In general we agree with the three main categories of conditions for the setting of higher risk weights (paragraph 1) and the setting of higher minimum LGD values (paragraph 2). The RTS could, however, include the combined result of the assessment of the appropriateness of the risk weights referred in the CRR and the methodology proposed in the consultative paper, financial stability considerations and the other conditions referred in Art. 4 and Art. 6 of the proposed Regulation.
However, we consider the categories of conditions specified in the Regulatory Technical Standards (RTS) potentially offer too wide a framework for National Competent Authorities (NCAs) risk weighting and Loss Given Default (LGD) assessment processes. The field of data to be potentially analysed is large, and their characteristics are not defined precisely. Also, the assessment processes could require an administrative burden and significant cost of adapting available databases or building new databases (including anticipation models), and the impact this will have on capital allocation inside the institutions are of significance. We hope for precise governance frameworks detailing the processes as they are established between the NCAs and the institutions of their jurisdictions. These governance frameworks for assessments would determine precisely the data and ratios to be analysed, and define more precisely the rules according to which the NCA would decide to raise risk weighting and LGD requirements or not. A transparent publication of these process frameworks would be required.
Also, we believe there should be a confirmation that the NCAs mentioned in the RTS are national NCAs, even for institutions supervised at the ECB level. It seems implicit, since national specificities have to be taken into account according to this draft RTS. But from this point of view it must be noticed that institutions having activities in different European countries could end up having to comply with different risk weighting and LGD assessment processes, and with different risk weight and LGD requirements.
Finally, and most importantly, residential loans are mostly secured in France by guarantees (“cautions”). This type of guarantee, that have proved to be sufficient security similar to a mortgage, and recognized in EU regulation, might be permanently named and considered an equivalent to a mortgage in European financial regulation in general, and in this RTS on NCAs risk weighting and LGD assessment processes in particular.
We agree with the conditions for specification of the loss experience and the loss expectations, but we do not agree completely on what is specified in paragraph 3 of Article 2, as it seems redundant considering the paragraph 2 conditions.
Paragraph 1 describes how competent authorities shall determine the loss experience relating to one or more property segments of exposures secured by immovable property based on the data indicators referred to therein, as a ratio of losses stemming from those exposures values; they should determine the losses which they expect to be realised as a ratio of losses expected for those exposures to those exposures values and consider the indicative benchmarks proposed in the consultative Regulation in paragraph 4.
Paragraph 2 describes as the required adjustments to reflect the forward-looking immovable property market development are to be based on: a) the historical evolution in the immovable property market, b) the expected evolution in immovable market prices and the expected volatility in those prices, including an assessment of the uncertainty around these expectations, c) the time horizon over which property market developments are expected to materialize, d) fundamental drives i.e. loan-to-value ratio and debt service-to-income ratio, e) structural and cyclical characteristics of the immovable property market, f) the impact in terms of increase of total risk-weighted exposure amounts for exposures secured by immovable properties.
Finally, according to paragraph 3, the National Competent Authorities may be more conservative when there is a high degree of uncertainty regarding the expectations in the immovable property market and/or one or several indicators of losses experience of fundamental drivers (i.e. loan-to-value and debt service-to-income ratio) are not available over a sufficient long period.
This disposition risks doubling the effects of the prudential adjustments required on the basis of a forward-looking analysis of the market, for the specific issues described in the above mentioned paragraphs b) and d). In addition, we would like to outline that debt service-to-income ratio is a fundamental indicator for personal lending and household mortgage lending, but not for commercial real estate lending and, as far as we know, in the non-residential mortgage sector there are no publicly available statistics concerning debt service-to-income ratio.
Also, the reports that are mentioned in Article 2.1a) exist since January 2014. They are periodic reports and there has been three collections to date. It could become of major signification for NCAs when determining a possible rise in risk weights and LGD minimal values, it would be important that the regulator gives feedback on the quality and significance of the data collected.
Article 1(a) refers to determining loss experience relating to one or more property segment secured by immovable property. It seems that the segmentation between residential and commercial immovable properties is efficient enough. We consider it would be too complex and hardly efficient to look for loss experience data on too narrow segmentation. It would become difficult to handle for institutions if too fine data were required. Moreover, some concerned institutions are larger banks subsidiaries. They may use the standardised approach whereas the mother bank uses the IRB approach. It seems important that the segmentation used in the assessment processes for risk weight and LGD are the same, or at least remain coherent. Finally, it seems of major importance that the segmentation rules used in the assessment processes are coherent with the current work undertaken by the Basel committee on the revision of the standardised approach.
The data to be collected and analysed for the assessment processes are widely defined in the draft RTS. From a quantitative point of view, it seems insufficient to list the categories of data required for the determination of expected losses in the assessment process. It’s important that institutions are aware of the types of data that will be required, and their level of detail. If required data is too detailed, it will raise questions concerning the ability of the institutions to deliver the data. It might oblige institutions to build models that at the end could be closer to an IRB model. These questions raise the issue of more precise and transparent governance frameworks for the assessment processes, as exposed in Q1. The choice of data collected for the risk weight assessment process should be relative to the risk factors that will finally be determined when current work on the revision of the standardised approach is finalised.
From a qualitative point of view, it seems that data used for the calculation of loss expectation should not only rely on macro –economic considerations and figures on the one hand (article 2.1), and on single mathematical formulas on the other hand (article 2.3). It seems data should be better linked with the immovable property financing market and the term of their portfolios. For instance, account should be taken of historical and dynamic data on Loan-to-Value (plus prices actualisation) and Debt to Income ratios series, over longer periods of time.
Furthermore, the frequency of the risk weights assessment process is important to determine. CRR requires that it is at least once a year. It would mean that once risk weights have been raised, institutions would have to wait at least a year to see them reduced. However, it’s worth mentioning that if it were too frequent that could become difficult to handle for smaller institutions.
We agree with the two thresholds mechanisms for the standardised approach. Looking at the two possible methods used to calculate the initial threshold (page 23-24 of the document), the first one, based on the Basel 2 formula for the calculation of capital requirements and based on the fact that capital would be required to absorb unexpected losses (estimated at 4% for a 50% weighting for non-residential real estate), seems to be the most appropriate. We agree with the 2% benchmark of expected losses for a 50% weighting of non-residential real estate leasing exposures and we would suggest a 6% benchmark as a limit for the 100% weighting of these exposures.
We think that the other method, based on the 0.3% level of losses mentioned in paragraph 3 (a) in Articles 125 and 126 CRR, is too prudential and not appropriate because it was introduced with another purpose, in order to derogate from the requirement according to which the risk of the borrower shall not materially depend upon the performance of the underlying property or project, but on the underlying capacity of the borrower to repay the debt from other sources.
In addition, in the second method, the assumption made at the beginning of the EBA’s calculation appears to be inconsistent with the “CRD IV package” applicable risk weights for RRE and CRE under the Standardised Approach are respectively 35% and 50% instead of 100%. This implies that the 0.105% lower band of the loss interval proposed for Residential Real Estate (RRE) and the 0.15% respective bound for Commercial Real estate (CRE) should more correctly be 0.3% in both cases?
Notwithstanding of that, in our opinion, it is better to propose an higher benchmark than a wide range of benchmarks that could be approached differently by different National Authorities.
As far as the application of the indicative benchmarks for the assessment of the risk weight, we ask that the increase of the risk weighting is proportional to the increase in the losses, with a “gradual” adjustment process towards higher risks weights. As an example, a level of 2.5% expected losses in the market would not automatically lead to the maximum level of weighting for that class (100%), but only to proportional increase as respect to the 50% risk weighting. Therefore, we ask you to introduce, or to allow National Authorities to introduce, a proportional mechanism for the RE risk weighting assessment. Moreover, the demonstration underlying this second argument seems biased since the note #5 in paragraph 1 page 24 mentions that 100% is the applicable risk weight for exposures fully secured by immovable properties, whereas it seems that the right percentage is rather 80%.
Therefore, for the identification of the levels of losses to which the risk weights of 35% and 50% for exposures secured by RRE and CRE are appropriate, we would consider the first argument more accurate. It leads to consider risk weights of 35% and 50% as generally sufficient for an average loss of 1.4% or 2%.
It would be important to confirm that these percentages have to be applied on the global portfolio, and not on a loan to loan basis.
We agree with the principle according to which financial stability considerations exist when setting higher risk weights or higher minimum LGD values and we also agree with the specification of the term of “financial stability considerations” illustrated in Article 3 of the draft regulation.
We agree with the other conditions for setting of higher risk weights; as far as the indicative benchmarks specified in paragraphs 3 and 4 go, we estimate that, for the real estate leasing market, a loss expectation above 2%, but lower than 6% is generally appropriate for increasing the 50% risk exposures fully and completely secured by commercial immovable property up to but below 100% and that a loss expectation equal to or above 6% is generally appropriate for increasing the 50% risk weight of such exposures to a risk weight ranging from 100% to 150%.
The provision by NCAs of explanations and assessment of procyclical effects seem essential. Yet, it implies the ability for institutions to “backtest” the data analyses of the NCAs, and so it’s important that the NCAs processes are transparent and precise.
We agree with the conditions for specification of the exposure weighted average LGD and the LGD expectation and with the adjustments allowed to be made to the average exposure weighted LGD on the basis of the forward-looking immovable property market developments. As mentioned referring to the standardised approach, also for the minimum LGD values the competent authorities may be more conservative when there is uncertainty around the expectation in the immovable market and/or one or several indicators of loss experience of fundamental drivers (i.e. loan-to-value and debt service-to-income ratio) are not available over a sufficient long period. This disposition risks doubling the effects of the prudential adjustments required on the basis of a forward-looking analysis of the market, for the specific issues described in Article 2, paragraph 2 b) and d).
We also agree that it is not appropriate to set indicative benchmarks for the setting of higher minimum LGD values because of the specificities of national immovable property markets and because of the relationship of the LGD parameter with the other internal model parameters.
We agree with the “other conditions” for the setting of higher minimum LGD values, especially regarding Article 1.2 c) requirement to assess the potential pro-cyclical effects of setting higher minimum LGD values in the current stage of the economic cycle on the financial stability considerations referred to increasing the minimum LGD in order to mitigate the financial stability considerations.
We agree with the decision to avoid specifying in this RTS a wide set of data indicators that should be considered in the forward-looking analysis of the immovable property market. An eventual list wouldn’t help in pointing competent authorities to the essential variables and wouldn’t be appropriate for any immovable market or segment of the market, considering that any national market differs from the others, that the residential real estate mortgage market differs from non-residential and that non-residential real estate mortgages’ market also differs from real estate leasing market.
As mentioned in answer to question 3, among the two different arguments proposed for the indicative benchmarks for setting higher risk weights, we consider the second one too conservative as it is not built for the final purpose of this RTS.
Obviously the risk weights must be adequate to the level of risk. However we would like to point out that the proposed solution will increase regulatory capital requirements. Therefore we expect that the financial supervision authorities will govern both ways, for example, they would set higher minimum LGD levels if the forward looking immovable property market development is negative and set lower minimum LGD levels if the forward looking immovable property market is positive. Additionally we hope to see an agreed framework and mechanism with the financial sector for the monitoring of the immovable property sector.