On February 12, Banco Santander did not call a perpetual instrument that many in the market had expected it not to call. The price of those bonds briefly dipped from their pre-announcement level of 98.7 cents to just below 97c but had recovered to 98.367c by the following day.
On paper, this doesn’t look like much of a story. But this is the subordinated debt market, which does not react well to things not panning out exactly as expected. Particularly when it comes to calling bonds. Subordinated debt instruments are priced to their first call date: it is taken as read that they will then be called.
In 2008, when Deutsche Bank announced that it would not call a €1 billion lower tier-2 issue because this would be more expensive that extending it, the entire European iTraxx subordinated index widened out by more than 10 basis points. In 2016, both Standard Chartered and Commerzbank decided against calling legacy hybrid tier-1 instruments, and StanChart’s notes immediately fell by 14 points to 83c.
The first sign that the Santander situation might not be going entirely by the book came when the Spanish bank launched a new RegS dollar 7.5% non-call five $1.2 billion additional tier-1 deal on Wednesday February 6, but did not announce the call of its outstanding €1.5 billion 6.25% AT1 notes at the same time.
The new deal had a two-day settlement period, however, which stoked expectation that Santander would announce the call of the euro notes by their February 12 deadline. When the day came, however, the notes were not called.
While many bondholders railed against the decision, the subsequent performance of the bond suggests that its long-term impact might be muted. The possibility of non-call was already priced in as the bonds were trading at 95 in early January.
What is more significant is Santander’s own behaviour. The bank already has a reputation for being relatively subordinated bondholder unfriendly, and the communication process around the non-call of the bond was bizarre.
When the dollar deal was launched, instead of announcing whether or not the euro bonds would be called, the bank simply clarified when the deadline for this decision was: even though they must have already made the decision not to call.
The timing of the dollar issue – which was launched during Chinese New Year when healthy Asian demand for AT1 would have been largely absent – certainly looks interesting, and could suggest that the bank simply wanted to get the dollar deal done before it revealed that it was not going to call the euro one.
The key point here is that just because the market expects AT1 deals to be called on the first call date, there is no obligation for issuers to do so. The AT1 market rallied strongly in January, which fuelled the expectation that Santander would call and refinance. But even after the rally a new euro deal would probably have priced at around 550bp – not far off the existing trade. So, Santander presumably decided it was not economically beneficial to call.
Santander has a further €1.5 billion of 6.375% AT1s that reach their first call date in May, and the next call date for these 6.25% notes falls in June. If there is another failure to call, buyers could revolt.
They shouldn’t. AT1 bonds have coupon and extension risk – and investors are generously paid for both. Banks know this, and Santander must be judging that any future premium it might have to pay for the non-call will be small enough to manage. And it is probably right.
Investor demand for anything with yield is as strong as ever, and subordinated bank paper from a highly rated issuer is just about as good as it gets.
Bondholders may complain, but they will keep on buying.