What lies beneath

IFR IMF/World Bank Report 2014
18 min read

At the height of the financial crisis it seemed a comprehensive rethink of the financial system was inevitable. It was not only the public, goaded on by an outraged press, that assumed such an outcome was unavoidable, but bankers themselves. Five years on and there is no evidence of any new paradigm in banking. But there are new hybrid instruments, such as CoCos, that regulators hope will make the sector a safer place.

One of the many causes of the crisis was the inadequate levels of capitalisation at the banks. A solution to this problem has been the development of a new generation of hybrid instruments, such as contingent capital – CoCos – debt instruments that are either written off or turned into equity when a certain trigger is hit.

Essentially, CoCos increase a bank’s capital buffer at less expense than equity. They are structurally riskier than other non-equity capital instruments, even than their pre-crisis forebears. But there is nothing shady about them, or immoral about banks selling them, insist bankers.

“Their risks are fully disclosed in offering materials,” said Peter Jurdjevic, head of balance sheet solutions at Barclays.

Europe’s enthusiasm for CoCos is in contrast to developments in the US, which has developed its own instruments to increase bank capital.

“Compared with similar instruments issued by banks in Europe, US banks have been raising Additional Tier 1 capital using perpetual preferred securities, which have the advantage of being far simpler products,” said Nik Dhanani, head of capital solutions at HSBC.

This is no small advantage: some bankers are quietly concerned that the complexity of these structures is leading regulation in the opposite direction of its stated destination.

“While one of the aims of the new framework was to simplify capital structures, the final rules in some regions are in practice promoting greater levels of sophistication and complexity,” said one London-based banker. “This raises the potential for new layers of capital to be created.”

Bespoke

Part of the reason for the complexity is the highly bespoke nature of the products being issued, which banks argue must be tailored to suit their own very specific circumstances and requirements. This makes comparison between different issues very difficult.

“The term CoCo has been used to describe various loss-absorbing instruments, and could mean different things to different people,” said Dhanani.

“Generically, contingent capital securities are instruments that may be written down or converted into equity-based, on predefined triggers. With the advent of statutory bail-in frameworks, that definition suggests a wider range of unsecured liabilities are effectively CoCos. The only variable is the trigger and the severity of the loss absorption.”

Yet pricing is failing to account for the significant differences that exist between products, which have different triggers and different outcomes once the trigger is hit. It also seems to be playing down differences in credit risk. Many peripheral names have been trading tighter to core European credits.

Skeletons in the closets

Banks such as BBVA issued with coupons of around 11.5%, reflecting the risk associated with such institutions at the time, but have now traded down to below 7%. Many in the market acknowledge that these levels have become detached from the true risk of the investment. Yet many also believe banks domiciled in core European countries such as Germany, France and the UK have skeletons that are yet to be revealed in their own closets, and considerable leverage on their own balance sheets.

Structural peculiarities mean some hybrids look riskier than equity. For example, if buffer capital is approaching the trigger and shareholders know that trigger will see them get diluted, that might dissuade them from raising new share capital.

Or if the trigger will see AT1 capital holders wiped out, shareholders may push the company towards that trigger out of self-interest, meaning AT1 is de facto junior to equity. This presents a can of worms of unintended consequences as shareholders look to game the system to protect their own interests above those of other parts of the capital structure.

No sense in standardisation

All of this would be easier to keep an eye on if products were more easily comparable, but a move towards greater standardisation of products does not seem to be on the cards.

“AT1s were created legislatively by CRD IV so they are actually quite standardised, and I would not expect to see further consolidation of structures,” said Jonathan Weinberger, head of capital markets engineering at Societe Generale.

“CRD IV is quite prescriptive but leaves room for decisions at the national regulator level, so differences between instruments are likely to be largely geographical, though issuers and investors will also have their preferences,” said Weinberger. “It is the same in the high-yield market, where issues have different covenant packages but investors can still compare them. It works for high-yield and AT1s can be the same.”

However, writedown instruments do seem to have the upper hand.

”A number of different approaches to pricing have already been developed, some taking equity as the starting off point and bridging from there while others start from debt”

“Over time we expect to see more issuance of products that write down than convert, especially as today they price broadly in line,” said Jurdjevic. “Banks will tend to prefer this option where permitted. If investors start to differentiate more or for higher trigger instruments, we may see more variety.”

Weinberger agreed that writedown securities have been gathering momentum. “If only one type was going to survive it would probably be them, but I think they can easily coexist with equity conversion securities. It just requires better investor education; at the moment it is not clear they are focused on the differences.”

The difference between writedown and conversion will also become less critical as the bank sector gets healthier, said Sam Theodore, managing director of financial institutions at European rating agency Scope Ratings, adding: “Once the AQR has been completed, investors may be reassured still further.”

But there is still the issue of pricing. The starting off point should be the yield on equities on the basis of their respective structures, but as an asset class pricing has come too far inside. However, this is not an issue for CoCos alone: other forms of debt also look mispriced, with the relative mispricing between them less clear-cut.

“A number of different approaches to pricing have already been developed, some taking equity as the starting off point and bridging from there while others start from debt,” said Weinberger. “Some model the risk and others use algorithms measuring the spread between senior and subordinated debt or credit spread diffusion. There are a lot of data points and time will tell which method performs best.”

One banker warned that the increasing complexity of CoCos might accelerate the escalation of problems within banks. Before the crisis a Core Tier 1 ratio of 5.5% was seen as robust, but now many securities trigger at 7%, which, with more conservative deductions and calculations of risk-weighted assets, might be closer to 8.5% in old world terms.

“If an institution approaches the higher trigger it is perceived to be in trouble, but is it?” he asked. “Triggering the conversion could create volatility and become a self-fulfilling prophecy, it could undermine the institution’s ability to raise new capital.”

The BES event

Even if the products do not exacerbate problems within banks, there are other questions regarding their effectiveness. Portuguese bank BES has encountered significant and persistent problems, including a loss of confidence within the financial community because it has been honest about the state of its balance sheet.

But its experience illustrates the limits of capital buffer effectiveness: the bank went from a double-digit Core Tier 1 ratio to below 5% in a matter of days, calling into question how much additional capital is needed to really make a bank safe. It demonstrated how quickly a capital buffer can be eradicated by a single event, which may not even be directly related to the bank itself.

According to Weinberger, little can be concluded from studying the BES case. “The regulators ran the process by implementing a good bank, bad bank structure so it was difficult to see how the instruments behaved on their own,” he said.

Issue now, worry later

Regardless of the misgivings some bankers may have, issuance has been very healthy, despite a summer lull that saw Banco Popular Espanol pull its deal in July, citing poor market conditions. Most of Europe’s big banks have issued AT1s, with Banco Santander and UniCredit both preparing deals late in the summer.

“There is a clear case for bank issuance of hybrid securities in meaningful size,” said Jurdjevic. “There is an economic imperative to issue 1.5% of their risk weighted assets, but in addition there are leverage ratios and potential pillar two requirements.”

The abundant demand seen among investors also ensures that banks will keep coming back to the market. Sitting between traditional debt securities and equity, AT1s see demand from both sets of investors. A fixed income investor that would never invest in equities might consider AT1 because of the additional yield on offer. Equity investors might see them as a hedge on their equity positions.

Analysing hybrid instruments requires investors to understand the specifics of the highly complex structures in question. But as credit instruments, investors must also be comfortable with the name. The lack of differentiation between banks may explain the uncertainty investors feel about the sector, or imply that investors have retained a stubborn belief that governments will always stand behind their banks, regardless of what they say.

Another argument is that investors are failing to even attempt to analyse credits or are buying paper they know is riskier than they are being paid for, be that CoCos or cov-lite loans, because their need for yield leaves them little choice.

“In a super-low yield environment these are one of the few products that really offer yield,” said Gerald Podobnik, global head of capital solutions at Deutsche Bank.

If the big driver for demand really is yield, it remains to be seen how pricing will respond in an increasing rate environment that makes other products more attractive. If the cost becomes prohibitive, banks will look to other sources of capital, such as the equity market, and it is possible these securities will disappear the way the old Tier 1 market did. But that would leave banks needing to retain more profits or do more rights issues.

However, Podobnik stressed that investors have spent considerable time preparing for CoCo investment, and understand the risks.

“Investors compare them to a high-yield corporate and see the choice between a Single or Double B rated credit or a structurally more risky investment in a bank,” he said. “Given the level of regulation around banks now and the high level of public scrutiny, many think the bank investment looks the better bet.”

If regulators are worried about investor enthusiasm for this risky product, they face a dilemma. Regulators have promoted these securities as a way for banks to increase their capital buffers and believe they will reduce systemic risk, but they will only work if there is a healthy market for them. Yet they do not wish to see investors load up on risk either.

The solution has been to generally avoid distributing AT1s among retail investors, although some have slipped through the net: Bankia sold its own issue to customers via its own branches.

“Distributing AT1s to retail customers is especially problematic if they are marketed as safe instruments that are an alternative to deposits,” said Scope Rating’s Theodore. “Regulators need to make sure they are marketed as risky investments but not to prohibit their sale, after all individuals can buy shares that are risky as well.”

Broad acceptance

However, retail investors have not been the predominant buyers of hybrids. A year ago, hedge funds and real money accounted for roughly equal amounts of secondary turnover, according to Barclays figures. But more recently real money has accounted for a larger proportion of total turnover.

“The increasing prominence of real money accounts in the buying and selling of CoCos shows the asset class has won wider acceptance and bodes well for future growth,” said Jurdjevic.

Yet even if hybrids are being bought by sophisticated investors, that does not itself resolve the potential problem. Hybrids may shift the loss away from banks and on to institutional investors but it does not mean a systematically important institution will not be left holding losses in the event of a bank failure. The issue does seem to be investors’ need for yield, and their lack of options in securing it.

“Pre-Lehman, investors had more choices, they had CDOs, RMBS and some of these products no longer exist,” said Theodore.

The comparison may alert investors to the nature of the risk they are taking on.

“It was thought pre-crisis that securitisation was bullet-proof but the crisis showed us that things don’t always work out the way they are expected or designed to,” said Matt Pass, head of European FIG DCM at RBC. “Correlation risk is always very high in these extreme circumstances.”

However, it is important to acknowledge the limitations of that comparison.

“CDOs were considered an appropriate part of a liquidity portfolio,” said Theodore. “AT1 instruments are not considered part of a liquidity portfolio and they are of course risky, but they are less risky than equity.”

The biggest risk with AT1s, he said, is principal cancellation. “But this risk will in time be marginalised as capital ratios inch up, with the current 7% triggers much less threatening in relative terms in the future.”

The risk of coupon cancellation is ever-present and eventually an institution will be unable to pay the coupon due on its AT1s. When this happens it is likely to trigger a repricing across the market as investors reassess the risk of other institutions doing the same.

Until that happens it is impossible to predict how investors will react. But overall bankers are unsurprisingly optimistic.

“A shock may trigger repricing and volatility but we should see the market retain some level of liquidity,” said Jurdjevic. “There will be selling but there will also be buyers even as the price falls.”

“I have no doubt AT1s will perform as intended in a crisis situation,” said Weinberger. However, there are lingering uncertainties, the biggest of which is how conversion will impact the rest of the capital structure. “What happens to the price of the issuer’s covered bonds or senior debt?” he asked.

“The overall framework is untested as the rules have only recently been finalised,” said one banker. “It remains to be seen how equity markets might react to a scenario where multiple instruments are converting into equity, which represents a material amount of the outstanding capital base. Will investors want to hedge their exposure by short-selling the equity before the trigger is breached?”

There is at least a universal acceptance that the current generation of hybrids are an improvement on the last, which circulated pre-crisis. Those instruments were never properly understood either and were considered just another fixed income product by investors, not proper capital, said Theodore. Investors were ill-prepared for the losses they incurred during the crisis. At least next time round investors will know what to expect.

Ultimately, if investors or regulators ever hoped the next financial crisis could be averted by means of a new bank product, hybrids were always set up to fail.

“If a systemic institution fails, there is no single capital markets product that can make a crisis go away,” said Pass. “Preventing bad lending is the only sure fire ‘cure’ for future financial problems.”

“AT1s are just a mechanism to increase capital,” said Weinberger. “The system is safer now, not because of AT1s but because banks have increased their capital. They are only one way of doing that but they are efficient because they allow the risk to be shifted to a new set of investors.”

To see the digital version of this report, please click here.

To purchase printed copies or a PDF of this report, please email gloria.balbastro@thomsonreuters.com.

What lies beneath