Bcbs Risk-Weights

[Ah-so]

10 hours ago, Neverwhere said:

This is how that works out for the IRB banks:

 

To put that into some further context:

 

It is a helpful ballpark figure from 2014, but I think that the RWA percentage for the LBG BTL portfolio is now quite a bit higher. Clearly the bank must have reassessed the amount of capital it holds against that portfolio for some reason 

[Neverwhere]

10 hours ago, Ah-so said:

It is a helpful ballpark figure from 2014, but I think that the RWA percentage for the LBG BTL portfolio is now quite a bit higher. Clearly the bank must have reassessed the amount of capital it holds against that portfolio for some reason 

Interesting, thanks.

I wonder whether that represents a change in the performance of the portfolio itself, or if it might be an attempt to get ahead of Basel III capital requirements?

[spyguy]

1 hour ago, Neverwhere said:

Interesting, thanks.

I wonder whether that represents a change in the performance of the portfolio itself, or if it might be an attempt to get ahead of Basel III capital requirements?

Itll be brutal as bhl loans default, sucking in more capital fir the rest of the book.

We cant be far off Nationwide stating itll do no more btl.

[mrtickle]

In that table, what does "SA" mean please... actually I don't know how to read the whole table

Eg in the "below 50% column" (50% of what, I don't understand) it starts at 15% (of what?) in 2022 and changes to 21.75% (of the same thing?) in 2027.

What does that mean? Is that change from 15% to 21.75% a good thing or a bad thing (for BTLers, and then for banks)?

Also I can't see any way of linking it to the "Lloyds 2014 Pillar 3" table or how it changes the numbers in that table?

TIA.

 

Edited December 8, 2017 by mrtickle

[Neverwhere]

45 minutes ago, mrtickle said:

In that table, what does "SA" mean please... actually I don't know how to read the whole table

Eg in the "below 50% column" (50% of what, I don't understand) it starts at 15% (of what?) in 2022 and changes to 21.75% (of the same thing?) in 2027.

What does that mean? Is that change from 15% to 21.75% a good thing or a bad thing (for BTLers, and then for banks)?

Also I can't see any way of linking it to the "Lloyds 2014 Pillar 3" table or how it changes the numbers in that table?

TIA.

SA stands for "standardised approach".

Banks can use the risk weights set out under this approach to calculate how much capital they have to hold, or they can use an internal ratings based (IRB) approach based on their own models. The higher the risk weight the more capital the bank has to hold against that particular loan, and therefore, effectively, the more expensive it is for them to lend.

Under Basel III risk weights will vary by loan-to-value (LTV) which means, amongst other things, that the amount of capital a bank has to hold against a repayment mortgage will go down as the principal is paid off.

From 1st January 2022 this is what the risk weights for BTL will look like for each LTV band under the standardised approach (currently, IIRC, there is a blanket RW of 35%):

Smaller banks tend to use the standardised approach, hence the challenger banks lending into the riskier end of the BTL market are likely to be using this approach, and so will need to hold more capital against their BTL loan books.

Larger banks tend to use the internal ratings based approach. One of the interesting things that Basel III introduces is an output floor based on the standardised approach, which limits how much IRB can differ from SA in terms of capital requirements. This will be phased in from 1st January 2022, and will start by requiring that the amount of capital held based on an IRB approach will be no less than 50% of the capital that would have been required had the standardised approach been used. This will rise to 72.5% by 1st January 2027.

So in the table below the row in bold is the BTLers' LTV, the row in red is the risk weight that a bank taking the standardised approach would have to apply to that loan, and the rows underneath are the minimum risk weights that a bank using the internal ratings based approach will have to apply during each of the years that the output floor is being phased in:

The 'Residential mortgage exposures by major portfolio' table from Lloyds has an '2014 Average risk weight %' column so as an example we can see that in 2014 Lloyds's internal ratings based approach resulted, at the time, in an average risk weight for their UK buy-to-let portfolio of just 9.33%.

This is all to the best of my knowledge so if anyone else wants to chip in?

[Sancho Panza]

1 hour ago, mrtickle said:

In that table, what does "SA" mean please... actually I don't know how to read the whole table

Eg in the "below 50% column" (50% of what, I don't understand) it starts at 15% (of what?) in 2022 and changes to 21.75% (of the same thing?) in 2027.

What does that mean? Is that change from 15% to 21.75% a good thing or a bad thing (for BTLers, and then for banks)?

Also I can't see any way of linking it to the "Lloyds 2014 Pillar 3" table or how it changes the numbers in that table?

TIA.

 

I believe

SA=Standardized approach-where they lay out risk weights as per the calculus in Basel 3.

As ooposed to IRB where they can base it on the banks own data.

PS I'd wait for Neverwhere or Ahso to confirm. This is one of the best threads on this site.I have to reread it to make myself familiar with it before entering the discussion.

Edited December 8, 2017 by Sancho Panza

[Sancho Panza]

17 minutes ago, Neverwhere said:

SA stands for "standardised approach".

Banks can use the risk weights set out under this approach to calculate how much capital they have to hold, or they can use an internal ratings based (IRB) approach based on their own models. The higher the risk weight the more capital the bank has to hold against that particular loan, and therefore, effectively, the more expensive it is for them to lend.

Under Basel III risk weights will vary by loan-to-value (LTV) which means, amongst other things, that the amount of capital a bank has to hold against a repayment mortgage will go down as the principal is paid off.

From 1st January 2022 this is what the risk weights for BTL will look like for each LTV band under the standardised approach (currently, IIRC, there is a blanket RW of 35%):

Smaller banks tend to use the standardised approach, hence the challenger banks lending into the riskier end of the BTL market are likely to be using this approach, and so will need to hold more capital against their BTL loan books.

Larger banks tend to use the internal ratings based approach. One of the interesting things that Basel III introduces is an output floor based on the standardised approach, which limits how much IRB can differ from SA in terms of capital requirements. This will be phased in from 1st January 2022, and will start by requiring that the amount of capital held based on an IRB approach will be no less than 50% of the capital that would have been required had the standardised approach been used. This will rise to 72.5% by 1st January 2027.

So in the table below the row in bold is the BTLers' LTV, the row in red is the risk weight that a bank taking the standardised approach would have to apply to that loan, and the rows underneath are the minimum risk weights that a bank using the internal ratings based approach will have to apply during each of the years that the output floor is being phased in:

The 'Residential mortgage exposures by major portfolio' table from Lloyds has an '2014 Average risk weight %' column so as an example we can see that in 2014 Lloyds's internal ratings based approach resulted, at the time, in an average risk weight for their UK buy-to-let portfolio of just 9.33%.

This is all to the best of my knowledge so if anyone else wants to chip in?

That's what I was talking about .....

Super post NW and thanks as ever for the clarifications.I'll have a look through when I have time over the weekend after I've refreshed my understanding of risk wieghts through a reread.

I still am amazed how bigger banks can have loan boks skewed by their own data...........it just doesn't make any sense at all.

Edited December 8, 2017 by Sancho Panza

[mrtickle]

48 minutes ago, Neverwhere said:

So in the table below the row in bold is the BTLers' LTV, the row in red is the risk weight that a bank taking the standardised approach would have to apply to that loan, and the rows underneath are the minimum risk weights that a bank using the internal ratings based approach will have to apply during each of the years that the output floor is being phased in:

 

Thank you (both) very much indeed! Very clear explanation and the "aha!" moment was after reading the above para
 

[Neverwhere]

14 minutes ago, mrtickle said:

Thank you (both) very much indeed! Very clear explanation and the "aha!" moment was after reading the above para

54 minutes ago, Sancho Panza said:

That's what I was talking about .....

Super post NW and thanks as ever for the clarifications.

You're welcome

54 minutes ago, Sancho Panza said:

I'll have a look through when I have time over the weekend after I've refreshed my understanding of risk wieghts through a reread.

I still am amazed how bigger banks can have loan boks skewed by their own data...........it just doesn't make any sense at all.

This is exactly what the new output floor looks set to address, although internal models do at least have to be signed off by the bank's national supervisor in any case.

During the transition period supervisors will also have the option of allowing IRB banks to take advantage of a transitional cap that would prevent the output floor from increasing their risk weighted assets by more than 25%. However, this cap will be removed on 1st January 2027, at which point the full 72.5% output floor will apply, so its use could create something of a cliff edge.

Probably worth noting at this point that capital requirements (Capital Adequacy Ratios) are set as a percentage of risk weighted assets. Digging back through the thread (which is something I also have to do periodically ) this is also going to be rising from 8% to 10.5% under Basel III:

On 15/04/2016 at 10:14 AM, Beary McBearface said:

The RWAs allow the calculation of the capital adequacy ratio.

If you had a balance sheet which was all loan assets with a 100% RW then you'd need £8 of capital to lend £100

A RW of over 100% is basically saying that if you were to make loans of that type only then £8 of capital wouldn't be enough to cover your loans, you'd need more than £8.

The risk-weights work in tandem with the capital adequacy ratio (Capital/RWAs).

Post Basel 3 you'll need £10.50 of capital for each £100 of RWAs, so a 150% RW is saying that for lending of that type you'll need to have £15.75 of capital to cover each £100 of lending.

 

[Bland Unsight]

Bit tangential but always worth repeating that bank capital in this context has a very simple meaning. When considering the loans made by the bank what we are interested in is how the bank funds those loans. Whilst the banking system as a whole creates  money ex nihilo de novo (new and from nothing) through the process of simultaneously creating customer debts and deposits (the bank lends you £100, giving you £100 to spend and writing into its ledgers your debt of £100 to them) each individual bank still has to match money in and money out. For example, imagine that my bank (Nationwide, say) lent me £100 billion and I hired a team of Saudi princes and Premier League footballers to spend as much of it as possible on whatever they wanted. I gave my princes and footballers only one rule; they couldn't spend the money with any company or individual that banked with Nationwide. My princes and centre forwards would use the payment cards I had issue to them and each time they spent money the seller's bank would send a demand to Nationwide asking for the payment to be honoured by transferring some cash. Sooner or later the demands to Nationwide from other banks would start to pour in and Nationwide would have to meet the demands. It would have a small amount of cash on hand (about £10bn) and some near cash assets they could sell to provide more cash (again about £10bn) but as the demands for payment continued to pour in they would need more cash. This availability of cash to back the loans made is called funding.

Banks have two key sources of funding; money borrowed (from depositors, bondholders and other banks etc) and equity. Equity includes both the money the lender received from shareholders when the bank issued shares and also any profits which are kept in the bank instead of being distributed to shareholders as dividends.

When talking about capital within the context of the BCBS risk weights and capital adequacy regime what we are interested in is how the lending is funded. To what extent is is funded with money the bank has borrowed (leverage) and to what extent is the lending funded by equity ("capital")?

If enough of the bank's lending is backed by equity ("capital") then it can lose money on its lending but still be in a position to repay its depositors. This is important because in reality the pressure on a bank to ensure that its loans are funded does not come from me and my team of spendthrifts it comes from the temporal mismatch between when depositors (funding the bank with money lent to the bank) want their money back and when borrowers pay back the money that they owe. This was the mechanical detail of how Northern Rock blew up; the short term wholesale money funding NR's loans refused to rollover the money they'd lent to NR (which is the same as withdrawing a deposit) and the amount of money being withdrawn in this way was colossal compared to any ongoing loan repayments from NR's own borrowers.

This temporal mismatch is particularly problematic for any lender who funds 30 year mortgages with deposits that are repayable on demand. The problem of funding the balance sheet with a good balance of debt and equity is further complicated if you are a mutual not private company; it was one thing for Lloyds and the rest to turn to their shareholders in 2008 and raise some more equity via a rights issue, it would be another thing entirely for Nationwide or The Coventry to turn to their retail depositors for a bailout.

Another dynamic that we might see here is that a bank can reduce its need for capital and leverage by shedding lending and the easy way to shed lending in a market where there is competition is to become less competitive. In the BTL game competition between lenders for the pool of borrowers is conducted by lending money at low mortgage interest rates and the way to shed lending is to make sure that the interest rates you offer to customers are higher than the rates offered by the other banks.

[guest_northshore]

Thanks for the analysis/explanations above.

[mrtickle]

Ditto. Very educational, much appreciated.

 

[Ah-so]

On 08/12/2017 at 6:40 PM, Neverwhere said:

Interesting, thanks.

I wonder whether that represents a change in the performance of the portfolio itself, or if it might be an attempt to get ahead of Basel III capital requirements?

I have just checked - it is now 13.06%, so about a 40% increase on the capital they had to hold previously (http://www.lloydsbankinggroup.com/globalassets/documents/investors/2016/2016_lbg_pillar_3_report_v2.pdf (p.63)

It would be nice to think that the PRA realised that holding fewer RWAs for BTL than  prime residential was f***ing insane and told them to get real. I doubt the bank would want to do it voluntarily. And remember, this has happened at a time of rising house prices when LTVs are falling and the mortgage portfolio therefore supposedly becoming safer.

Edited December 9, 2017 by Ah-so

[Neverwhere]

29 minutes ago, Ah-so said:

I have just checked - it is now 13.06%, so about a 40% increase on the capital they had to hold previosly (http://www.lloydsbankinggroup.com/globalassets/documents/investors/2016/2016_lbg_pillar_3_report_v2.pdf (p.63)

It would be nice to think that the PRA realised that holding fewer RWAs for BTL that prime residential was f***ing insane and told them to get real. I doubt the bank would want to do it voluntarily. And remember, this has happened at a time of rising house prices when LTVs are falling and the mortgage portfolio therefore supposedly becoming safer.

It would indeed.

[Neverwhere]

Could be worth keeping an eye out to see if there is anything from the PRA on this?

Quote

IMLA - Buy to let: under pressure

Basel capital requirements for buy-to-let In December 2015, the Basel Committee on Banking Supervision published the second consultative document on Revisions to the Standardised Approach (SA) for credit risk. The consultation proposed changes to the way banks and building societies on the standardised approach would calculate minimum capital ratios. Most contentiously, the consultation proposed the introduction of a separate scale of risk weights for income producing residential property that would see higher LTV buy-to-let loans assigned higher risk weights than unsecured personal loans.

In December 2017 the Basel Committee published Basel III: Finalising post-crisis reforms which contained details of the risk weight scales for firms on the standardised approach for loans secured on ordinary residential property and income producing residential property. The risk weights are higher for income producing property, for example for 80-90% LTV loans the risk weight is 40% for standard residential property and 60% for income producing residential property. Although lending to an individual with a small (undefined) number of rented properties are excluded from the higher risk weights, this change disadvantages buy-to-let lending to portfolio landlords relative to conventional mortgage lending.

Quote

60. To apply the risk-weights in Tables 11, 12, 13 and 14 and the approaches set out in paragraphs 65 and 71, the loan must meet the following requirements:

[. . .]

• Ability of the borrower to repay: the borrower must meet the requirements set according to paragraph 61.
• Prudent value of property: the property must be valued according to the criteria in paragraph 62 for determining the value in the loan to value (LTV) ratio. Moreover, the value of the property must not depend materially on the performance of the borrower.
• Required documentation: all the information required at loan origination and for monitoring purposes must be properly documented, including information on the ability of the borrower to repay and on the valuation of the property.

61. National supervisors should ensure that banks put in place underwriting policies with respect to the granting of mortgage loans that include the assessment of the ability of the borrower to repay. Underwriting policies must define a metric(s) (such as the loan’s debt service coverage ratio) and specify its (their) corresponding relevant level(s) to conduct such assessment.(36) Underwriting policies must also be appropriate when the repayment of the mortgage loan depends materially on the cash flows generated by the property, including relevant metrics (such as an occupancy rate of the property). National supervisors may provide guidance on appropriate definitions and levels for these metrics in their jurisdictions.

62. The LTV ratio is the amount of the loan divided by the value of the property. The value of the property will be maintained at the value measured at origination unless national supervisors elect to require banks to revise the property value downward.(37)

[. . .]

67. When the prospects for servicing the loan materially depend(46) on the cash flows generated by the property securing the loan rather than on the underlying capacity of the borrower to service the debt from other sources, and provided that paragraphs 74 and 75 are not applicable, the exposure will be risk-weighted as follows:

• if the requirements in paragraph 60 are met, according to the LTV ratio as set out in Table 12 below; and
• if any of the requirements of paragraph 60 are not met, at 150%.

The primary source of these cash flows would generally be lease or rental payments, or the sale of the residential property. The distinguishing characteristic of these exposures compared to other residential real estate exposures is that both the servicing of the loan and the prospects for recovery in the event of default depend materially on the cash flows generated by the property securing the exposure.

[. . .]

64. Where the requirements in paragraph 60 are met and provided that paragraphs 67, 74 and 75 are not applicable, the risk weight to be assigned to the total exposure amount will be determined based on the exposure’s LTV ratio in Table 11.

[. . .]

68. The following types of exposures are excluded from the treatment described in paragraph 67 and instead, subject to the treatment described in paragraphs 64 to 66:

• An exposure secured by a property that is the borrower’s primary residence;
• An exposure secured by an income-producing residential housing unit, to an individual who has mortgaged less than a certain number of properties or housing units, as specified by national supervisors;
• An exposure secured by residential real estate property to associations or cooperatives of individuals that are regulated under national law and exist with the only purpose of granting its members the use of a primary residence in the property securing the loans; and
• An exposure secured by residential real estate property to public housing companies and not-for-profit associations regulated under national law that exist to serve social purposes and to offer tenants long-term housing.

 

(36) Metrics and levels for measuring the ability to repay should mirror the FSB Principles for sound residential mortgage underwriting practices (April 2012).

(37) If the value has been adjusted downwards, a subsequent upwards adjustment can be made but not to a higher value than the value at origination.

[. . .]

(46) It is expected that the material dependence condition would predominantly apply to loans to corporates, SMEs or SPVs, but is not restricted to those borrower types. As an example, a loan may be considered materially dependent if more than 50% of the income from the borrower used in the bank’s assessment of its ability to service the loan is from cash flows generated by the residential property. National supervisors may provide further guidance setting out criteria on how material dependence should be assessed for specific exposure types.

 

[spyguy]

6 hours ago, Neverwhere said:

Could be worth keeping an eye out to see if there is anything from the PRA on this?

 

Like all things financial and economic ... theres a lot of words and the like that sound good and look grown up ... til you step back and see the obvious risk - a rental depends on the income  of the tenant!

BTL was a massive exercise in both the idiot LL and idiot banks who somehow thought that the risk was mitigated  by stopping lending to what would have been the OO - young, low income, and lending to the LL, who put some equity in, and leasing it to the young now renter. The mortgage is still paid by the same person but the loan as twice as risky!!!!!!

[Sancho Panza]

10 hours ago, Neverwhere said:

Could be worth keeping an eye out to see if there is anything from the PRA on this?

 

When the LTV's start heading northward on decreasing property values,it'll be interesting to see how./whether they respond,as surely logic dictates there will be some sort of capital call alongside a change in risk weighting.

3 hours ago, spyguy said:

Like all things financial and economic ... theres a lot of words and the like that sound good and look grown up ... til you step back and see the obvious risk - a rental depends on the income  of the tenant!

BTL was a massive exercise in both the idiot LL and idiot banks who somehow thought that the risk was mitigated  by stopping lending to what would have been the OO - young, low income, and lending to the LL, who put some equity in, and leasing it to the young now renter. The mortgage is still paid by the same person but the loan as twice as risky!!!!!!

Also worth noting the taxpayer backed LL's receiving HB on the back of a 5% fiscal deficit.

[Sancho Panza]

On 08/12/2017 at 23:03, Sancho Panza said:

That's what I was talking about .....

Super post NW and thanks as ever for the clarifications.I'll have a look through when I have time over the weekend after I've refreshed my understanding of risk wieghts through a reread.

I still am amazed how bigger banks can have loan boks skewed by their own data...........it just doesn't make any sense at all.

Any talk of risk weighting IRB/SA always makes me think of the education this thread has given to me.Thanks to all involved.
 

As above,worth noting that the data banks are basing IRB on is based on the housing market/economy of the last 20 years...........what could go wrong?

Edited May 23, 2018 by Sancho Panza

[Neverwhere]

On 23/05/2018 at 07:09, spyguy said:

Like all things financial and economic ... theres a lot of words and the like that sound good and look grown up ... til you step back and see the obvious risk - a rental depends on the income  of the tenant!

BTL was a massive exercise in both the idiot LL and idiot banks who somehow thought that the risk was mitigated  by stopping lending to what would have been the OO - young, low income, and lending to the LL, who put some equity in, and leasing it to the young now renter. The mortgage is still paid by the same person but the loan as twice as risky!!!!!!

Exactly!

So now the person actually earning the income to pay the mortgage isn't party to the mortgage contract at all, and can walk away from it entirely on very little notice, and choose to reduce their consumption of housing by joining an existing household in the process.

Not to mention that most of this stuff is non-amortising, or that the larger the portfolio the less likely the BTLer will be to have sufficient other assets or income to cover any shortfalls.

[Neverwhere]

On 23/05/2018 at 11:09, Sancho Panza said:

When the LTV's start heading northward on decreasing property values,it'll be interesting to see how./whether they respond,as surely logic dictates there will be some sort of capital call alongside a change in risk weighting.

I may have gotten the wrong end of the stick but I think that Tier 1 capital at least has to be immediately available to absorb losses on an ongoing basis, so it would be rundown rather than called in?

And as LTVs and RWs should head northwards before many losses were realised then the bank should have already raised additional capital to cope with those losses?

There are also various mechanisms for the FPC and the PRA to require banks to hold additional capital, which they can increase if they think risks are accumulating, or reduce if they want to free up capital for other purposes:

Quote

The Financial Policy Committee’s approach to setting the countercyclical capital buffer

2.4 How does the countercyclical capital buffer fit with the rest of the regulatory framework?

The CCyB is part of a broader framework of equity and other loss-absorbing capital requirements that apply to UK banks, introduced in the aftermath of the financial crisis.(4) Some elements of this framework are currently in effect; other elements are being phased in and will take full effect by 2019.

The framework of risk-based capital requirements comprises three elements.

First, there are minimum levels of going concern capital that must be met at all times, for which banks follow internationally agreed methods for calculation and calibration.

(4) See ‘Supplement to the December 2015 Financial Stability Report: The framework of capital requirements for UK banks’ for further details on the elements of this framework.

 

Going concern capital is able to absorb losses in the normal course of business. The minimum Tier 1 capital requirement is 6%, 4.5 percentage points of which must be met with common equity Tier 1, the highest quality of loss-absorbing capital.

In addition to this common minimum requirement, the PRA applies supervisory requirements that vary by bank (referred to as ‘Pillar 2A’) to compensate for shortcomings in existing measures of risk-weighted assets. In terms of Tier 1 capital, these currently average 2.4% of risk-weighted assets across major UK banks.(1)

Second, there are system-wide buffers of equity, which sum to the ‘combined buffer requirement’ in CRD and PRA rules.(2) These buffers can be used to absorb losses, reducing the need in stressed conditions for banks to withdraw services such as credit provision to the real economy. They are based on internationally-agreed methods for calculation and calibration.

The combined buffer is comprised of:

• The capital conservation buffer, which will be set at 2.5% of risk-weighted assets as of 2019;
• The CCyB, which effectively extends the capital conservation buffer. This will vary through time depending on the risk environment facing banks;
• Additional buffers for banks that are judged to be systemically important for either the global or domestic economy. Banks judged by the Financial Stability Board to be globally systemic will have buffer requirements that range between 1% and 2.5% of risk-weighted assets. Ring-fenced banks and large building societies will be subject to a domestic systemic risk buffer of between 0% and 3% of risk-weighted assets.(3)

Third, in addition to the ‘combined buffer’, some individual banks are subject to a supplementary supervisory buffer, calibrated to capture specific risks they face that are not captured in other buffers. This is the PRA buffer. It applies to banks whose balance sheets are more sensitive to a given level of economic risk than the system as a whole. Banks whose risk management and governance have weaknesses will also be subject to a PRA buffer.

The capital framework also includes a simple leverage ratio, which sets a floor of 3% for the level of Tier 1 capital a bank must have relative to its total (un-weighted) exposures. The Government has given the FPC powers to supplement this leverage floor by making Directions over a countercyclical leverage ratio buffer (CCLB).(4) As a guiding principle, the FPC intends to move the CCLB in line with its setting of the CCyB, with the CCLB rate set at 35% of a bank’s institution-specific CCyB rate. This will help to maintain overall consistency between the risk-weighted capital and leverage ratio frameworks.

(1) In addition, sectoral capital requirements provide the FPC with a means for varying the risk weights on banks’ exposures to residential property, commercial property and other parts of the financial sector. The FPC expects to apply this tool if exuberant lending conditions in one of these sectors pose risks to financial stability. The FPC’s strategy for deploying sectoral capital requirements is described in ‘The Financial Policy Committee’s powers to supplement capital requirements: A Policy Statement’, January 2014.

(2) See Chapter 4, ‘Capital Conservation Measures’, in the Capital Buffers Part of the PRA Rulebook.

(3) For details on the FPC’s proposals for applying systemic risk buffers to domestic systemically important banks, see ‘The Financial Policy Committee’s framework for the systemic risk buffer: A Consultation Paper’, January 2016.

(4) For further details of the leverage ratio framework, see ‘The Financial Policy Committee’s powers over leverage ratio tools: Policy Statement’, July 2015.

 

[. . .]

 

Stage 3: Risks in the financial system become elevated: stressed conditions become more likely

As risks in the financial system become elevated, borrowers are likely to be stretching their ability to repay loans, underwriting standards will generally be lax, and asset prices and risk appetite tend to be high. Often risks are assumed by investors to be low at the very point they are actually high. The distribution of risks to banks’ capital at this stage of the financial cycle might have a ‘fatter tail’, such as that shown by the magenta line in Figure 1. Stressed outcomes are more likely. In such environments, the FPC would expect to increase the CCyB rate beyond the region of 1%. There is no upper bound on the rate that can be set by the FPC. But under EU law and internationally agreed standards, foreign authorities are mandated to reciprocate increases in the rate on UK exposures only up to 2.5% of risk-weighted UK exposures.

The FPC recognises that, while historical relationships contain significant information about the link between risk indicators and crises, it must take into account how those relationships evolve in response to structural changes in the financial system. For instance, active use of the CCyB will itself help to reduce the likely losses in high-risk environments because, in having the capacity to absorb shocks, the banking system will be less of an amplifier of economic shocks than in the past. Moreover, structural reforms introduced since the financial crisis, notably ring-fencing, but also limits on large counterparty exposures and reforms to derivatives markets, will reduce the impact on banks of even elevated risk levels. Historical levels and growth rates of credit and asset prices may also not be a good guide to sustainable future levels and growth rates.

The absence of reliable historical guides to the appropriate CCyB rate in higher-risk environments is another factor driving the FPC’s gradual approach. It is also consistent with the FPC’s intended approach to informing the setting of the CCyB using the annual stress test of major UK banks.

Stage 4: Risks in the financial system crystallise

Should a stress materialise, the FPC may cut the CCyB rate, including where appropriate to 0%. Reducing the CCyB rate pre-emptively before losses have crystallised may reduce banks’ perceived need to hoard capital and restrict lending, with consequent negative impacts for the real economy. And if losses have crystallised, reducing the CCyB allows banks to recognise those losses without having to restrict lending to meet capital requirements. This will help to ensure that capital accumulated when risks were building up can be used, thus enhancing the ability of the banking system to continue to support the economy in times of stress.

FPC decisions on the CCyB are made each quarter and a decision to reduce the CCyB takes immediate effect. The FPC is required to accompany such a decision with an indication of the period during which no increase to the CCyB is expected and its rationale for choosing that period.