Description

This is an idea that won't make anyone rich but we believe it is a very attractive credit event-driven situation with low credit risk, a nice yield (for EUR) and the potential for a double-digit IRR upon the event playing out. We see this as a short(ish) term idea with a strong catalyst which should be uncorrelated to overall markets and have low downside (particularly interesting in the current environment). 

The FBAVP Perp (ISIN BE0933899800), also known as CASHES, is a legacy capital instrument of BNP Paribas Fortis (hereinafter referred as BNPPF) . It trades at 92c with a coupon of E+2% (~5.5%), providing investors with a base yield of 6%. We expect the CASHES to be redeemed at par in the near term (6-12monhts), providing a significant uplift to the IRR (base case of 10-25% IRR) for an instrument with little credit risk.

In a nutshell, our thesis is:

1. The CASHES are not compliant with current bank capital regulations and we believe they create what the EBA defines as “infection risk” (a more detailed explanation below). As a result, we believe the bank will be forced to redeem the CASHES in the near term (providing bondholders with the pull-to-par on top of the coupon).

2. Since the CASHES are not compliant with capital regulations, they have been excluded from the bank’s capital / loss-absorbing capacity since 2022. These instruments currently rank high in the capital structure and thus have very low credit risk (e.g. cannot be bailed-in in case of a bank resolution). Moreover, Ageas (European insurance) is a co-obligor of the CASHES, further reducing their credit risk.

3. Recent developments on other legacy instruments and discussions with the treasury teams at BNP and other banks lead us to believe that the redemption of the CASHES will come in the relatively near term (3-12 months). A redemption in the next 6-12 months (our base case) would lead to an IRR of 15-25%. Even in the cases of potential delays of 1 year / 2 years (downside cases) the IRR would still be 10-11%.

4. The current base yield of the CASHES (even without pull-to-par) actually looks high for their credit risk even without the event. The senior unsecured debt of BNP Paribas (which is junior to the CASHES)  trades at 3.3%. Note that the CASHES have an even lower credit risk as the “regular” unsecured bonds are part of MREL (so can be bailed-in) and do not have the co-obligor. The fact that BNPPF can fund itself cheaper than the CASHES is also a positive economic incentive (although ultimately not as relevant as the regulatory pressure).

5. There are some complications around redeeming the CASHES (explained below), which are the likely reason why 28% of the original issue is still outstanding. However, these are not insurmountable as discussed below. We expect a solution to be found as regulatory pressure continues to tighten for these to be redeemed.

Below we explore the above thesis in more detail, organizing it in four main sections:

• A) Technical background: Quick summary of what “legacy instruments” are, their regulatory treatment, and what is the potential “infection risk”;

• B) Regulatory treatment of the CASHES: Application of the general considerations of legacy instruments discussed above to the CASHES;

• C) Important read-across from other legacy BNP Bonds: Recent developments in the other two legacy instruments of BNP that point to a near-term solution;

• D) Complications around redemption of the CASHES: Brief discussion on the challenges in redeeming these bonds and how we believe they can be addressed;

A. Technical Background:

Some technical background on the topic is important to understand the thesis - we are happy to go through anything in more detail in the comments/questions.

Legacy instruments:

• As capital requirement rules for European Banks evolved over time (particularly post GFC) certain capital instruments previously issued stopped complying with the definition of "Own Funds". These are called legacy instruments.

• These instruments were initially grandfathered for a transition period (to expire in Dec-21), to be gradually phased out from "Own Funds" and the total loss-absorbing capital of banks.

• One of the mandates of the EBA (European Banking Authority) is to monitor the quality of own funds for European banks and the eligibility of capital instruments issued. Consequently, the EBA has been working with banks to analyze their outstanding legacy instruments, the materiality of their problems, and the issuer’s intentions at the end of the grandfathering period.

• Although there was a significant legislative effort to create uniform and precise regulations on what in theory would constitute non-compliance with the definition of "Own Funds", in practice, given the complex nature of the instruments, significant uncertainty remained regarding the compliance (or not) of each individual security. In fact, over the last few years, multiple banks have expressed different views on the compliance of their own instruments (often completely identical). The general trend has been that the institutions that had taken the view that their securities are compliant have been forced to review their position (often explicitly by the EBA) and exclude those instruments from their "Own Funds".

• As of today, there is significantly more clarity on the topic with the majority of these instruments having now been declassified from "Own funds" (including the CASHES). As a result, they are no longer included in the calculation of the banks’ Total Loss Absorbing Capacity (TLAC). There are still a number of instruments where there is no such clarity.

“Infection risk”:

• Up to 2020, it was widely believed that, even in instances of non-eligible legacy instruments, issuers could still choose to keep those instruments as non-loss absorbing unsecured liabilities (i.e. funding). These would, of course, now rank significantly ahead of their original ranking (ahead of even the senior unsecured bonds) and so issuers should in theory repay those instruments in cases where it would be cheaper to refinance them with cheaper funding.

• In October 2020 the EBA published an opinion paper (Link) on the treatment of legacy instruments in which it introduced the idea of “Infection Risk” (link). In a very simplified way, this consists of the idea that certain of the non-compliance elements described above could, not only compromise the eligibility of the particular security as "Own Funds" but also threaten the eligibility of the remaining instruments in the bank’s capital layer (and the application of the bank resolution framework). There was an update provided on that paper in July 2022 (Link).

• In its October 2020 paper, the EBA concluded that issuers with securities that had “Infection Risk” had three options:
  i) to call, redeem, repurchase, or buyback the instrument;
  ii) to amend the T&C of the instrument with approval from bondholders;
  iii) to keep the instrument on the balance sheet as a non-regulatory instrument.
  This third option however, should only be used as a last resort/backstop option for residual amounts and/or situations in which the issuer can strongly demonstrate that it was unable to pursue either of the prior options (to be only used in limited cases).

B. Regulatory treatment of the CASHES:

With the dry section on the technical background out of the way (congratulations to the ones still reading this) the implications for this instrument are relatively straightforward:

• The CASHES (BE0933899800) are a legacy instrument issued in 2007 that is no longer eligible as capital and has been declassified as such by BNP since Jun-22.

• As per the principles communicated by the EBA, this instrument creates “Infection Risk” on BNPPF and BNP’s balance sheets (we have discussed this with the BNP Treasury team which has not disputed this).

• The EBA has been increasing pressure on BNPPF and BNP (and other European banks) to deal with their remaining outstanding legacy securities. After a significant push in 2023, most European banks have started to address these securities, resulting in the call and redemption of nearly ~€4bn of these bonds.

• Recent discussions we had with BNP and some of the remaining issuers strongly indicate that they acknowledge that these liabilities have to be repaid (although so far there has not been an explicit deadline to do so).

C. Important read-across from other BNP legacy instruments:

During Q4’23/Q1’24 there were material developments on the other two legacy instruments of BNP that have a very strong read-across to the bond we are discussing. The ISINS for these bonds are FR0008131403 and FR0000572646.

• In 2023, these bonds were still classified by the bank as eligible capital (T2) - despite these bonds having various characteristics that make them non-compliant.

• In Sep’23 a group of noteholders of the first bond, represented by the law firm Quinn Emanuel, wrote a letter to the EBA (Link). In this letter, they highlighted the non-compliance of the bond and why they believed it creates “Infection Risk” under EBA’s definition. They asked EBA to look into this in more detail.

• In the last days of Dec’23, BNP announced that it was declassifying both those bonds from its "Own Funds" (clear victory for the noteholders)  and shortly after exercised the option to call and redeem the first bond (it was then repaid in Mar’24). BNP does not have the option to call and redeem the second bond, which is likely why it did not. In our last discussion with BNP, shortly after these events took place, we were told that a legal team was being assembled to review options and solutions for that bond (as there is no option to call). During the same call they also confirmed that, in their view, those two bonds and the CASHES have the same threat of “Infection Risk”.

• A few days later in Jan’24, the EBA replied to the noteholders' letter (Link), essentially agreeing with their argument that the bond in question should have not been included in capital. The following passage (page 2) is particularly noteworthy as it stresses the need to redeem these instruments:

“In this regard, it is recalled that the option to keep the instrument in the balance sheet as a non-regulatory instrument should be treated as a last resort option, i.e. if the redemption of the instrument or the amendments of the terms and conditions are not possible (...)”

D. Main complications/risks to the thesis:

As discussed above, although the three legacy instruments above are all ineligible for capital and create the same threat of “Infection Risk”, only one of them has a call option that allows the issuer to unilaterally redeem the bonds. We believe that this is the reason why, after the de-classification from capital at year-end 2023, that bond was called/redeemed while the other two remain outstanding. It is important to note that only €831m of the CASHES are currently outstanding (which corresponds to ~28% of the original size).

 1. How to repay a perpetual bond without call features:

Although the non-call nature of these bonds makes the process more complicated for the issuer, we believe that there are a few options to solve the situation (this is likely what the bank is currently analyzing):

1. Amend T&C to solve non-compliance and “Infection Risk”: We believe, that, in reality, this is not feasible as the larger noteholders would likely not consent (as they want to be repaid). Of all the numerous cases of consent solicitations attempted over the last 2 years to change documents for legacy instruments, we have yet to see one that succeeded.

2. Bond buybacks: While this would be effective at reducing the amount of bonds outstanding (maybe even at a small discount to par), it would be unlikely to fully resolve the problem entirely (ie. fully redeem the bond).

3. Tender offer (likely at 100% or above): A tender offer presents in our view the most effective way to try to resolve the situation. The tender could be made contingent on bondholders agreeing to insert a call option on the notes, which would provide the issuer a way to deal with any hold-out / non-tendered bonds (provided that they get a 75% acceptance rate).

Based on the instructions provided by the EBA, we believe that BNPPF and BNP would only be able to keep any amount of these CASHES outstanding after “strongly demonstrating” that it was impossible to redeem them. In our view, it would be hard to do so without trying a par tender.

We should also note that in theory, the tender offer may happen above 100% to get bondholders to accept (creating further upside for the investment), although we prefer to assume a par repayment. Likewise, in theory, the tender offer could also happen <100%. We believe that a tender <100% would not be likely as most bondholders would not accept it (not solving BNP's "infection risk") and the regulator would be unlikely to consider that BNP has done everything it could to redeem the instrument.

It is also worth noting that this bond is quite small in the balance sheets of both BNP (2,700bn of assets) and BNP Paribas Fortis (374bn of assets). If there is indeed strong regulatory pressure, saving a few % on the tender price will unlikely be critical.

2. Finding an agreement between BNP and Ageas:

The second complication arises from the complex nature of this instrument. While the CASHES are a liability on the balance sheet of BNPPF/BNP (which pays the entire coupon to noteholders), there is a compensation mechanism between both institutions called the Relative Performance Note (RPN), under which Ageas has a liability to BNP Paribas Fortis (and thus BNP Paribas Fortis has an asset that partially offsets its own liability to noteholders). The RPN only remains outstanding to the extent that the CASHES are outstanding. The RPN varies in size (depending mostly on the price of the bonds) and has a coupon of E+90bps. As of the 30-Jun-2024 the value of the RPN in Ageas balance sheet was €432m.

What this means in practice is that when these bonds are redeemed the RPN (a liability for Ageas and an asset for BNPPF/BNP) disappears. As a consequence, realistically BNP would be looking for Ageas to also contribute to the repayment of the bonds - as seen in the examples below.

Previous tenders/purchases:

• In February 2012, BNP launched a tender offer at a price of 47.5c and purchased €1.9bn of bonds (63% of the original amount). Ageas agreed to pay a €287m indemnity to BNP (vs. a total payment made by BNP of €897m). It is important to note that at this time there were still T1 instruments (so significantly more junior), which partially explains the big discount.

• During 2016, under an agreement with Ageas (that expired in the same year), BNP Paribas Fortis purchased €164m of the CASHES in the secondary market. While the price was undisclosed, we estimate that it was ~66c. Ageas agreed to pay BNP €44m (vs. a total payment made by BNP of €108m).

• In H2’22 BNP was approached by a noteholder looking for a bid and bought back ~€116m of bonds (our estimate) for an estimated price of 79c. Ageas agreed to pay BNP €47m (vs an estimated total payment made by BNP of €92m).

In prior tender purchases, we believe that Ageas agreed to pay roughly 100% of the RPN being redeemed as compensation to BNPPF.

In reality, an agreement between BNPPF and Ageas (assuming a 100% redemption). should require Ageas to pay ~€432m and BNPPFs to pay the remaining €400m. The incentives from both sides to do so are:

• BNP/BNPPF (discussed in detail above): Significant regulatory pressure due to “Infection Risk” + economics of redemption (the bank can fund itself significantly cheaper). The significant savings in the cost of funding (>2% per year on a perpetual liability) should allow BNP/BNPPF to offer Ageas a potential discount on their compensation (particularly given the regulatory pressure faced by the bank).

• Ageas: We believe there is also some economic incentive from Ageas to repay and refinance this liability (which costs them E+90bps, significantly higher than the cost of funding for Ageas) - particularly if BNP/BNPPF agrees to a “sweeter-deal”. Moreover, the way that this liability is accounted for in Ageas’ balance sheet (its value changing with the price of the CASHES and the value of Ageas shares) creates extra earnings volatility for Ageas (e.g. from Dec’22 to Jun’24 the change in value of the liability resulted in a loss of almost €100m for Ageas which is not immaterial vs. the company’s ~€1bn net income).

Disclosure: We own CASHES (ISIN BE0933899800). Everything in this post is our own opinion and does not construe investment advice. Please do your own due diligence.

Catalyst

• The ultimate catalyst here is the tender announcement at 100% which, in our view, could happen at any point

However, it is possible that there are some intermediate catalysts before the final outcome (which would increase the visibility into that outcome), including:

• More regulatory pressure from ECB/EBA to repay the legacy bonds (both behind the scenes specifically to BNP but also to the market through publications);
• Tender/other solution for the FR0000572646 (which also does not have a call option but does not have the extra complication of Ageas as co-obligor). We believe this what solution BNP found to address legacy bonds with no call features;