image missing on blob storage!!!

4%. Again?

17 May 2019
Richard Kemmish

Ahead of the Asian covered bond conference earlier this year I wrote a couple of pieces on Singapore’s 4% cap on covered bond issuance. Do the same factors apply to Canada’s 4% limit?


If you want, the articles about Singapore are here and here. But, to briefly summarise, a 4% cap can be to prevent subordination of unsecured creditors, to ensure that banks have sufficient unencumbered assets for use in a crisis, for macro-economic reasons, to prevent too rapid changes in funding models or to ensure an orderly market.  Some of these are weaker than others, none are compelling.

What should replace a cap? Another (higher) cap, an issuer specific number, an increase in pillar 2 capital to ‘punish’ excess issuance or a smarter way of calculating the ratio are all possible approaches. With another arbitrary cap the worst.

Summary over.

Do all of the same arguments apply in Canada?

One major difference between the countries is the greater use of securitisations in Canada. The fact that securitisations don’t attract the same sort of regulatory limits as covered bonds is perhaps cultural – just as there are no covered bond caps in countries where banks traditionally use the product, so there is more regulatory acceptance of securitisations in Canada, Australia or the UK.

Interestingly Singapore is the only country to cap covered bond issuance that doesn’t have a vibrant securitisation market already. That makes the cap there even more conservative.

The forms of encumbrance that we see in a securitisation and a covered bond are different – the latter is more problematic due to the open- ended commitment to replace failing assets (good for investors, bad for issuers). But, crucially, the greater potential impact on total encumbrance of any given quantity of covered bonds is offset by the proven greater likelihood of the covered bond market providing funding for a bank in extremis.

Similarly, in practice a bank in trouble is likely to use covered bonds or securitisations as collateral for repo operations. If the commitment to provide more mortgages for a covered bond is problematic it is nothing compared to the commitment to mark securities to their market value in a repo agreement, and the far greater volatility in the value of the securitisations used for that purpose than we see in the covered bond market. 

The reason for the Canadian use of securitisations is linked to another major difference: the greater reliance on retail deposits in Singapore and capital markets in Canada. Banks in both countries have plenty of retail depositors – to whom we owe the greatest duty of care – but Canadians have far more wholesale senior bonds to bail-in to protect those depositors in a resolution situation. 

Arguably, if a bank has senior bonds owned by rational third-party investors the ‘cost’ of covered bond encumbrance should be factored into the amount of interest that bank has to pay for its senior debt (presumably via the mechanism of its credit rating). So why should a regulator care? 

Deposit insurance on the other hand is a public cost – everyone pays for it, usually via bank levies. If deposit insurance costs more because of an excess of covered bonds this should be reflected somewhere. (I’ll ignore the fact that this is asymmetrical – no bank gets a rebate on their deposit insurance levy because they have reduced their probability of default by issuing robust covered bond funding). The ‘tipping point’ – the number of covered bonds where more the product becomes detrimental rather than positive for the deposit insurance fund - is presumably far higher in a country, like Canada, with significant senior debt whose bail-in probability is already priced in by investors. 

At the risk of sounding facetious, one very obvious difference between Canada and Singapore is that one is very big and the other very small. What does that have to do with covered bond limits? One of the valid criticisms of the covered bond market in Spain was that it facilitated speculative and ultimately useless real estate development. In Singapore it is more or less impossible to build useless real estate; in Canada it is a very real risk. But a covered bond limit is one of the worst ways of controlling excess lending to the real estate market, pillar 2 capital, securitisation caps or floors on the risk weight of residential mortgage lending are all far more appropriate tools.  

Because it is so big, Canada has a very heterogenous banking system, unlike Singapore. We all know the big players but there are also a vast number of smaller banks and credit unions effectively cut off from the covered bond market by the 4% limit. In Australia – with a similar tail of smaller regionals – it was argued that a covered bond law was just another way for the big banks to increase their cost advantage. I haven’t heard this argument made in Canada yet, whether this is because Canadians are more supportive of their big banks or just naturally more polite than Australians, I wouldn’t like to say.

I don’t envy whoever has to design a new covered bond limit for Canadian banks – so many variables to take into account. But I’m looking forward to the discussion. 

Event, Articles and Videos that might interest you

America should not have a covered bond law

10 June 2019 |

At the recent Euromoney North American covered bond conference we held an Oxford Union style debate on the motion that “This house believes that America should have a covered bond law”. To many people’s surprise I opposed the motion. Here’s why.

Tags: Covered Bond

The devil is in the cliché (4): Extension triggers

15 May 2019 | Richard Kemmish

The covered bond directive has a surprisingly large number of paragraphs that make normal people question the intent. The definition of the conditions to trigger an extendable maturity are clearly a political compromise. That’s the wrong starting point.

Tags: Covered Bond