Residential Mortgage Risk Weights – Cp29/16

[Ah-so]

One for those of you who take a geeky interest in risk weighted assets. The PRA has released a consultation paper that seems to have the implication that banks will need to hold higher RWAs against mortgages and will be a back-stop against those banks that really abuse the Basel modelling approach to IRB.

http://www.bankofengland.co.uk/pra/Pages/publications/cp/2016/cp2916.aspx

Summary of proposals

The PRA proposes to amend the SS11/13 such that firms would be expected to adopt probability of default modelling approaches for their residential mortgage portfolios that avoid the lack of risk capture identified in the point-in-time and through-the-cycle models currently used by firms, and instead calibrate their models using a consistent and appropriate assumption for the level of model cyclicality.

The PRA also proposes to expect firms not to apply a house price fall assumption of less than 25% in their UK residential mortgage loss given default models.

[spyguy]

One for those of you who take a geeky interest in risk weighted assets. The PRA has released a consultation paper that seems to have the implication that banks will need to hold higher RWAs against mortgages and will be a back-stop against those banks that really abuse the Basel modelling approach to IRB.

http://www.bankofengland.co.uk/pra/Pages/publications/cp/2016/cp2916.aspx

Summary of proposals

The PRA proposes to amend the SS11/13 such that firms would be expected to adopt probability of default modelling approaches for their residential mortgage portfolios that avoid the lack of risk capture identified in the point-in-time and through-the-cycle models currently used by firms, and instead calibrate their models using a consistent and appropriate assumption for the level of model cyclicality.

The PRA also proposes to expect firms not to apply a house price fall assumption of less than 25% in their UK residential mortgage loss given default models.

So, 75% LTV max going forward then.

[eek]

So, 75% LTV max going forward then.

How easy is it to get a 75%+ mortgage without right to buy....

[spyguy]

How easy is it to get a 75%+ mortgage without right to buy....

No so.

HSBC, who I use as my test lender, seem to be wary of LTV 60%+

HSBC seem to want as little exposure to the UK housing market as possible.

[Ah-so]

The smaller lenders tend to offer better deals for high LTVs.

[Venger]

The PRA also proposes to expect firms not to apply a house price fall assumption of less than 25% in their UK residential mortgage loss given default models.

Implementation

The PRA proposes that these changes will come into effect by 31 March 2019 with firms allowed until 31 May 2018 to submit for approval adjusted residential mortgage models meeting these expectations.

:yawn:

[spyguy]

The smaller lenders tend to offer better deals for high LTVs.

Small, undercapitalised banks offer riskier mortgages.

[Ah-so]

Small, undercapitalised banks offer riskier mortgages.

They aren't necessarily under-capitalised. It is just that the small banks are usually not able to use IRB models and so their relative competitive advantage is in riskier mortgages.

But firms using IRB models nearly always hold proportionality lower capital against the mortgages than the big banks do.

[Neverwhere]

One for those of you who take a geeky interest in risk weighted assets. The PRA has released a consultation paper that seems to have the implication that banks will need to hold higher RWAs against mortgages and will be a back-stop against those banks that really abuse the Basel modelling approach to IRB.

http://www.bankofengland.co.uk/pra/Pages/publications/cp/2016/cp2916.aspx

Summary of proposals

The PRA proposes to amend the SS11/13 such that firms would be expected to adopt probability of default modelling approaches for their residential mortgage portfolios that avoid the lack of risk capture identified in the point-in-time and through-the-cycle models currently used by firms, and instead calibrate their models using a consistent and appropriate assumption for the level of model cyclicality.

The PRA also proposes to expect firms not to apply a house price fall assumption of less than 25% in their UK residential mortgage loss given default models.

Meaning that IRB approaches may continue to be pro-cyclical if falls exceed 25%?

[Neverwhere]

Potentially interesting re. BTL, although considering what the BCBS have proposed elsewhere it may be taken out of IRB altogether:

2.15 In order to calibrate PD models firms need to calculate long-run average default rates through an economic cycle using a representative mix of good and bad years. In the United Kingdom, the PRA considers this to include the recession in the early 1990s. Because firms do not have detailed granular data for their models going back to this period, firms need to infer default rates for years where sufficient internal data are not available.

[. . .]

2.21 The detailed changes to SS11/13 are set out in appendix 1, of which the material elements are:

[. . .]

(vi) Paragraph 10.14 would make explicit that the economic cycle used for calibrating long-run averages would be expected to include economic conditions equivalent to that experienced in the early 1990s in the United Kingdom

[. . .]

(viii) For some residential mortgages, such as buy-to-let, self-certification and sub-prime, the absence of external data over a representative economic cycle represents an additional challenge. For such portfolios, firms would be expected to model how book level default rates would have performed under the economic conditions experienced in a representative economic cycle (paragraph 10.15). This modelling would be expected to include a level of conservatism and the PRA would then, as part of its review of these models, assess whether the degree by which long-run average PDs of these assets exceeds prime long-run average PDs is sufficient (paragraph 10.16);

[Ah-so]

Meaning that IRB approaches may continue to be pro-cyclical if falls exceed 25%?

If house prices fall 25% then the lenders would need to further increase capital to prepare for another 25% (possibly).

I think that this has been introduced because all models will have to include an input for house price falls, but these vary and presumably some are assuming a less than 25% fall in their models. Putting a back-stop on aggressive assumptions seems pretty sensible.

[Neverwhere]

If house prices fall 25% then the lenders would need to further increase capital to prepare for another 25% (possibly).

I think that this has been introduced because all models will have to include an input for house price falls, but these vary and presumably some are assuming a less than 25% fall in their models. Putting a back-stop on aggressive assumptions seems pretty sensible.

Absolutely. I'm just thinking about the London market and considering that from where we are now falls could significantly exceed 25%, and trying to work out how the IRB approach might interact with that scenario.

[spyguy]

They aren't necessarily under-capitalised. It is just that the small banks are usually not able to use IRB models and so their relative competitive advantage is in riskier mortgages.

But firms using IRB models nearly always hold proportionality lower capital against the mortgages than the big banks do.

Do we have many small banks left?

The number of small, medium and big banks that have gone t1ts up over the last 10+ years is massive.

[Ah-so]

Do we have many small banks left?

The number of small, medium and big banks that have gone t1ts up over the last 10+ years is massive.

There are quite a lot of them, once you include specialist lenders and building societies. The likes of Paragon, Aldemore,Yorkshire BS have quite a lot of the market.

[spyguy]

There are quite a lot of them, once you include specialist lenders and building societies. The likes of Paragon, Aldemore,Yorkshire BS have quite a lot of the market.

Ah Paragon and Aldemore are not banks. They are unregulated speculative investments.

Yorkshire BS is NAB big mistake.

[Si1]

And Yorkshire bs is now Yorkshire bs Group. It's still a mutual but seems to try to operate as a PLC, runs seperate btl lending arm and IT consultancy.

[spyguy]

You sure about Paragon? They seem very yo not use 'bank' on their website.

[spyguy]

Did you see my link?

EDIT: The 'who we are' is probably clearer...

http://www.paragonbank.co.uk/who-we-are/

Its the 'Paragon group of companies' that muddies the water. Hows the regulated bank bit separated from the unregulated bit? What is the banks size , relative to the rest?

It seems a regulation hellhole. Id have forced the bank to be separate from the rest.

[Ah-so]

Ah Paragon and Aldemore are not banks. They are unregulated speculative investments.

Yorkshire BS is NAB big mistake.

You're right that the BTL lender Paragon is not technically a bank, but Paragon Group has a banking licence and takes deposits so the regulator considers it a bank for regulatory purposes. .

Aldemore is a bank.

The Yorkshire BS is a building society and should not be confused with Yorkshire Bank, part of the NAB group.