As we know, the calculation of coverage across covered bond regimes can be on either a nominal or a present value basis. That neither approach is better - ‘it just depends’ - is an assertion so often repeated that it has become trite. But is it true?
Inherent in the distinction between nominal and present value (PV) is a fundamental difference. Nominal based coverage calculations compare outstanding principal only. Present value based coverage calculations compare the amount and timing of that principal and interest payments. Lets think about those differences.
The time at which the principal will arrive in the cover pool is, surely, irrelevant if it is going to be reinvested in new mortgages? Anyone from a fixed-rate mortgage environment would say no because we dont know what yield the new mortgages will have when the cover pool needs to reinvest that returning principal. Anyone from a floating-rate mortgage environment will say that the timing of the repayment is irrelevant because the mortgages repaying, and the new ones available for the cover pool at the time, are both on market.
The value of future fixed-rate interest payments relies on them being received. That is the mortgage will not prematurely repay, or if it does repay, it is with a pre-payment penalty calculated on the basis of the economic value of the lost coupon payments. It usually does.
But this also assumes that the borrower will not default. In floating-rate environments sudden rate hikes presage high defaults. To some extent fixed-mortgage rates reduce this risk. But not totally. Five or ten year fixed-rate mortgages for example run the risk of much higher monthly repayments when they come to the end of their term. Similarly, commercial borrowers (buy-to-let landlords or corporates) probably have interest rate exposure elsewhere. Companies fail more often when rates rise whether mortgages are floating or fixed.
Another overlooked factor is whether repayment penalties are enforceable in insolvency. Or does the mortgage lender have a claim of just the nominal value? It differs by jurisdiction and between theory and practice.
Two practical considerations to add into the decision. Firstly with a coverage calculation based only on present value, in theory the principal of the bonds can exceed that of the assets bonds can be issued against future excess interest. I suspect that this does happen in practice (it is a thing in some parts of the securitisation market. Oddly nothing in the STS rules seems to prohibit it). It feels wrong.
Secondly, at what interest rate should PV be calculated? Swaps flat is supposed to be a risk-free rate so appears too low. But as the mismatch between asset and liability duration can be in either direction, and both must be calculated on the same curve, using a high discount rate is not necessarily the more conservative option.
So yes, it is true that the nominal or PV choice is a function of many things, including the duration of the underlying assets, the nature of the credit (the borrowers interest rate sensitivity) and legal analysis of enforcement. National specificities apply, but it isnt always a straightforward decision.
Event, Articles and Videos that might interest you
No related events !