be related to “their relevant obligations under the instruments or the swap agreements”, rather than more generally, makes it less immediate to capture, for example, the simple need to replace the FRESH instrument with other instruments.

While the possibility of an Increased Burden Event cannot be ruled out, it appears low probability.  The reason for this is because in order for this to occur all three of the following would need to occur:

• The ECB decides to change the capital treatment of this transaction (i.e. a different interpretation to that taken by the Bank of Italy).  Importantly, any ECB action would also impact the much larger €3B Unicredito CASHES transaction, which has the same structure as the FRESHES bonds.  Given the desire for stability of the financial system it is difficult to see the incentive for the ECB to take an aggressive interpretation that results in these banks needing to come to market to raise further equity capital;
• MPS would need to be able to legally trigger the Increased Burden Event due to the change in capital treatment and have this be upheld by the courts.  Even if MPS ultimately prevailed it is virtually certain there would be costly and lengthy litigation beforehand; and
• MPS would need to decide that it is in its interest to trigger the Increased Burden Event and impose losses on subordinated creditors.  Given the bank’s potential future need for Additional Tier 1 capital, doing itself significant reputational damage in the market seems irrational, even if they were legally able to do so.

 It appears very unlikely that all three of these events will transpire.

• 5 (f) Automatic Exchange at Maturity: The bond does not mature at par at the maturity in 2099, rather bondholders receive the underlying shares.  This should be modelled appropriately in thinking about the potential IRRs / fair value.  But, given the discount rates, the net present value of the bonds at maturity in 84 years is effectively meaningless.

VALUATION

The simplest way to think about the valuation of the FRESHES bonds is the implied yield or IRR on the bond for the remaining coupons.  If MPS is profitable for 2015 (in line with company guidance sell-side coverage expectations) and pays a dividend on those profits (which is highly unlikely) – then the yield at the current price of 19 is approximately 29%.  Depending on for which fiscal year the bank first declares dividends, the yield is as follows: 

However, a better way to think about the value of the FRESH bonds is based on several factors:

• Probability that MPS fails:  5 year subordinated CDS trades at approximately 575bps.  At a recovery of 20% (in line with market convention), this implies approximately 32% cumulative risk-neutral probability of default.  Using 0% recovery implies approximately 26% cumulative probability of default.  Arguably there is non-immaterial risk premium in these numbers and the true probability is substantially less.  However, the CDS market provides the opportunity to hedge this risk at this price, so using the 32% probability appears a reasonable assumption;
• Probability that the documentation weakness described above result in zero recovery:  This is the truly idiosyncratic element of this investment and the key element to assess.  Namely, what is the probability that MPS does not default on subordinated debt but the documentation weaknesses described above result in the FRESH bonds recovering zero anyway?  A reasonable assessment of this probability is approximately 10% (although the market-implied probability – as seen in the sensitivities below – is much higher);
• Fair value of the bonds in all other states:  What is the appropriate spread over the risk-free rate to discount the remaining cash flows in the event that neither of the above two events takes place?  Using spreads for European AT1 instruments as a benchmark and assuming MPS has been restored to profitability, a 500bps spread appears reasonable as a ‘stabilized’ spread.  The other variable that will impact fair value is when coupon payments resume (recall that MPS is required to be both profitable and be paying a dividend in order for coupon payments on the FRESH bonds to resume).  Assuming coupon payments resume in April 2018 on the basis of the 2017 fiscal year the NPV for the FRESH bonds is approximately 70, or a 284% upside to current levels. 

Taking all the above assumptions together, the probability-adjusted fair value for the FRESHES bonds is approximately 49 or a 135% upside from current levels.  Looking at the sensitivity of the bond’s fair value to the timing of coupon resumption and probability of documentation failure, we see that the market seems to be implying a very high probability that the documentation hole results in an Automatic Exchange of the bonds (approximately 60% if coupon payments resume after the 2017 year) and/or a much later commencement of coupon payments. 

Catalyst

• Continued improvement in the health of the bank
• Continued builing of equity capital
• Resumption of coupon payment