Drawn to the attractions of subordinated and hybrid bonds, investors have flocked to the FIG market in recent years, driving in spreads. This has resulted in questionable valuations of non-step-up structures and features such as mandatory deferral triggers. Some investors now consider the market overvalued, although the recent credit market weakness is providing selective opportunities.
In the past two years one sector of the credit markets has outperformed all others: subordinated insurance paper. It left sectors such as consumer goods companies, industrials, and utilities trailing, not to mention the auto sector, which was stuck in reverse gear.
Apart from a good showing by utilities in 2005, the only other sectors threatening to overtake the subordinated insurance sector in the past two years have been from the financial institutions asset class as well: in 2004 tier one was just behind in second place, and last year the runner-up was lower tier two paper.
The outperformance of financial institutions paper – with other tiers of capital putting in strong showings as well – has two possible causes.
Firstly, that the asset class has fundamental credit qualities that have supported it in the past two years or so. And secondly, that investors have poured money into the sector because they believe it to have such fundamental credit qualities, which, according to simple supply and demand dynamics, has driven spreads tighter.
"There is the perception, which I believe to be true, that at a similar rating you are better off owning a financial institution because it operates in a regulated market under the supervision of the prudential or monetary authorities," says Eric Cherpion, director in Société Générale's bond syndicate in charge of SSA, financials and hybrid securities in London. "This belief became increasingly prevalent three to four years ago and is one of the factors that has driven the rally in subordinated paper by financial institutions."
Others agree. "If you take a lower tier two transaction and a senior transaction, the probability of default on both is the same," says one DCM official in London. "The loss given default is different, but not the probability of default, and within the European financial institutions world this is perceived to be very low.
"There is a risk of default, of course, but everybody knows that in most countries the regulators would step in before an effective default situation occurred. That is not written anywhere, but it is understood and a few years ago people realised that the gap between the spreads of senior and subordinated paper was too large."
This has resulted in a widening of the investor base for subordinated and hybrid paper. "When the market started in the late 1990s there were maybe 30 accounts that were ready to invest in lower tier two transactions from banks and who understood subordination in Europe," says one hybrids specialist. "Now you have a base of 150 investors who are regular buyers of any kind of capital product.
The sector is also seen as immune to problems that have made the non-financial corporate market a tricky place to invest. "Most investors have taken a hit on positions that have suffered from M&A or LBO activity and they feel safer owning financial institutions paper," says Cherpion. "The mergers that have occurred in the banking world, for example, have been rather friendly for bondholders.
"And that has even been driving in the spreads on banks that are seen as potential takeover targets, a typical example being Italy. There is an M&A premium incorporated in the spread."
One fund manager in London agrees that M&A is more likely to be a credit positive in the financial institutions sector. "For some of the weaker names, it tends to give a boost to their credit quality if they get taken over by a double-A bank or a strong European incumbent," he says.
Standard & Poor's, meanwhile, highlighted the positive credit dynamics in the European banking industry in a research piece in late March entitled "Fundamentals to drive positive ratings momentum in 2006.
"Western European banks enjoyed a vintage year in 2005," said the rating agency, "as strong income growth, well controlled costs, and falling bad debts combined to produce strong increases in pretax profits."
Stick or twist?
However, several market participants feel that after the strong tightening of financial institutions spreads in the past two years and subordinated paper having become so widely accepted, there has been little upside in the sector this year.
"We are at the end of the bull market rally in subordinated and hybrid spreads," says David Marks, head of the financial institutions DCM at JP Morgan in London. "What we have seen in the past couple of months is that for the first time in five years the relative bid for hybrids is less than it is for corporate paper.
He says that this is because of largely technical reasons. "It is the fact that every investor has been putting on the same trade for the last few years, so whenever there has been any weakness there have been marginal investors coming in to support spreads," he says. "But with everyone invested in the sector, where does the marginal investor now come from?
"It does feel like we are in a different paradigm to where we have been before. Its reminiscent of when your taxi driver tells you what a good investment bank hybrid capital is... maybe its time to sell."
Vincent Martinez, credit analyst at Henderson Global Investors in London, agrees. "If you look at hybrid financials, the current problem for me is that you might have a lack of new institutional investors coming into the market," he says. "If you look at last year, everyone seems to have been overweight tier one or hybrid securities across Europe.
"Now when you have volatility you don't have new investors coming in and everybody turns neutral."
Indeed Martinez says that Henderson has been generally neutral on tier one and hybrid issuance in both euros and sterling since February. "Valuations have been stretched since then," he says.
However, he says that it does not make sense to maintain an underweight position. "It is quite difficult to have an absolute underweight on tier one because of the carry you miss out on," he says.
Martinez says that the way they have protected themselves against tight valuations has been by holding a combination of short and long positions, being more selective on new issues and looking at relative value to avoid the high beta names that pose the greatest risk.
Other investors agree that spreads are very tight. When asked if there any names were trading cheap based on fundamentals, one replies: "If only."
Credits that have been trading too tight are, however, easier to spot. One fund manager in London says that the argument that M&A is a credit positive in the banking industry has gone too far in some cases.
"During the beginning of April there were all these rumours about Italian banks being taken over, like BNP Paribas' purchase of Banca Nazionale del Lavoro," he says. "There were rumours about Crédit Agricole buying Intesa or San Paolo and their tier one paper ended up trading in line with Crédit Agricole and other strong double-A banks in Europe, almost flat to HBOS.
"Why? Because the market was considering it too likely that these Italian banks would be taken over by double-A banks. However, I do not believe that this is going to happen soon; I think they will probably go for an Italian solution first with Italian consolidation."
And even if the rumours did come true, he adds, there was no upside in the paper. "If Crédit Agricole did take over Intesa, the tier one would not get any tighter," he says.
Many investors have been reappraising their views of the financial institutions sector in light of the greater volatility in credit markets since May. Shaken by falls in equity markets, potential interest rate rises, and inflation fears, credit spreads widened sharply, and then ebbed and flowed on short term change in sentiment.
Several subordinated and hybrid bank and insurance transactions were launched into this uncertain market in the first week of June, testing how appetite for such paper had been affected by the volatility and widening. The deals for Allied Irish Banks, Fortis, Generali and HBOS sent out mixed messages.
"The recent volatility has meant that we have preferred to stay away from the new issue market," says one fund manager in Paris. "Spreads remain very tight, and any long term widening will likely happen on more volatile paper, such as subordinated.
"Considering the volatility, we are staying away from tier one and will look more at lower tier two paper."
But others have turned positive on the sector. "We have been underweight since the turn of the year, mostly on valuations and expectations of issuance," says one credit analyst in Edinburgh. "We have just recently seen some paper from Generali and others, and spreads have backed up on that issuance, as well as on the back of weaker equity markets and simply a retreat from the tight spreads.
"So we are more comfortable with the sector than we at the turn of the year. I won't go as far as saying that we are overweight, but we are looking at selectively adding paper."
Andrea Hossó, head of financial institutions and emerging markets research at European Credit Management in London, agrees. "The huge overhang has now passed and technicals look good," he says "Still, although not generous, the transactions were fairly priced, particularly looking at the fact that we don't expect much supply going into the summer, although M&A may prompt some issuance, and the deals should therefore perform.
"Real money has actually added to bank positions in May as well as subordinated insurance paper," he adds. "There has been a positive net increase of volumes. Furthermore, if there is risk aversion, it should help financials over the summer."
Another fund manager feels that investors finally have the upper hand. "If you look at the first quarter of the year, there was a lot of cash still coming into the tier one market – just because investors had a lot of money to put to work – and price guidance was being tightened all the time," he says. "Now when deals come to the market it is issuers who are more pressing and so they come with wider price guidance that puts pressure on secondary spreads.
"That was the case for Generali in euros and HBOS in sterling, as they are being forced to compensate investors for the fact that in the short term the paper may widen. So if you have a neutral or slightly positive view on the market, it is the right time to take some positions in new transactions."
Other investors, while in agreement with some of these dynamics, are more cautious on the supply outlook.
"If you are looking on a three to six month basis, some of these deals are offering reasonable value," says Neil Sutherland, investment manager at Axa Investment Managers in London. "But in the short term they may struggle because of upcoming supply."
Sutherland's expectations of further supply are partly based on the attractiveness of the non-step-up, non-innovative structure to banks.
"Every time a new deal comes, it reprices the market," he says, "but given that these securities are still significantly cheaper than common equity, a 10bp-15bp move in bond spreads doesn't make a huge amount of difference to the issuers.
"We saw that with the HBOS deal. They repriced the market by 5bp-10bp, but that, for them, is a small price to pay."
This is of particular concern given the room that banks have for subordinated and hybrid bonds in their capital structures. "In non-innovative tier one hybrid capital they can only do 15% of total tier one capital," says Sutherland, "but in other hybrid securities there is possible capacity to increase this to a total of 50% of their tier one capital. That would suggest almost limitless supply – although in practice the rating agencies would clamp down on it quickly if anyone did too much."
The non-step-up tier one structures have proven amongst the hardest for investors to value since they introduced by Barclays in November 2004. Previously the structure had been targeted at the retail market, but Barclays convinced institutional investors to buy it by paying a spread over its step-up tier one.
However, the question of how much of a pick-up investors require to be compensated for the risk that the deal is not called has, as yet, not firm answer. This has been one contributing factor towards the structures being among the most volatile in the market in any asset class.
"Some investors do not seem to have been fully aware of the duration risks and volatility that you have with certain names or certain structures like the non-step-up transactions," says Martinez at Henderson.
After initially achieving was is now regarded as tight pricing, the first non-step structures widened on the back of wider credit market volatility in March and April 2005. Issuers such as HBOS and Barclays then paid higher spreads to get their deals away, but recently levels have returned to those seen when the structure was first pioneered.
"We took quite large positions in the non-step tier ones last year and were happy to do so because there was quite a significant discount to the tier one with a step-up, and they were trading at a bit of a discount to the secondary market," says one fund manager in London. "But more recent deals, for Alliance & Leicester and BNP Paribas have been priced at much tighter Libor spreads."
This is not just a relative value question. "Everyone started off with the presumption that the bonds are going to be called after 10 years," says the fund manager, "but there is a risk that if corporate bond spreads are significantly wider in 10 years time that the banks do not call them. There is also an outside risk that the regulator might question a bank refinancing them at significantly more expensive levels.
"We felt that the wider levels of last year's deals gave you more insulation against this risk, and that it was not being accounted for in some of the tighter issues this year, where there will be less of a motivation for the issuers to call them."
Most investors do still consider that the transactions will be called, with the rationale being the consequences the banks might face for disappointing institutional investors' expectations.
"It is mostly a reputational issue," says the credit analyst in Edinburgh. "These deals are not sold as if they will be called, but the banks will need to have access to the market and to the institutional investors who make up the market and who are expecting the issue to be called. This is particularly true of names such as HBOS or Barclays who have to have access to the market all the time."
Expressing a view
There can be other issuer specific arguments to persuade investors that a non-step-up issue will be called. Crédit Logement, for example, which launched a Eu800m non-innovative perpetual non-call deal in March via BNP Paribas, HSBC and SG, is not expected to need so much capital from 2010, and is therefore seen as having a clear motivation to call the transaction.
"In the case of Crédit Logement, you cannot be sure that it will be called – that is the principle of the structure," says one investor who met the issuer on its roadshow. "But when we were talking with the management, we got the feeling that according to their Basel II calculations that will come into force very soon they will require less capital overall in their business.
"Therefore when it comes close to the call date it appears that they will have an excess of capital and will redeem the non-step-up tier one."
That this view is shared by other accounts is backed up by the relatively stable trading of the Crédit Logement issue compared to other non-innovative tier one deals.
Sutherland at Axa points out that UK banks may have a different incentive to refinance their non-step-ups in the future.
"One reason why we have been more comfortable with the UK banks than perhaps the European ones is that these non-step hybrid securities are not yet tax-deductible in the UK, whereas they are in the rest of Europe," he says. "There is a reasonable likelihood that the UK securities will be brought into line with European regulations, which means that non-step hybrids in future could be tax-deductible.
"That would give UK banks much more of a motivation to redeem the outstanding ones at the call date."
The lack of agreement on just how much of a premium the structure should require has made it an ideal trading instrument for accounts.
"Because these transactions are the most volatile, they are going to underperform or outperform the market," says one fund manager in London. "They will overshoot any market movement.
"We saw that recently with HBOS non-step paper. It widened 10bp and then came back 5bp, which was more than the rest of the market, and that kind of overreaction is interesting for fast money accounts looking to take a view on the market."
The addition of non-step-up paper to the tier one bucket of indices also explains why in 2005 the sector became more volatile overall and performed more poorly than in 2004.
"Historically liquid subordinated insurance debt has been a high beta sector, so if investors, such as hedge funds, wanted to express a directional view on the credit market, they would take a long or short position in it," says Damien Régent, insurance analyst at UBS in London. "But that is less the case now as investors now use the non-step bank hybrids as well as iTraxx indices and corporate hybrids."
While the market is getting to grips with non-step-up tier one structures, it has generated rules of thumb for valuing other differences between the different tiers of instruments in the bank capital and subordinated insurance sectors.
This would typically involve assigning a spread pick-up or multiple to different levels of subordination or deferral mechanisms, and then either adding these to the spread at the senior level or calculating a multiple of it.
However, some market participants argue that such methodologies are fundamentally flawed. "The trouble with this approach is that it does not pay much attention to the details of the language involved, particularly with regard to coupon deferral," says Julien Turc, head of SG's quantitative strategy team in Paris. "It there does not differentiate between bonds that involve very different types of risks if they have a similar maturity."
Turc has attempted to capture more exactly the risks involved in different structures in a quantitative model, which uses subordinated CDS pricing and assigns recovery rates to different tiers of capital, puts a number on reputational cost, and looks, for example, at the history of measures in mandatory deferral triggers.
He says that the model produces results that are applicable to the market. "If the model highlights a cheap asset, then in theory an arbitrageur should be able to lock in a premium by buying the security and hedging it with a portfolio of sub CDS and spread options," he says. "As not all risks can be hedged, the premium should compensate for non-hedgeable risks plus a risk premium.
"As a result, the model has a very concrete interpretation in terms of trading subordinated issues against CDS."
But Turc acknowledges that the market appears content to use rules of thumb for the time being. "I think that the market should focus much more on the discrepancies between structures," he says, "but the market seems to be saying that as stress situations will never materialise there is no need to do so."
Other market participants agree that there are anomalies in spreads, particularly in the insurance sector where, because the industry is tailoring deals to rating agency and not regulatory rules, structures are more diverse.
Martinez at Henderson highlights the levels at which two subordinated Allianz issues in euros are trading, one a 4.375% perpetual non-call 2017, the other a 5.5% perpetual non-call 2014. "If you look at the Allianz 2014 with mandatory deferral," he says, "it has historically traded inside the 2017, even adjusted for maturity, which only has optional deferral."
He says that this is because of other factors that a quant analysis will not typically take into account. "It comes down to the placement of the bonds and the overall risk perception on the issuer," he says, "even if it is basically a nonsense given the different structures of the bonds."
The Allianz and other examples also show how important a factor maturity is in pricing. "The Allianz 2014 has got quite severe mandatory deferral language," says one credit analyst in London, "but the traditional undated 2017 bond trades wider partly because it is a couple of years longer to the call.
"There is not sufficient compensation for the deferral language, but the market is just distinguishing between perpetuals and dated issues, and when the call date is."
The underlying theme to the market's valuations is its benign view of the banking and insurance industries. "Nobody looks at the structures of bank and insurance hybrids because the operating environment has been pretty favourable," says one credit analyst. "If you look at all the names that are present in the market, they are doing well, so investors are in a mindset where they feel there is no need to differentiate between them."
But some market participants caution against such complacency. "I remember two or three years ago when the markets were weak," says one fund manager in London. "In 2002-2003 the differential between tier one and upper tier two in some names was 50bp. Now that has come down to 10bp-15bp."
"People would go into the minutiae of the documentation of these different issues," he adds. "But in the market at the moment, with spreads so compressed, people tend to ignore the different details of the structures."
This could change, he predicts, if, for example, the current softer market turns into something nastier. "I am sure that if we do get into a weaker market, with a widening spread environment, people will start to look in more detail and there will be discounts in the pricing for features such as harsher deferral language," he says. w