Basel IV – Digital Finance Analytics (DFA) Blog

Australia Leads the Way on ‘Basel IV’, Banks Well Placed

The Australian Prudential Regulation Authority (APRA) has continued to burnish its conservative credentials by being the first regulator to publish proposals to implement the Basel III endgame standard, also known as “Basel IV”, says Fitch Ratings.

APRA proposes to implement a stricter variant of the international reforms published last December on an accelerated timeline. We do not expect Australian banks to have any difficulty meeting the proposed capital requirements, but changes in risk weights could lead to modest adjustments to the mix of their loan portfolios over time.

The changes published in two discussion papers on 14 February 2018 amend elements of the “unquestionably strong” requirements that were first outlined in July 2017. APRA proposes to adopt tougher risk-weighted asset (RWA) standards, including the controversial output floor, which will prevent modelled capital requirements from falling below a certain level, by 1 January 2021, without the five-year phase-in period that most countries are likely to allow. The discussion paper details extensive “gold-plating” of capital requirements, with increased requirements for retail exposures, loan and credit card facilities, and investment and interest-only residential mortgages compared to international norms. Similarly, APRA has proposed a minimum 4% leverage ratio, which is above the globally agreed 3% minimum.

Australian banks are already well on the way to meeting APRA’s “unquestionably strong” requirements, and the regulator has indicated that the latest changes should not require additional capital beyond what is needed to meet those requirements. However, the “unquestionably strong” 10.5% common equity Tier 1 ratio target for banks using internal models is likely to be reduced to offset the increase in RWAs that will most likely result from the changes in these discussion papers. This is in line with what Fitch had expected when the target was first announced. APRA said it will conduct a quantitative impact study with the banks before finalising the changes to ensure that they are in line with the benchmarks outlined in July 2017.

APRA’s proposed minimum leverage ratio should have no impact on the banks’ capital holdings – reported leverage ratios are already well above the 4% minimum at banks using the advanced models, and well above 3% at the standardised banks.

The proposed standards are unlikely to result in significant changes to banks’ business models, but they may lead to modest adjustment to loan portfolios. For example, the proposed changes to residential and, possibly, commercial mortgage risk-weights would increase charges for riskier loan types, such as investor and interest-only lending, relative to amortising owner-occupier loans for both advanced and standardised banks. This may result in banks increasing pricing for the riskier loan types or emphasising growth in owner-occupier loans. Similarly, increased capital charges for off-balance sheet facilities and credit cards could result in banks repricing products, or raising fees.

APRA also plans to further narrow the gap between the mortgage risk-weights under the advanced and standardised approaches, which is likely to provide some competitive benefit to the smaller banks that use the standardised approaches. APRA expects to release draft revised prudential standards after it has received feedback on its proposals, which is due by mid-May 2018.

RBNZ Reviews Capital Adequacy Framework

The New Zealand Reserve Bank has announced it is undertaking a comprehensive review of the capital adequacy framework applying to locally incorporated registered banks over 2017/18. The aim of the review is to identify the most appropriate framework for setting capital requirements for New Zealand banks, taking into account how the current framework has operated and international developments in bank capital requirements.

The Capital Review will focus on the three key components of the current framework:

The definition of eligible capital instruments

The measurement of risk

The minimum capital ratios and buffers

The purpose of this Issues Paper is to provide stakeholders with an outline of the areas of the capital adequacy framework that the Reserve Bank intends to cover in the Capital Review, and invite stakeholders to provide initial feedback on the intended scope of the review, and issues that might warrant particular attention. As feedback is received and decisions are made, some of these issues might fall away or be given a lower priority.

Detailed consultation documents on policy proposals and options for each of the three components will be released later in 2017, with a view to concluding the review by the first quarter of 2018.

Basis and framework for capital regulation

The Reserve Bank has powers under the Reserve Bank Act 1989 to impose capital requirements on registered banks. The Reserve Bank exercises these powers to promote the maintenance of a sound and efficient financial system, and to avoid significant damage to the financial system that could result from the failure of a registered bank.

The capital adequacy framework for locally incorporated registered banks is set out mainly in documents BS2A and BS2B of the Reserve Bank’s Banking Supervision Handbook. The framework is based on, but not identical to, an international set of standards produced by the Basel Committee on Banking Supervision.

The framework imposes minimum capital ratios. These are ratios of eligible capital to loans and other exposures. Exposures are adjusted (risk-weighted) so that more capital is required to meet the minimum requirement if the bank has riskier exposures.

The high-level policy options raised in this Issues Paper have the potential to result in reasonably significant changes to the New Zealand capital framework. It is expected, however, that any changes are likely to occur within a Basel-like framework.

The Reserve Bank invites submissions on this Issues Paper by 5pm on 9 June 2017

New Report On Mortgage Industry With JP Morgan Released

The report, released today highlights that property investors will be hit hard as banks re-price their mortgages.

Volume 24 of the mortgage report, a collaboration between J.P. Morgan and Digital Finance Analytics (DFA), explores how to practically define the term ‘materially dependent on property cash flows’ and looks to translate that into potential incremental capital requirements for the Australian major banks. The report also considers different re-pricing strategies and competitive dynamics, particularly around the issue of dynamic Loan-to-Value Ratios (LVR).

Significant changes are afoot for investor loans defined as being ‘materially dependent on property cash flows’ to repay the loan. Amidst the transition to Basel 4, these mortgages will see the most extreme effects on their capital intensity and pricing – with capital levels somewhere between 3x and 5x current requirements, which could have a significant impact on pricing of investor loans down the track.

The report draws heavily on modelling completed by Digital Finance Analytics from our household surveys, as presented in the recently published The Property Imperative 8, available here. Our survey is based on a rolling sample of 52,000 households and is the largest currently available. It includes data to end February 2017.

“The dispersion of impacts across the portfolio highlights the fact that assessing the mortgage by Probability of Default band or LVR band isn’t necessarily ‘good enough’. Although banks may have access to significant pools of data, the new regulatory regime is forcing them to become ever-more granular in their analysis – top-down portfolio analytics just won’t cut it anymore,” said Martin North, principal, DFA.

“Rather than managing the portfolio with ‘macro-prudential’ drivers, banks need to move to the other end of the analysis spectrum and become ‘micro-prudential’,” Mr North concluded.

Unfortunately because of compliance issues, the JPM report itself is only available direct from them, and not via DFA.

DFA is not authorized nor regulated by ASIC and as such is not providing investment advice. DFA contributors are not research analysts and are neither ASIC nor FINRA regulated. DFA contributors have only contributed their analytic and modeling expertise and insights. DFA has not authored any part of this report.

Basel IV Proposals May Lead to EU/US Divergence

Basel Committee discussions on “Basel IV” at the end of November may expose deep divisions between national members and could lead to further differentiation between the EU and internationally agreed Basel standards, Fitch Ratings says. This would exacerbate challenges for market participants attempting to compare the capital strength of banks globally and could undermine confidence in a framework aimed at promoting a level playing field.

The latest Basel III amendments (sometimes referred to as “Basel IV”) seek to restrict the use of internal risk models, which are associated with wide divergences in risk-weighted measures, and potentially set overall capital requirements using the revised standardised approaches. EU politicians have expressed significant concerns in their parliamentary submissions ahead of a vote on the EU’s finalisation of Basel III on 24 November. Meanwhile, regulators such as Thomas Hoenig of the US FDIC recently warned against the dilution of the reforms and support tougher equity-based requirements.

If enacted the proposals are likely to increase capital requirements for lower risk-weight portfolios, such as mortgage loans, despite the committee’s intention not to significantly raise capital requirements for banks globally. European banks generally hold larger mortgage portfolios and would be more affected. EU regulators seek to balance the economic consequence of any increase in capital requirements against the challenges banks face to boost capital internally in light of negative yields and low growth prospects. EU lawmakers say they are aiming for “global standards with local calibration”.

In contrast, US banks would be less affected as they typically sell their mortgage loans to US government agencies and larger firms already hold capital on the higher of the standardised and internal ratings-based (IRB) approaches. US supervisors also might be more willing to set higher capital standards due to the more supportive operating environment and lower reliance on loan-based finance than the EU.

These tensions are likely to be reflected in two key areas in the 28-29 November Basel discussions. The proposal for a capital floor based on a certain percentage of the standardised approaches has the greatest potential for EU divergence. Whether the committee sets an overall floor or one for each risk category is important. EU lawmakers have publicly protested that a floor would severely punish stable banks focused on traditional low-risk lending, while banks with higher-risk assets would be less affected. The impact on European mortgage banks may be a key consideration here as residential mortgage loans are potentially most vulnerable, especially if a permanent capital floor were to be applied on a risk-category basis.

We believe outright elimination of internal ratings models for certain credit exposures is unlikely. The speech by Stefan Ingves, chair of the Basel Committee on 10 November did not mention this. EU lawmakers are concerned about the impact on credit flow to the real economy and have argued that working to enhance trust in IRB models is preferable to abandoning them for portfolios such as large corporates. We believe there would be a generous transitional phase-out period even if the committee does move away from using models for these portfolios.

We think it is more likely that specific constraints will be introduced to reduce risk-weight variability, as proposed for modelling mid-sized corporate and retail exposures, and to ensure a minimum level of conservatism, for example through the introduction of specific loss given default floors in models for residential and commercial real estate exposures. But other measures, such as a model floor for unconditionally cancellable commitments, which may effectively increase capital charges for these exposures, face opposition from both sides of the Atlantic due to concerns it would constrain banks’ lending capacity.

Revised Model-Based Market Risk Rules Costly for Banks – Fitch

The overhaul of the internal models approach – used by most banks with large trading books to calculate market risk capital requirements – will be costly, says Fitch Ratings. The Basel Committee on Banking Supervision’s revised market risk framework, published in January and effective from 2019, fundamentally changes the approach.

The model revisions should improve risk assessment capabilities, lead to higher capital charges for hard-to-model trading positions and make it easier to compare banks’ results. But the model approval process and governance are being thoroughly revised and implementing the changes will require considerable investment in technology and risk management.

Banks will need to obtain approval for internal models desk by desk, rather than bank-wide. This will make it easier for supervisors to decline approval for a particular trading desk, if, for example, the desk is unable to satisfy model validation criteria due to back-testing failures or an inability to properly attribute profits and losses across products. But Fitch thinks costs associated with building and running the more sophisticated models will be high.

Instead of running a single bank-wide model for a range of stressed and unstressed risk factors, multiple new models will need to be built, validated and run daily. This will multiply the number of model reviews and operational runs and add to subsequent data analysis and reporting procedures. Additional risk personnel will be required for review, oversight, and reporting purposes.

The amount of regulatory capital models-based banks will need to cover potential market risks following the revisions is uncertain. The Basel Committee’s latest studies show that, for a sample of 12 internationally active banks with large trading books, all of which provided high-quality data, market risk capital charges under the revised approach were 28% higher. But for a broader sample of 44 banks using internal models, the median market risk capital requirements fell by 3% under the revised models.

Fitch thinks the result for the 12 banks could reflect greater concentrations of less liquid credit positions that require more capital, or larger trading positions lacking observable transaction prices, which are subject to a stressed capital add-on. Banks facing higher charges under the regime may re-assess whether certain activities remain profitable.

The new internal models approach replaces value at risk (VaR) with an expected shortfall (ES) measure. VaR does not capture the tail risk of loss distribution, which can arise during significant market stress. The use of ES models for regulatory capital is positive for bank creditors because they could lead to better capitalisation of tail-risk loss events and might motivate risk managers to limit trading portfolios that could lead to outsized losses.

When calculating ES measures, banks will have to use variable market liquidity horizons – to a maximum of 120 days for complex credit products, against the current fixed 10-day period. We think model inputs will be more realistic, by acknowledging that some instruments take longer to sell or hedge without affecting prices. ES will also constrain recognition of diversification and hedging benefits, extensively used in VaR models to reduce capital charges. We think this will make model outputs more prudent and force banks to better capitalise potential trading losses.

Structural flaws in the way banks calculated capital charges for market risk were exposed during severe market stresses in 2008-2009. The Basel Committee subsequently undertook a fundamental review of the trading book. The original proposals were watered down, but we think the final revised minimum capital standards for model-driven market risk are positive for creditors because improved model standards and more prudent methods employed to capture risk should mean trading risks are more accurately capitalised.

The Capital Schmozzle

“schmozzle (plural schmozzles). (informal) A disorganized mess; (informal) A melee”.

When the FSI inquiry was handed down last year, with recommendations mostly later accepted by Government, a cornerstone was to avoid the risk of a failing bank needing to be bailed out by tax payers, as happened in a number of countries during the GFC.

A year later, we can look back to see significant changes to the current capital rules for “Advanced” banks (those that use their own approved internal models) and significant capital raisings of more than $30bn by the industry. This has translated into higher interest rates on mortgages, especially investment property loans.

Currently underway are discussions about the next iteration of the capital rules, with the expectations that “Advanced” banks’ rules will be tightened, and the rules for other banks will get more complex, with capital ratios being determined for example by the loan-to-value (LVR) of loans, as well as differentiation between loans serviced by income, and those materially serviced by rental income flows.

APRA last week said they would look to encourage more banks to move towards the “Advanced” methods, with a series of potential interim steps, at the time when the rules are under review. They also said that the current counter cyclical buffer would be set to zero.

Five Australian Banks have “Advanced” capital management, and a number of other banks are already on the journey to “Advanced” capital methods; but it is a complex and twisted path, and the destination is not yet clear. Better data, models and systems are required to meet the necessary hurdles.

What is likely though is that the journey to hold more capital is far from over, whether “Advanced” or “Standard”, and that the light between the two systems is closing, as the “Standard” system gets more complex, and the “Advanced” ratios are lifted higher.

We think that there will be only limited upside to be gained from moving to “Advanced”, as the gap closes, and complexity increases in the standard approach. We also think significant further capital will be required – some are suggesting an additional $30-40bn in the next couple of years, enough to force mortgage prices across the board significantly higher again. As these adjustments are essentially across the board, to a greater or lesser extent, we doubt that smaller players will actually get much differential benefit. Indeed, if investment mortgages require higher capital still. (that depends on the definition of what is “material” – yet to be made clear; and the LVR), some banks could need much more capital than is currently assumed.

It is also worth noting that spreads on overseas bank capital raisings are rising, indeed, spreads are wider now across the market, as we noted last week.

So what is the potential impact of lifting capital ratios? We already mentioned the uplift in mortgage rates, as banks seek to cover the additional costs involved. Households should expect to pay more. Shareholders may also have to take a haircut in future returns, as the economics of banks change. The super profits banks have enjoyed may be trimmed a little.

But a recent paper from the Bank of England has also highlighted that lifting capital may not reduce systemic risks much. The study, a working paper “Capital requirements, risk shifting and the mortgage market” looked at what happened when capital ratios were lifted. They found that whilst the average value of a loan made fell a little, there was no reduction in higher risk lending, despite the requirements for higher capital, because the lender was looking to protect overall margins and still chose to take more risks. If this is true, higher capital requirements does not necessarily reduced systemic risks. Banks may still need Government support in a crisis.

But wait, was that not the whole reason for lifting capital ratios in the first place? Looks like a schmozzle to me!

BIS Capital Proposals Revised Again, LVR’s and Investment Loans Significantly Impacted

The second consultative document on Revisions to the Standardised Approach for credit risk has been released for discussion.

There are a number of significant changes to residential property risk calculations . These guidelines will eventually become part of “Basel III/IV”, and will apply to banks not using their internal assessments (which are also being reviewed separately).

First, risk will be assessed by loan to value ratios, with higher LVR’s having higher risk weights. Second, investment property will have a separate a higher set of LVR related risk-weights. Third, debt servicing ratios will not directly be used for risk weights, but will still figure in the underwriting assessments.

There are also tweaks to loans to SME’s.

These proposals differ in several ways from an initial set of proposals published by the Committee in December 2014. That earlier proposal set out an approach that removed all references to external credit ratings and assigned risk weights based on a limited number of alternative risk drivers. Respondents to the first consultative document expressed concerns, suggesting that the complete removal of references to ratings was unnecessary and undesirable. The Committee has decided to reintroduce the use of ratings, in a non-mechanistic manner, for exposures to banks and corporates. The revised proposal also includes alternative approaches for jurisdictions that do not allow the use of external ratings for regulatory purposes.

The proposed risk weighting of real estate loans has also been modified, with the loan-to-value ratio as the main risk driver. The Committee has decided not to use a debt service coverage ratio as a risk driver given the challenges of defining and calibrating a global measure that can be consistently applied across jurisdictions. The Committee instead proposes requiring the assessment of a borrower’s ability to pay as a key underwriting criterion. It also proposes to categorise all exposures related to real estate, including specialised lending exposures, under the same asset class, and apply higher risk weights to real estate exposures where repayment is materially dependent on the cash flows generated by the property securing the exposure.

This consultative document also includes proposals for exposures to multilateral development banks, retail and defaulted exposures, and off-balance sheet items.The credit risk standardised approach treatment for sovereigns, central banks and public sector entities are not within the scope of these proposals. The Committee is considering these exposures as part of a broader and holistic review of sovereign-related risks.

Comments on the proposals should be made by Friday 11 March 2016.

Looking in more detail at the property-related proposals, the following risk weights will be applied to loans against real property:

which are finished properties
covered by a legal mortgage
with a valid claim over the property in case of default
where the borrower has proven ability to repay – including defined DSR’s
with a prudent valuation (and in a falling market, a revised valuation), to derive a valid LVR
all documentation held

If all criteria a met the following risk weights are proposed.

For residential real estate exposures to individuals with an LTV ratio higher than 100% the risk weight applied will be 75%. For residential real estate exposures to SMEs with an LTV ratio higher than 100% the risk weight applied will be 85%. If criteria are not met, then 150% will apply.

Turning to investment property, where cash flow from the property is the primary source of income to service the loan

Commercial property will have different ratios, based on counter party risk weight.

But again, those properties serviced by cash flow have higher weightings.

Development projects will be rated at 150%.

Bearing in mind that residential property today has a standard weight of 35%, it is clear that more capital will be required for high LVR and investment loans. As a result, if these proposals were to be adopted, then borrowers can expect to pay more for investment loans, and higher LVR loans.

It will also increase the burden of compliance on banks, and this will likely increase underwriting costs. Finally, whilst ongoing data on DSR will not be required, there is still a need to market-to-market in a falling market to ensure the LVR’s are up to date. This means, that if property valuations fall significantly, higher risk weights will start to apply, the further they fall, the larger the risk weights.

Finally, it continues the divergence between the relative risks of investment and owner occupied loans, the former demanding more capital, thus increasing the differential pricing of investment loans.

The Committee notes that the SA is a global minimum standard and that it is not possible to take into account all national characteristics in a simple approach. As such, national supervisors should require a more conservative treatment if they consider it necessary to reflect jurisdictional specificities. Furthermore, the SA is a methodology for calculating minimum risk-based capital requirements and should in no way be seen as a substitute for prudent risk management by banks.

Now, some will argue that in Australia, this will not impact the market much, as the major banks use their own internal models, however, as these are under review (with the intent of closing the gap somewhat with the standard methods used by the smaller players) expect the standard models to inform potential changes in the IRB set. Also, it is not clear yet whether banks who use lenders mortgage insurance for loans above 80% will be protected from the higher capital bands, though we suspect they may not. Non-bank lenders may well benefit as they are not caught by the rules, although capital market pricing may well change, and impact them at a second order level. We will be interested to see how local regulators handle the situation where an investment loan is partly serviced by income from rentals, and partly from direct income, which rules should apply – how will “materially dependent” be interpreted?

Basel IV – Is More Complexity Better?

In December 2014, The Bank For International Settlements issued proposed Revisions to the Standardised Approach for credit risk for comment. It proposes an additional level of complexity to the capital calculations which are at the heart of international banking supervision. Comments on the proposals were due by 27 March 2015. These latest proposals, which have unofficially been dubbed “Basel IV”, is a continuation of the refining of the capital adequacy ratios which guide banking supervisors and relate to the standardised approach for credit risk. It forms part of broader work on reducing variability in risk-weighted asset. We want to look in detail at the proposals relating to residential real estate, because if adopted they would change the capital landscape considerably. Note this is separate from the proposal relating to the adjustment of IRB (internal model) banks. Whilst it aspires to simplify, the proposals are, to put it mildly, complex

For the main exposure classes under consideration, the key aspects of the proposals are:

Bank exposures would no longer be risk-weighted by reference to the external credit rating of the bank or of its sovereign of incorporation, but they would instead be based on a look-up table where risk weights range from 30% to 300% on the basis of two risk drivers: a capital adequacy ratio and an asset quality ratio.
Corporate exposures would no longer be risk-weighted by reference to the external credit rating of the corporate, but they would instead be based on a look-up table where risk weights range from 60% to 300% on the basis of two risk drivers: revenue and leverage. Further, risk sensitivity would be increased by introducing a specific treatment for specialised lending.
The retail category would be enhanced by tightening the criteria to qualify for the 75% preferential risk weight, and by introducing a fallback subcategory for exposures that do not meet the criteria.
Exposures secured by residential real estate would no longer receive a 35% risk weight. Instead, risk weights would be determined according to a look-up table where risk weights range from 25% to 100% on the basis of two risk drivers: loan-to-value and debt-service coverage ratios.
Exposures secured by commercial real estate are subject to further consideration where two options currently envisaged are: (a) treating them as unsecured exposures to the counterparty, with a national discretion for a preferential risk weight under certain conditions; or (b) determining the risk weight according to a look-up table where risk weights range from 75% to 120% on the basis of the loan-to-value ratio.
The credit risk mitigation framework would be amended by reducing the number of approaches, recalibrating supervisory haircuts, and updating corporate guarantor eligibility criteria.

Real Estate Capital Calculation Proposals

The recent financial crisis has demonstrated that the current treatment is not sufficiently risk-sensitive and that its calibration is not always prudent. In order to increase the risk sensitivity of real estate exposures, the Committee proposes to introduce two specialised lending categories linked to real estate (under the corporate exposure category) and specific operational requirements for real estate collateral to qualify the exposures for the real estate categories.

Currently the standardised approach contains two exposure categories in which the risk-weight treatment is based on the collateral provided to secure the relevant exposure, rather than on the counterparty of that exposure. These are exposures secured by residential real estate and exposures secured by commercial real estate. Currently, these categories receive risk weights of 35% and 100%, respectively, with a national discretion to allow a preferential risk weight under certain strict conditions in the case of commercial real estate.

Residential Owner Occupied Real Estate

In order to qualify for the risk-weight treatment of a residential real estate exposure, the property securing the mortgage must meet the following operational requirements:

Finished property: the property securing a mortgage must be fully completed. Subject to national discretion, supervisors may apply the risk-weight treatment for loans to individuals that are secured by an unfinished property, provided the loan is for a one to four family residential housing unit.
Legal enforceability: any claim (including the mortgage, charge or other security interest) on the property taken must be legally enforceable in all relevant jurisdictions. The collateral agreement and the legal process underpinning the collateral must be such that they provide for the bank to realise the value of the collateral within a reasonable time frame.
Prudent value of property: the property must be valued for determining the value in the LTV ratio. Moreover, the value of the property must not be materially dependent on the performance of the borrower. The valuation must be appraised independently using prudently conservative valuation criteria and supported by adequate appraisal documentation.

The current standardised approach applies a 35% risk weight to all exposures secured by mortgage on residential property, regardless of whether the property is owner-occupied, provided that there is a substantial margin of additional security over the amount of the loan based on strict valuation rules. Such an approach lacks risk sensitivity: a 35% risk weight may be too high for some exposures and too low for others. Additionally, there is a lack of comparability across jurisdictions as to how great a margin of additional security is required to achieve the 35% risk weight.

In order to increase risk sensitivity and harmonise global standards in this exposure category, the Committee proposes to introduce a table of risk weights ranging from 25% to 100% based on the loan-to-value (LTV) ratio. The Committee proposes that the risk weights derived from the table be applied to the full exposure amount (ie without tranching the exposure across different LTV buckets).

The Committee believes that the LTV ratio is the most appropriate risk driver in this exposure category as experience has shown that the lower the outstanding loan amount relative to the value of the residential real estate collateral, the lower the loss incurred in the event of a default. Furthermore, data suggest that the lower the outstanding loan amount relative to the value of the residential real estate collateral, the less likely the borrower is to default. For the purposes of calculating capital requirements, the value of the property (ie the denominator of the LTV ratio) should be measured in a prudent way. Further, to dampen the effect of cyclicality in housing values, the Committee is considering requiring the value of the property to be kept constant at the value calculated at origination. Thus, the LTV ratio would be updated only as the loan balance (ie the numerator) changes.

The LTV ratio is defined as the total amount of the loan divided by the value of the property. For regulatory capital purposes, when calculating the LTV ratio, the value of the property will be kept constant at the value measured at origination, unless an extraordinary, idiosyncratic event occurs resulting in a permanent reduction of the property value. Modifications made to the property that unequivocally increase its value could also be considered in the LTV. The total amount of the loan must include the outstanding loan amount and any undrawn committed amount of the mortgage loan. The loan amount must be calculated gross of any provisions and other risk mitigants, and it must include all other loans secured with liens of equal or higher ranking than the bank’s lien securing the loan. If there is insufficient information for ascertaining the ranking of the other liens, the bank should assume that these liens rank pari passu with the lien securing the loan.

In addition, as mortgage loans on residential properties granted to individuals account for a material proportion of banks’ residential real estate portfolios, to further increase the risk sensitivity of the approach, the Committee is considering taking into account the borrower’s ability to service the mortgage, a proxy for which could be the debt service coverage (DSC) ratio. Exposures to individuals could receive preferential risk weights as long as they conform to certain requirement(s), such as a ‘low’ DSC ratio. This ratio could be defined on the basis of available income ‘net’ of taxes. The DSC ratio would be used as a binary indicator of the likelihood of loan repayment, ie loans to individuals with a DSC ratio below a certain threshold would qualify for preferential risk weights. The threshold could be set at 35%, in line with observed common practice in several jurisdictions. Given the difficulty in obtaining updated borrower income information once a loan has been funded, and also given concerns about introducing cyclicality in capital requirements, the Committee is considering whether the DSC ratio should be measured only at loan origination (and not updated) for regulatory capital purposes.

The DSC ratio is defined as the ratio of debt service payments (including principal and interest) relative to the borrower’s total income over a given period (eg on a monthly or yearly basis). The DSC ratio is defined using net income (ie after taxes) in order to focus on freely disposable income. The DSC ratio must be prudently calculated in accordance with the following requirements:

Debt service amount: the calculation must take into account all of the borrower’s financial obligations that are known to the bank. At loan origination, all known financial obligations must be ascertained, documented and taken into account in calculating the borrower’s debt service amount. In addition to requiring borrowers to declare all such obligations, banks should perform adequate checks and enquiries, including information available from credit bureaus and credit reference agencies.
Total income: income should be ascertained and well documented at loan origination. Total income must be net of taxes and prudently calculated, including a conservative assessment of the borrower’s stable income and without providing any recognition to rental income derived from the property collateral. To ensure the debt service is prudently calculated, the bank should take into account any probable upward adjustment in the debt service payment. For instance, the loan’s interest rate should (for this purpose) be increased by a prudent margin to anticipate future interest rate rises where its current level is significantly below the loan’s long-term level. In addition, any temporary relief on repayment must not be taken into account for purposes of the debt service amount calculation.

Notwithstanding the definitions of the DSC and LTV ratios, banks must, on an ongoing basis, have a comprehensive understanding of the risk characteristics of their residential real estate portfolio.

The risk weight applicable to the full exposure amount will be assigned, as determined by the table below, according to the exposure’s loan-to-value (LTV) ratio, and in the case of exposures to individuals, also taking into account the debt service coverage (DSC) ratio. Banks should not tranche their exposures across different LTV buckets; the applicable risk weight will apply to the full exposure amount. A bank that does not have the necessary LTV information for a given residential real estate exposures must apply a 100% risk weight to such an exposure.

Some points to note.

Differences in real estate markets, as well as different underwriting practices and regulations across jurisdictions make it difficult to define thresholds for the proposed risk drivers that are meaningful in all countries.
Another concern is that the proposal uses risk drivers prudently measured at origination. This is mainly to dampen the effect of cyclicality in housing values (in the case of LTV ratios) and to reduce regulatory burden (in the case of DSC ratios). The downside is that both risk drivers can become less meaningful over time, especially in the case of DSC ratios, which can change dramatically after the loan has been granted.
The DSC ratio is defined using net income (ie after taxes) in order to focus on freely disposable income. That said, the Committee recognises that differences in tax regimes and social benefits in different jurisdictions make the concept of ‘available income’ difficult to define and there are concerns that the proposed definition might not be reflective of the borrower’s ability to repay a loan. Further, the level at which the DSC threshold ratio has been set might not be appropriate for all borrowers (eg high income) or types of loans (eg those with short amortisation periods). Therefore the Committee will explore whether using either a different definition of the DSC ratio (eg using gross income, before taxes) or any other indicator, such as a debt-to-income ratio, could better reflect the borrower’s ability to service the mortgage.
There are no specific proposal to treat loans that are past-due for more than 90 days.

Investment Loans

Bearing in mind that 35% of all loans are for investment purposes in Australia, the proposals relating to loans for investment purposes are important. So how will they be treated under Basel 4?

There are a number of pointers in the proposals, though its not totally clear in our view. First, we think the proposals would apply to separate loans where repayment is predicated on income generated by the property securing the mortgage, i.e. investment loans rather than a normal loans where the mortgage is linked directly to the underlying capacity of the borrower to repay the debt from other sources. Such loans might fall into a special commercial real estate category, specialist lending category, or a fall back to the unsecured category, each with different sets of capital weights.

The Committee proposes that any exposure secured with real estate that exhibits all of the characteristics set out in the specialised lending category should be treated for regulatory capital purposes as income-producing real estate or as land acquisition, development and construction finance as the case may be, rather than as exposures secured by real estate. Any non-specialised lending exposure that is secured by real estate but does not satisfy the operational requirements should be treated for regulatory capital purposes as an unsecured exposure, either as a corporate exposure or other retail exposure, as appropriate.

Specialised lending exposure, would be defined so if all the following characteristics, either in legal form or economic substance were met:

The exposure is typically to an entity (often a special purpose entity (SPE)) that was created specifically to finance and/or operate physical assets;
The borrowing entity has few or no other material assets or activities, and therefore little or no independent capacity to repay the obligation, apart from the income that it receives from the asset(s) being financed;
The terms of the obligation give the lender a substantial degree of control over the asset(s) and the income that it generates; and
As a result of the preceding factors, the primary source of repayment of the obligation is the income generated by the asset(s), rather than the independent capacity of a broader commercial enterprise.

On the other hand, in order to qualify as a commercial real estate exposure, the property securing the mortgage must meet the same operational requirements as for residential real estate. If the loan is a commercial real estate category, the risk weight applicable to the full exposure amount will be assigned according to the exposure’s loan-to-value (LTV) ratio, as determined in the table below. Banks should not tranche their exposures across different LTV buckets; the applicable risk weight will apply to the full exposure amount. A bank that does not have the necessary LTV information for a given commercial real estate exposure must apply a 120% risk weight.
LTVNote, if this LTV refers to market value, the threshold should be set at a lower level: eg 50%.

Where the requirements are not met, the exposure will be considered unsecured and treated according to the counterparty, ie as “corporate” exposure or as “other retail”. However, in exceptional circumstances for well developed and long established markets, exposures secured by mortgages on office and/or multipurpose commercial premises and/or multi-tenanted commercial premises may be risk-weighted at [50%] for the tranche of the loan that does not exceed 60% of the loan to value ratio. This exceptional treatment will be subject to very strict conditions, in particular:

the exposure does not meet the criteria to be considered specialised lending
the risk of loan repayment must not be materially dependent upon the performance of, or income generated by, the property securing the mortgage, but rather on the underlying capacity of the borrower to repay the debt from other sources
the property securing the mortgage must meet the same operational requirements as for residential real estate
two tests must be fulfilled, namely that (i) losses stemming from commercial real estate lending up to the lower of 50% of the market value or 60% of loan-to value (LTV) based on mortgage-lending-value (MLV) must not exceed 0.3% of the outstanding loans in any given year; and that (ii) overall losses stemming from commercial real estate lending must not exceed 0.5% of the outstanding loans in any given year. This is, if either of these tests is not satisfied in a given year, the eligibility to use this treatment will cease and the original eligibility criteria would need to be satisfied again before it could be applied in the future. Countries applying such a treatment must publicly disclose that these and other additional conditions (that are available from the Basel Committee Secretariat) are met. When claims benefiting from such exceptional treatment have fallen past-due, they will be risk-weighted at [100%].

Implications and Consequences

We should make the point, these are proposals, and subject to change. But it would mean that banks using the standard approach to capital could no longer just go with a 35% weighting, rather they will need to segment the book based on LTV and servicability at a loan by loan level. Investment loans may become more complex and demand higher capital weighting. The required data may be available, as part of the loan origination process, but additional processes and costs will be incurred, and it appears net-net capital buffers will be raised for most players. The capital would be determined using two risk drivers: loan-to-value and debt-service coverage ratios with risk weights ranging from 25 percent to 100 percent. Investment loans may require different treatment, (and the RBNZ discussion paper recently issued may be relevant here, where investment loans are handled on a different basis.)

Finally, a word about those banks on IRB. Currently, under their internal models, they are sitting on an average weighting of around 17% (compared with 35% for standard banks). There are proposals to lift the floor to 20% minimum, and the FSI Inquiry recommend higher. Indeed, Murray called for the big banks to lift the average mortgage risk weighting to a range of 25% to 30%. This would bring them closer to the average mortgage risk weighting used by Australia’s regional banks and credit unions, though as described above, these, in turn, may change. Incidentally, the Bank of England thinks 35% is a good target. Basel 4 will also reduce the variance between standardised banks and those using their own models by requiring the internal models not to deviate from the RWA number in the standardised model by a certain amount: the so-called “capital floor”.

Interestingly the US is focussing on an additional measure, The Tier 1 Common measure, which is unweighted assets to capital, and has set a floor of 5%, or more. The Major Banks in Australia carry real, or non-risk-weighted, equity capital of just 3.7% of assets. Some banks are leveraged over seventy times the equity capital to loans, which is scarily high, but then the RBA (aka the tax payer) would bail them out if they get into trouble, so that’s OK (or not). This means that just $1.70 in assets will now support a $100 loan.

We wonder if the ever more complex models being proposed by Basel are missing the point. Maybe we should be going for something simpler. Many banks of course have invested big in advanced models to squeeze the capital lemon as hard as they can. But stepping back we need approaches which allow greater ability to compare across banks, and more transparent disclosure so we can see where the true risks lay. Certainly capital buffers should be lifted, but we suspect Basel 4, despite the best of intentions, is going down the wrong alley.