HIGHER SCHOOL OF ECONOMICS
National Research University
(Moscow, Russia)
December 8, 2011
Financial Stability Board
Email:
Dear Sirs,
Financial Stability Board Consultation Paper on
Principles for Sound Residential Mortgage Underwriting Practices
The National Research University Higher School of Economics (HSE) is one of the leading Russian economic research and educational establishments actively carrying out the empirical research and policy response analysis.
Banking sector is one of HSE primarily research focus. The continued research in the area of banking within The Laboratory of the Banking Institute of HSE includes the analysis of banking regulation, approaches to banking risks management, banking sector performance analysis, efficient banking branches allocation etc.
The National Research University Higher School of Economics (HSE) is pleased to submit comments on the Consultation Paper ‘Principles for Sound Residential Mortgage Underwriting Practices’ [further on –CP, or Principles’] released by the Financial Stability Board on October 26, 2011 and available at
The comments are presented in two parts:
(a) ‘General Comments’ referring to the Consultation Paper on a whole;
(b)‘Pointwise Comments’ referring to particular points of the Consultation Paper.
We hope our comments would be of use for promoting financial stability at the worldwide and countrywide levels when dealing with the residential mortgages underwriting.
In case of further questions, please, do not hesitate to get in touch through email (), telephone (+7.495.772.95.90, ext. 2081) or fax (+7.495.772.95.90, ext. 2101).
Dr. H. Penikas
Lecturer
Economics Department
General Comments
It is highly demanded to draw the timeline of the document implementation, as well as the timeline for adopting corresponding local regulatory documents.
March 30, 2012 is thought to be the appropriate deadline for Principles’ implementation, whether June 30, 2013 might be the adequate date for local regulation adoption.
Pointwise Comments
P. i – Definitions:
“Equity: Difference between the appraised value of the property and the total claims held against the property.”
To avoid miscommunication when using term ‘equity’ for capitalizing bank’s exposures a term ‘quasi-equity’ is proposed as actually the mortgage borrower does not provide a formal balance sheet and his equity can not be reliably estimated not accounting for his endowment, for example.
“Collateral: The property or property rights upon which the residential mortgage loan is secured”, “Mortgage: A loan that is collateralised against a residential property, including purchase, home equity lines of credit (HELOCs) and refinancings”.
We would welcome terms “property” and “real estate” to be explained in greater detail. Particularly, it is desirable to highlight whether the term “property” is only limited to “real estate” or not.
“A borrower’s underlying income capacity is a key input into effective mortgage underwriting” (P. 2 – Principle 1).
Proper definition of ‘income’ is welcomed, i.e. whether it assumes the possibility of having several sources of income or the term ‘income’ is only limited to the principal place of work.
P. 1 – Introduction - “Internationally agreed Principles will help to strengthen residential mortgage underwriting practices and enable supervisors to more effectively monitor and detect the erosion of underwriting practices particularly when the housing market is booming”.
Apparently, the boom on the housing market can be caused by instability on other markets. That is why it is necessary to differentiate the treatment of countries non-equal by the level of housing market development.
P. 2 – Principle 1 – “While income verification can help to measure a borrower’s “ability to repay”, other financial information (e.g. credit scores, credit registers) can help to measure aborrower’s historical “propensity to repay”.
(1) Those borrowers with zero credit history should not be discriminated in favour of those who have one. It is suggested to use comparability approach for those who do not have a credit history to be treated as a borrower with similar characteristics.
(2) There are countries where in-house written IT procedures for scoring are preferred to the outsourced ones being cheaper and providing minimal risk of confidential data loss. Primarily for these very countries it is advised to list the data and algorithms requirements for providing proper scoring calculation.
Basel II quantitative and qualitative requirements for using IRB-approaches (including the overall statistics history of 5 years and reliable, including no-missing data of 3 years) might be used as the starting point for mortgage data and scoring procedures requirements. Certification of procedures by local Central Banks would provide additional guarantee on the reliability and soundness of the lending institutions.
(3) International research[1] suggests that households and individuals tend to underestimate (deliberately or non) their income received. That is why it is proposed to extend the presented documents list with the data on expenses made (particularly, on durable goods purchases; households tend to keep proving documents to be able to apply for guarantees or discounts).
P. 2 – Principle 1.1 – “For borrowers who are self-employed or entrepreneurs, lenders may require even more extensive history or third-party verification to document income and profit capacity.”
It is recommended to amend the ‘third-party verification’ by the possibility of usingguarantors when applying for mortgage.
P. 2 – Principle 1.3 – “When misinformation provided by the borrower is detected, lenders should have recourse as appropriate to the jurisdiction’s legal system”.
The ‘best practice’ mechanisms of dealing with the mentioned situations is welcomed to be described in greater detail. Break-even clauses for the situations described might be listed being linked to the recovery and resolution plans of the lending banks.
P. 3 – Principle 2.1 – “Jurisdictions should ensure that lenders appropriately assess borrowers’ ability to service and fully repay their loans without causing them undue hardship.”
Foreign exchange risks are to be stress-tested when the mortgage lending occurs in a currency different from the local one. It might be important for a list of countries that demonstrated significant and unexpected local currency depreciation vs. US dollar during 2008-2011, e.g. 20-80% (cf. figure below).
Figure 1. RUR, KZT, UAH, BYR exchange rates to USD depreciated 20-80% from their Jan.01, 2008 values.
For particular occasions (countries with highly volatile exchange rates) local Central Banks may decide to limit mortgage lending to only local currency loans to prevent reinforcement of crisis effects. The recommendation is in line with the one by the European Systemic Risk Board of 21 September 2011 on lending in foreign currencies (ESRB/2011/1), OJ C 342[2] suggesting that “National supervisory authorities are recommended to: 1. monitor levels of foreign currency lending and of private non-financial sector currency mismatches and adopt the necessary measures to limit foreign currency lending;” (cf. Recommendation B, p. 2).
P. 4 – Principle 2.4 – “It is important that customer information be clear, concise, reliable, comparable, easily accessible, timely, and comprehensive (i.e. the information should also take into account the effect of variation in interest rates and the joint effect of the loan and any other product linked to it).”
As the previously mentioned research shows households from the US, Canada, Poland tend to demonstrate similar behavioral patterns when making decisions on vital issues like food or accommodation and consequently on mortgage loan borrowing. In order to work out proper scoring procedures the united international anonymous database for mortgage borrowers is to be created. Financial Stability Board is welcomed to grant access to the data base for research fellows encouraging the investigation of common international trends to be accounted for when granting mortgage loan in a particular country.
P. 4 – Principle 3 – “While it is common for individual lenders to apply a cap on LTV ratios, it is not necessary for regulators and supervisors to mandate such a cap if they satisfy themselves that the underwriting standards are sufficiently prudent and are unlikely to be eroded under competitive pressure. However, jurisdictions may consider imposing or incentivising limits on LTV ratios according to specific national circumstances.”
Central Banks might consider it reasonable to announce changes in LTV cap values same time when announcing changes in countercyclical capital buffer value.
P. 5 – Principle 4 – “Lenders may want to outsource certain tasks to external parties such as appraisers.” and p. 6 – Principle 4.6 – “appraisers meet certain qualification requirements”
Local regulatory authorities (The Central Bank or The Ministry of Finance) might consider it necessary to form a list of appraisers and their qualification requirements to be authorised for mortgage collateral assessment. That is regarded necessary for the countries where the market discipline is not developed enough.
P. 6 – Principle 4.3 – “Property appraisals should not reflect expected future house price appreciation.”
Decrease in value of mortgage underlying asset needs to be stress-tested to evaluate the prepayment risk as the bank would be faced with the necessity to pay out its long-term liabilities it borrowed to finance the mortgage portfolio.
P. 6 – Principle 5 – “Mortgage insurance is used in some jurisdictions as a form of credit support for mortgage loans”
It is welcomed to explain whether the term ‘mortgage insurance’ stands for insurance on behalf on the borrower (e.g. life insurance), on the side of a lender (e.g. relevant CDS contract purchase) or both.
P. 7 – Principle 5.3 – “Through the use of mortgage insurance, credit risks, particularly those for high LTV loans, are transferred from lenders to insurers and often concentrate the risks to a smaller number of institutions.”
This means that the mortgage insurers could become the sources of systemic risk for the financial system. This should be properly accounted for by the regulatory authorities by extending the treatment of systemically important financial institutions (SIFIs) by including mortgage insurers to the SIFIs list.
P. 8 – Principle 6.3 – “…the supervisor could specify that initial LTV ratios above 100 percent are not acceptable under any circumstances or that stated income is not acceptable and lenders should always conduct due verification”.
For example, in Russia LTV ratios vary from 0.3 to 0.9. On the contrary, in Australia or in the USA,low LTV ratios (which are below 0.8) are associated with the lower rates for lowrisky borrowers and allow lenders to deal separately with highly risky borrowers. Therefore, it is recommended to consider combined-loan-to-value ratio(CLTV) that is calculated as follows:
On the other hand, there is the issue of repricing induced by rapidly changing market environment. The minimum repricing period of 6 months is suggested to be introduced. Being more frequent it would increase the uncertainty for individual borrowers and might lead to worsening financial instability in times of crisis.
P. 8 – Principle 7 – “Jurisdictions should provide for appropriate monitoring and supervision of mortgage underwriting practices.”
Though “The FSB Principles for Sound Residential Mortgage Underwriting Practices aim to provide a framework for jurisdictions to set minimum acceptable underwriting standards”, it is important from the supervisory perspective to describe the treatment of mortgage portfolio securitisation. In particular, it seems logical to foresee restricting securitisation operations in a manner that the from abroad acquired mortgage loan portfolio is to satisfy the requirements of the bank home jurisdiction for the deal to be accepted.
P. 8 – Principle 7.1 – “These powers could include:collecting data … or requiring other agencies to collect data for this purpose on their behalf”
When stating the need to collect data for the benefit of regulation and supervision, changes to local regulation on confidential data processing are likely to be mentioned.
Research Team
Andrievskaya I.K.; Lvov N.P.; Malkov E.S.; Penikas H.I.
1
[1]E.g. Gardes F., Duncan G., Gaubert P., Gurgand M., Starzec Ch. Panel and Pseudo-PanelEstimation of Cross-Sectional and Time-Series Elasticities of Food Consumption: The Case ofU.S. and Polish Data // Journal of Business & Economic Statistics. 2005. Vol. 23.; Gardes F., Gaubert P., Langlois S. Pauvrete et convergence des consommations auCanada. CRSA/RCSA, 1999; Gardes F., Louvet Ph. La convergence internationale des structures de consommation.Paris: CREDOC, 1986.
[2]URL: