Contingent convertibles: when 'vanishing' bonds make good investments

The Santander CoCo

They were waiting for the call.

It was February 6, 2019, and investors in Banco Santander SA's EUR 1.5 billion contingent convertible bond were waiting for their bonds to be redeemed by the Spanish bank. The bond had been issued in 2014. For five years investors had been receiving the generous 6.25% coupon paid by the bank.

The coupon payments arrived like clockwork every quarter, and were credited straight into their accounts. Everything was fine and dandy.

Five years later, it was time for the bonds to be redeemed. Even though the bond technically bore no maturity date – it was a perpetual note – it had an optional redemption date, a 'call' date on which the bank could redeem the notes at face value. The call notice was scheduled to come in about a week’s time on 12 February.

Perpetual bonds were popular instruments in Europe and the European banks were all following the same playbook – issuing perpetual bonds with an optional call date, and redeeming the bonds when the first call date arrived. The banks were eager to demonstrate that their financial strength was beyond question and safeguard their reputation. For bond investors, it had become an accepted industry norm, an inviolable 'gentlemen's agreement'.

Moreover, Santander had just completed the issuance of another USD-denominated perpetual bond, which raised USD 1.2 billion for the bank, and market observers might have intuited that the proceeds would be used to redeem the EUR 1.5 billion issue. But when the dollar bond deal was completed without being accompanied by a call announcement for the EUR 1.5 billion notes, anxiety began to seep into the market.

The pundits went into overdrive and speculation reached a feverish pitch. Some said that skipping the call might work out cheaper for Santander, with the coupon likely resetting to around 5.5% versus the 6.25% that the bank was paying. It was a strong economic incentive for Santander not to call its bond.

Nonetheless, investors clung on to the belief that Santander would not leave them hanging. Santander was not facing any known crisis, was amply capitalized, and enjoyed easy access to capital markets.

There was nothing stopping the bank, at least from the perspective of financial resources, from redeeming the bonds. The perceived priority was to placate the market. Besides, any decision not to redeem the bonds on first call would be potentially catastrophic in terms of future funding costs for the bank.

Yields on Santander's perpetual notes could spike, in part to account for the increased non-call risk associated with the bonds. Santander's executives had to know that its reputation would take a beating. Surely they wouldn't gamble with the company's reputation?

The answer arrived on 12 February. In an announcement near the end of the day, the Spanish bank said that it would not exercise its option to call the EUR 1.5 billion contingent convertible issue the following month, making Santander the first European bank to fail to call this kind of hybrid debt on its first call date. The unprecedented development sent the market into a selling frenzy, with the bonds tumbling from a high of almost 98.5 to 97 cents following the announcement. The tersely worded statement issued by Santander made for sober reading, with a spokesman noting that Santander had an 'obligation to assess the economics and balance the interests of all investors.'

The gentlemen's agreement had been broken and the market for bank-issued contingent convertibles, at least in Europe, would never be the same again.

In the aftermath of the Santander affair, investors have become acutely conscious that their perpetual contingent convertible bonds face the possibility of not being redeemed after the customary five-year non-call period. A possibility that remains remote, but real nonetheless.

The Santander affair was not only a wake-up call for perpetual contingent convertible investors; it focused the world's attention on this popular, and rapidly growing asset class. In fact, it is estimated that the market for such debt is worth some USD 340 billion.

In the next section, we provide an introduction to this unique asset class and explain why it has become such a popular alternative to plain vanilla bonds. We invite you, our esteemed readers, to join us on this journey of exploration.

Contingent convertibles: why they exist and how they work

Contingent convertibles, known colloquially as CoCos, were first promulgated by European regulators in the aftermath of the 2008 financial crisis. In the years leading up to the financial crisis, European banks such as UBS in Switzerland and Royal Bank of Scotland in the United Kingdom had loaded up on collateralized debt obligations ('CDO') – complex financial products that pooled together cash-flow generating assets into a single package that could be sold to other investors. If that was not complicated enough, these banks underwrote and traded in all manner of derivatives based on the value of these products.

When Lehman Brothers filed for bankruptcy and AIG, a prolific trader of these complex derivatives, ran into trouble, the market for CDOs and their derivatives entered a deep freeze. Liquidity dried up and asset values plunged.

European banks, which were exposed to these instruments to the tune of billions of euros, were caught flat-footed. Even UBS, the venerable Swiss bank with a century-long history, took a USD 48 billion write-down and agreed to transfer USD 60 billion worth of illiquid subprime assets to a special vehicle supervised by the Swiss National Bank.

It was a sobering lesson for European regulators, one which highlighted the need for a way for European banks to raise their capital ratios without having to raise additional equity, or resort to taxpayer-funded public assistance. Contingent convertibles, or CoCos, offered a way to achieve this and the first CoCos were issued by Spanish bank BBVA in 2013. The market has not looked back since.

CoCos offer attractive yields compared to straight bonds. The Credit Suisse Capital Contingent Convertible Index, a series of indices that groups together more than 100 contingent convertible issues, indicates that at the end of April, the average yield of a USD-denominated CoCo was 5.82% and that of a EUR-denominated CoCo was 4.08%. Further details on the index are reproduced in Figure 1 for the benefit of our readers.

Figure 1: Credit Suisse Capital Contingent Convertible Index (CoCo)

But what are CoCos and how do they work?

CoCos are primarily hybrid securities; they are bonds that absorb losses when the capital of the issuer – which is usually a financial institution such as a bank – falls below a certain level. At this point, the debt of the bank is reduced, and the capitalization of the bank is restored.

By virtue of their ability to absorb losses at a moment where it is most needed, financial regulators generally regard CoCos as an important part of a bank's capital structure that help to satisfy regulatory capital requirements. CoCos absorb losses for the bank only upon the occurrence of specified trigger events that are clearly defined in the bond prospectus. We will discuss what these trigger events are later in this article.

What is more important, however, is to understand the mechanisms that enable CoCos to absorb losses. There are two possible means.

The first entails the securities converting into common equity (common shares) in the bank, in accordance with a pre-determined conversion ratio that specifies the conversion price to be applied in the event that a conversion is triggered. The second involves the bonds suffering a principal write-down.

When a principal write-down occurs, the amount owed to each CoCo bondholder is adjusted downwards. For instance, if a bondholder owns $250,000 worth of a CoCo bond and suffers a 20% principal write-down due to the occurrence of a trigger event, the bank no longer owes him $250,000, but $200,000 instead. The principal write-down could be permanent or temporary, with the latter being relatively less common than the former.

Earlier, we alluded to the fact that CoCos are only converted to equity or suffer a principal write-down upon the occurrence of specific trigger events clearly defined in the prospectus. Two categories of trigger events exist. The trigger could be mechanical in nature, where a conversion or write-down is triggered once a specific capital ratio of the bank falls below a certain figure. Or the trigger could be discretionary, where conversion or write-down is subject to supervisory (regulatory) judgement, which is by definition subjective in nature.

To provide our readers with a real-life example of a CoCo issue subject to a discretionary trigger, we consider the CS 5.625% Perpetual Corp (SGD), a perpetual security issued by Credit Suisse Group AG ('CSG'). An extract from the bond prospectus defines one of the trigger events as follows:

'The Regulator has notified CSG that it has determined that a write-down of the Notes, together with the conversion or write-down/off of holders' claims in respect of any and all other Going Concern Capital Instruments, Tier 1 Instruments and Tier 2 Instruments that, pursuant to their terms or by operation of law, are capable of being converted into equity or written down/off at that time, is, because customary measures to improve CSG's capital adequacy are at the time inadequate or unfeasible, an essential requirement to prevent CSG from becoming insolvent, bankrupt or unable to pay a material part of its debts as they fall due, or from ceasing to carry on its business.'

The primary reason why such a trigger is deemed discretionary is because any conversion or principal write-down is dependent on the subjective judgement of the regulators, in this case the Swiss Financial Market Supervisory Authority or any competent regulator recognized by the Swiss government.

Mechanical triggers work in a more straightforward way. As explained earlier, such triggers usually come into play when a clearly defined capital event occurs. For many CoCos, this entails the bank’s Common Equity Tier 1 ('CET1') ratio falling below a certain fixed threshold. CET1 capital consists largely of common (ordinary) shares, and does not include preference shares and debt issued by the bank.

Using the same Credit Suisse CoCo bond as an example, an extract of its prospectus reads:

'CET1 Write-down Trigger: 7.00%, based on Credit Suisse Group AG consolidated CET1 ratio.'

This means that when the CET1 ratio of the bank falls below 7%, calculated on a consolidated basis at the group level, the bonds are written-down automatically. Since the threshold was fixed at 7% when the bonds were issued, we consider this to be a mechanical or non-discretionary trigger.

Even though in our example a CET1 ratio threshold – the most commonly used metric in the CoCo market – was used as a trigger, by no means is it the only mechanical rule applied by CoCo bond issuers. Mechanical triggers fall into one of two broad categories: accounting-value triggers and market-value triggers.

As its name implies, accounting-value triggers rely on the calculation of book-value dependent metrics. The CET1 ratio, which is defined as the accounting-based value of the bank’s CET1 capital as a ratio of risk-weighted assets ('RWA'), falls into this category.

One of the reasons why accounting-based triggers – such as those that reference the CET1 ratio – are so popular is because they allow banks to respond directly to regulatory requirements. The reader might recall that in the aftermath of the 2008 financial crisis, financial regulators across the globe introduced a rash of regulatory measures designed to boost bank capital adequacy and promote stability in the international financial system.

The most notable is Basel III, which not only specifies minimum CET1 ratios banks have to maintain, but also defines the minimum CET1 trigger level required for a CoCo to qualify as Additional Tier 1 ('AT1') capital. To understand its significance, some context is required.

Basel III imposes a minimum Tier 1 capital ratio of 6%, of which 4.5% must be composed of CET1 capital and 1.5% of AT1 capital. Under the Basel standards, CoCos are recognized as AT1 capital, provided that they contain a CET1 trigger of 5.125% or higher. It is thus natural for banks seeking to raise capital that qualifies as AT1 securities to issue CoCos with a CET1 trigger at 5.125%.

The inquiring reader might wonder why the bank would choose to fix the threshold exactly at 5.125% instead of a higher level. The reason is simple. All things being equal, a higher threshold would make it more likely for the trigger event to occur and for bondholders to suffer loss through a conversion or write-down. We can imagine bondholders demanding a higher yield in order for them to accept the greater risk associated with a higher trigger point, which would in turn increase the associated borrowing costs.

Accounting-based triggers might be common, but they have one key flaw. The frequency of calculating metrics such as CET1 and publicly disclosing them can vary significantly across banks. Furthermore, book-value triggers such as CET1 are susceptible to guesswork or questionable manipulation. As such, these triggers may not be activated in a timely manner.

A less common category of mechanical triggers are market-value triggers. Such triggers fix the minimum market capitalization (which is a function of stock price) that the bank must maintain relative to its assets. Market-value triggers address some of the problems posed by accounting-value triggers, in particular those caused by inconsistent accounting valuations.

Unfortunately, a trigger based on the bank's stock price presents new problems of its own. For one, the structure introduces incentives, under certain circumstances, for CoCo holders to short-sell the underlying stock in order to force the price below the trigger threshold.

A minor point that might be of interest to our readers: CoCo bonds might be subject to a mechanical trigger, a discretionary one, or a combination of both. In the third case, the occurrence of either a discretionary or mechanical event could trigger a conversion or write-down.

Even though we did not state so directly, the attentive reader would have inferred by now that CoCo bonds actually convert outside bankruptcy. In other words, the conversion to equity or write-down of principal takes place before the company enters bankruptcy and is liquidated. At the risk of repetition, for many CoCos, this occurs when the issuer’s CET1 ratio falls below the pre-defined threshold level.

This observation is of great significance because therein lies the primary trait that distinguishes CoCos from regular bonds. Where plain vanilla bonds only suffer losses after equity investors see their stakes wiped out totally, in some cases of CoCos with high triggers, bondholders can actually suffer losses ahead of shareholders when the loss absorption mechanism of the instrument is activated. In this we see a potential inversion of the traditional hierarchy of investors.

An extract of the prospectus of a CoCo issued by DBS Group Holdings is elucidative, and reads:

'A write-off or conversion (if applicable) of any DBS Bank Common Equity Tier 1 Capital shall not be required before a DBS Bank Write-off or any DBS Bank Subordinated Notes can be effected in accordance with these conditions.'

At the risk of belaboring the same point, to grasp the full import of the CoCo structure is to appreciate that CoCos are bonds with equity characteristics, even if they might be classified as fixed income instruments.

Conversion of a CoCo: an illustration

Note: The content of this section has a slight technical bent. Readers who wish to directly access our bond recommendations may wish to skip this section and proceed to the next section entitled 'The main CoCo players'.

Earlier in the article, we indicated that CoCos absorb losses by converting into shares or suffering a principal write-down. We provided an example of a principal write-down but we did not explain in detail how a CoCo is converted into shares. That is the raison d'etre of this section.

In Figure 2, we provide a series of accounting snapshots of a hypothetical company — from CoCo issuance, to the moment when the conversion trigger of the CoCos is set off, and after the CoCos have been converted into common shares.

Figure 2: From CoCo issuance to conversion — a working example

We adopt a number of simplifying assumptions for this example, the most important being that the market value per share – on which the conversion trigger event is based – corresponds to the book value per share at all times, and there is no time lag. In our example, the CoCos are subject to a CET1 trigger with CET1 capital being linked to the share price of the company, which in turn corresponds exactly to its book value per share at any given time.

We fix the CET1 trigger at 5.125% — the CoCos are converted to common equity once the company’s CET1 ratio falls below 5.125%. There is no supervisory or discretionary trigger.

The CET1 ratio is calculated as the percentage of CET1 capital over RWA. We assume an arbitrary risk weight of 72.3% for the calculation of RWA. Risk-weighted assets simply refer to the total assets held by the bank, weighted by risk. It is a theoretical construct used by bank regulators to calculate CET1 adequacy and a detailed description of its calculation methodology is beyond the scope of this article.

Our starting point is the moment when the company has just issued $50 worth of CoCos. It has $2,550 in other debt obligations. These could include senior unsecured notes or other forms of debt that are not contingent convertible. For this example, the exact mix of securities is not of particular consequence.

The company has $200 in the book value of equity and its total liabilities – the sum of CoCos and other debt – are $2,600. In this example, the company has 200 shares outstanding which make the book value per share – equity divided by number of shares – $1.00.

Let’s suppose that the company's financial situation deteriorates significantly post-issuance of the CoCos. The second and third columns show the common equity declining from $200 to $100; the book value per share drops from $1.00 to $0.5. The nominal value of CoCo bonds and other debt remains unchanged at $2,600.

The reader would recall the basic accounting identity of Assets = Liabilities + Equity, an identity which implies that our company's total assets have declined from $2,800 to $2,700 over the same period. Applying the assumed 72.3% risk weight, we get RWA of $1952.10. Taking the CET1 capital of $100 and dividing it by the RWA of $1952.10 gives us 5.123% – the bank has breached the 5.125% threshold for conversion.

What happens after the CoCos are converted into common equity? In the fourth, fifth, and sixth columns, we paint three separate scenarios with different conversion price per share ('CPS') of $0.2, $0.5, and $1.0. Since the outstanding CoCos have a face value of $50, applying the conversion price of $0.2, $0.5 or $1.0 per share would yield 250, 100 and 50 converted shares in the respective scenarios.

If the conversion price is $0.2 per share, the total value of the 250 converted shares is $83.3 [number of shares times book value per share]. Applying the same reasoning, the values of converted shares when the conversion price are $0.5 and $1.00 are $50 and $30 respectively. Before conversion, the CoCo debt was worth $50, the value of which was fully converted into equity.

We make a few important observations here. For one, the conversion price does not affect the aggregate capital or value that is converted into equity. In our example, the entirety of CoCo debt went toward increasing the level of CET1 capital available to the company.

Secondly, the conversion price does affect the loss that the CoCo holders have to take. When the conversion price is fixed at $1.0, CoCo holders receive $30 worth of common shares, an impairment of 25% from the value of their CoCo holdings. No impairment occurs when the conversion price applied is $0.5, whilst a gain in value occurs when the conversion price is fixed at $0.2.

The astute reader might realize that a loss on the CoCo investor occurs only when the conversion price exceeds the book value per share of the company at the time of conversion. The higher the conversion price, the greater the loss of value suffered by CoCo bondholders.

The same reader thinking one step ahead would also recall that the value of the company remains the same regardless of the conversion price applied, meaning that the value lost/gained by CoCo holders cannot possibly disappear into thin air. This is not entirely an enigma: any value lost by CoCo holders has to flow to pre-existing common shareholders.

In a nutshell, when the conversion price of the CoCos corresponds to the company's book value, no redistribution of value occurs and the conversion is neither dilutive nor anti-dilutive to equity shareholders. But if the conversion price is fixed at a level that is higher than book value, the common shareholders benefit from the transfer in value from CoCo bondholders i.e. it is anti-dilutive for equity shareholders.

At this juncture, we would invite the reader to go one step further and think about what would happen if we were to fix the conversion price at an incredibly high level. A trillion dollars perhaps.

With such a price, the CoCo bondholders would receive an infinitesimal number of shares. The market value of these shares would be close to zero and virtually all the value of the CoCos would be transferred from bondholders to equity shareholders.

The above scenario is analogous to the situation where CoCos, instead of being converted into equity, suffer a principal write-down. Because holders of such CoCos are not compensated by way of converted shares, the exercise is anti-dilutive and favorable to common shareholders. This is one of the reasons why common shareholders favor the issuance of CoCos that have principal write-down as its loss absorption mechanism over those that convert into shares.

In the next section, we introduce some of the main players active in the CoCo space and highlight some interesting CoCo investment opportunities that investors might wish to consider.

The main CoCo players

CoCo issuance is largely concentrated in Europe, even though the proportion of CoCos issued by Asian banks is growing at a healthy clip. For the moment, we advocate selecting within the European space, partly because of the regulatory framework in Europe for these relatively new instruments is more established.

Defaults in the CoCo bond space are exceedingly rare. Save for a single instance where Versobank – a small financial institution in Estonia — defaulted on the back of regulators cancelling its operating license, we know of no other cases of default over the last two decades.

That said, it is nevertheless important to take a prudent approach to CoCo bond selection. This implies looking out for issuers with healthy balance sheets, minimizing the loss-absorption risk associated with CoCos.

It is clear that no CoCo bondholder wishes for a trigger event to occur and for his bonds to be converted or written down. Where the trigger event is defined by the company's CET1 ratio, the probability of CoCos being converted prior to redemption depends on how far the company's CET1 ratio is above the fixed CET1 threshold when the CoCos were issued. The greater the distance, the lower the likelihood of conversion.

Figure 3 presents a list of selected European CoCo issuers, their second-quarter CET1 ratios, and the trigger thresholds that apply to the respective issuers.

Figure 3: List of selected CoCo issuers

The reader would immediately observe that amongst our selected list of issuers, two Europeans banks stand out in the area of CET1 capitalization – Deutsche Bank and Standard Chartered. These two institutions have healthy CET1 ratios that are significantly higher than the trigger thresholds that apply on their CoCos. The gaps between their CET1 ratios and the CoCo triggers provide a significant margin of safety for investors who choose to invest in the CoCos of these two institutions.

One should note that we are not saying that other bank CoCos are at risk of being converted. Such an occurrence is unlikely. What we are saying, though, is that the wide CET1 gaps above their trigger levels posted by Deutsche Bank ('DB') and Standard Chartered make an already unlikely event even more unlikely, and that this provides a supplementary reason – other than yield – why an investor would favor DB and Standard Chartered.

Figure 4: Selected CoCo bonds

One standout opportunity that we would strongly recommend is Deutsche Bank's DB 6.250% PERPETUAL CORP (USD), which has a first call date in April next year and has a yield to call of 32.3%. A confluence of positive factors enable us to view this issue favorably. Not least is Deutsche Bank's strong CET1 capital position, something that we have already alluded to earlier. The second is the attractive potential upside in excess of 30%, should the issue be redeemed early next year.

The skeptical reader would counter that there remains the significant possibility that the bank might choose not to call the bond, mirroring Santander's example. Without doubt, we have to acknowledge that the possibility exists. But the reputational cost of not calling remains a significant consideration in the financial industry and the Santander incident remains somewhat of an outlier. But even assuming that the bond is not called on its first call date, at the current market price of 85.23, the bond boasts a very attractive yield to worst of 6.76%.

We also like the fact that even in the unlikely scenario where the CET1 threshold is breached, the bond provisions mandate that any write-down implemented would only occur to the extent that allows the bank to restore its CET1 ratio to 5.125%. Practically, even in this adverse scenario, bondholders should only experience a partial and limited loss of principal.

Unlike many CoCo issues that are subject to permanent write-downs, the DB 6.250% PERPETUAL CORP (USD) is unique in the sense that it offers investors the possibility of a write-up, where the company has the discretion to reverse any effected write-down at a later date.

Conclusion

Considered by many investors to be in a league of their own, contingent convertibles allow investors to capture significant yields, while possessing a structure that mitigates much of the volatility that comes with owning common equity.

We strongly recommend Deutsche Bank’s DB 6.250% PERPETUAL CORP (USD). We are motivated by the convergence of positive factors in its favor, chief among which is Deutsche’s strong capital position.

Declaration:

For specific disclosure, at the time of publication of this report, IFPL (via its connected and associated entities) and the analyst who produced this report hold a NIL position in the abovementioned securities.