Primary tabs

French Banking Federation

The EBA, in its consultation paper discusses two options for identifying the geographical location of credit exposure.
The first option, called (A1), is based on the principle of immediate risk, which is the place of residence of the obligor (immediate risk principle).
The second option, called (A2), is based on the principle of ultimate risk: therefore it retains the country of residence of the obligor in the case of the absence of any guarantees and collaterals, (the immediate and ultimate risks are identical) and retains the country of collateral or guarantor for guaranteed/collateralized facilities (i.e. the ultimate risk principle).

From a conceptual standpoint, the EBA favors the principle based on the country of the obligor/debtor. This solution allows the set-up of a countercyclical capital buffer regardless of where the collateral is located, because the residence of the obligor is, in most cases, closely linked to the economy of the country .

We consider that the EBA’s choice between the two options should be based on the analysis of which principle is consistent with the countercyclical capital buffer (CCB) objective.

According to the EBA’s RTS, the purpose of the CCB is:
1. “To protect the banking system against potential losses when excessive credit growth is associated with an increase in system-wide risk.”
2. “When risks appear, the additional capital will help the system to absorb losses while continuing to provide credit to the real economy.” However, this analysis is criticized by Repullo and Saurina (2011) according to whom”(…) banks may prefer to reduce credit extension rather than being subject to restrictions on capital distribution (dividend, shares, repurchases and especially discretionary bonus payment to staff) if they do not meet the additional capital requirement ”
3. “As a possible positive side effect, the CCB may help to counter the expansionary phase of the credit cycle by reducing the supply of credit and/or increasing its cost”
4. “The CCB should counter the pro-cyclical amplification of shocks via the banking system to the real economy”.
Therefore, the micro-prudential goal is put forward on points 1 and 2 whilst under points 3 and 4 the EBA recognizes an induced macro-economic effect of the CCB tool. The BIS also recognized the induced macro-economic effect of the CCB tool .

1. The ultimate risk is only relevant for immediate risk cases
The main argument in favor of the principle of immediate risk is the asymmetry between the immediate and ultimate risk. The choice of the ultimate risk may be justified only if the CCB instrument is intend for micro-prudential purpose. In this case, the objective is to provide a sufficient layer of capital to absorb potential losses. The ultimate risk, either on the collateral or on the guarantor measures rate of loss the bank is exposed in case of default of the counterparty.
However, concerns about the quality of the collateral or guarantor should intervene only in the event of default of the debtor/ obligor. Take the example of a loan granted by a credit institution to a debtor residing in a foreign country A and guaranteed by a guarantor residing in a foreign country B. The application of the ultimate risk principle would have as a consequence for the credit institution the requirement for additional countercyclical capital buffer (when the credit-to-GDB ratio grows excessively in country B) even though the risk and the likelihood the debtor defaults remain in country A, in the absence of an excessive credit growth of the credit-to-GDP ratio.

2. On the consistency of the obligor principle with that of credit-to-GDP ratio
The outstanding amount used to constitute the numerator of the credit -to- GDP ratio is based on the obligor principle (or immediate risk principle). Therefore, it seems adequate and logical, for consistency raisons, to maintain that principle in determining the geographical location of the credit institution exposure.

3. On the exemption for specialized financing
For specialized lending exposures, EBA proposes to waive the obligor principle, by using instead for the geographical location the country where the income is generated. The reasons for this are set out in the box of the Article 2 of the RTS ( For example, the systemic component of the credit risk related to specialized lending as defined in Article 147 (8) CRR, such as project finance, seems to be largely linked to the source of income, which is generated in the location of the project/ asset”, page 14). In our opinion, this analysis is arguable.
Thus, the credit-to-GDP ratio, may grow excessively in a particular country, without threaten the sustainability of revenue flow of the funded project. The relationship between the revenue flow of the project financed and the debtor's solvency should be considered. In addition, the economic environment may affect the project revenue flow regardless of any growth of credit-to-GDP ratio.
Moreover, if the ratio departs from its long-term trend exclusively due to a contraction of GDP (effect denominator), the losses generated due to the economic downturn will be hardly conducive to the creation of an additional layer of capital. Finally, the low proportion of specialized financing exposures in total loans to non- financial corporations (between 0.5 % and 1.1%), is a very arguable argument for justifying a derogation from the principle of immediate risk (not without mentioning the additional cost that would be incurred by credit institutions for the monitoring of geographical location of exposure by country of project income). Also, considering the low proportion of the specialized financing into the total credits in a given country, the credit- to- GDP ratio appears to be far too broad to be a relevant indicator of material risks for the project financing sector.



4. Verify the absence of redundancy in the calculation of exposures (for a consolidated group)
According to the article 129 of the CRD, "Member states require institutions to maintain (...) a capital conservation buffer of Common Equity Tier 1 capital equal to 2.5% of their total risk exposure amount calculated on an individual and consolidated basis ". It is therefore necessary that EBA confirms that only the direct exposure of an entity are subject a countercyclical capital buffer requirement in cases where its subsidiaries are already subject to a countercyclical capital buffer requirement for their own exposures. The Basel framework regarding the minimum capital requirements addresses this issue and it seems reasonable to consider that the same principle will apply for the countercyclical capital buffer requirement. We ask therefore the EBA to clarify this point.

The wording used by the BCBS is the following: "As with the minimum capital requirement and capital conservation buffer, host authorities would have the right to demand the countercyclical capital buffer be held at the individual legal entity level or consolidated level within their jurisdiction." This statement suggests that this is an option at the discretion of the national supervisor. We believe this is not specified in the consultation document of the EBA and we ask for clarification on this matter.

5. Why retain the immediate risk from a statistical point of view?

The statistical annex, based on the BIS’ foreign claims database aims at assessing the impact of the retained principle (the immediate risk or the ultimate risk) on the outstanding exposures amount and hence on the amount of countercyclical capital buffer. However, as the EBA mentions, the comparison of data based on immediate risk principle and ultimate risk principle is a challenging exercise due to differences in terms of scope (perimeters) and consolidation. Regarding the data based on the immediate risk principle, they are determined on a consolidated basis and therefore include foreign subsidiaries perimeters. However, the data based on ultimate risk principle take into account only resident entities. However, the paucity of data can lead to privilege the existing BIS statistics and ignore the caveats associated with these data.
According to calculations made by the EBA, the median of the relative deviation of the differences on foreign claims held by banks in countries reporting to the BIS towards the 28 member states of the EU on immediate risk and ultimate risk is low and remains below 5 % for the period Q1 2005 - Q4 2012. In other words, based on the available data, the choice of one or other of the two approaches will generally lead to similar results (i.e. modest impact in quantitative terms). For the purposes of simplification, it seems preferable, as suggested by the EBA, to retain the option A1 (i.e. the country where the obligor resides), or the immediate risk principle.
The EBA underscores that an exception to this general immediate risk principle is that where the guarantor plays an important role that the guarantor or collateral’s economic environment would represent a better reference for the purposes of the counter-cyclical buffer. The exception concerns the specialized financing (i.e. project finance), where the country of the source of income represents the country holding the ultimate risk. The exception to the immediate risk principle general seems problematic because the collection and reporting necessary to assess the ultimate risk for specialized lending could induce a significant additional cost significant ultimately borne by the credit institutions. Moreover, from a conceptual point of view, the guarantor's economic environment is not limited to the credit-to- GDP ratio."
Yes. We agree on using the obligor principle for the practical implementation of the countercyclical buffer. Nevertheless, we consider that only this principle should be applied even if collateral or guarantees exist. This principle is consistent with the objectives of the counter-cyclical buffer because it will help to build capital in the country of residence of the obligor, as the residence of the obligor will, in most but not all cases, be closely linked to the relevant economy.
For specialized lending exposures, EBA proposes to waive the obligor principle, by using instead for the geographical location the country where the income is generated. The reasons for this are set out in the box of the Article 2 of the RTS ( For example, the systemic component of the credit risk related to specialized lending as defined in Article 147 (8) CRR, such as project finance, seems to be largely linked to the source of income, which is generated in the location of the project/ asset”, page 14). In our opinion, this analysis is arguable.

Thus, the credit-to-GDP ratio, may grow excessively in a particular country, without threaten the sustainability of revenue flow of the funded project. The relationship between the revenue flow of the project financed and the debtor's solvency should be considered. In addition, the economic environment may affect the project revenue flow regardless of any growth of credit-to-GDP ratio.

Moreover, if the ratio departs from its long-term trend exclusively due to a contraction of GDP (effect denominator), the losses generated due to the economic downturn will be hardly conducive to the creation of an additional layer of capital. Finally, the low proportion of specialized financing exposures in total loans to non- financial corporations (between 0.5 % and 1.1%), is a very arguable argument for justifying a derogation from the principle of immediate risk (not without mentioning the additional cost that would be incurred by credit institutions for the monitoring of geographical location of exposure by country of project income). Also, considering the low proportion of the specialized financing into the total credits in a given country, the credit- to- GDP ratio appears to be far too broad to be a relevant indicator of material risks for the project financing sector."
No, the residence of the debtor should be the principle for the identification of the geographical localization of all credit exposures.
Yes. We agree on the inclusion of a proportionality threshold for credit risk exposures, to alleviate the operational burden for institutions with very limited aggregate cross-border activity.
For the calculation modalities, we propose to compute the threshold no more frequent than once a year based on the previous quarter
We request EBA to introduce a 0.5% RWA materiality threshold per country. As a safety net, the exemption (total of RWA for small perimeters) should not exceed 10% of total RWA.
We consider the approach of Article 3 based on a pro-rata of the standalone specific risk own funds requirements as an additional burden for internal model banks which bring no benefit. We therefore are not supportive of this approach.

Instead we support the fourth method that assigns the exposures using the same geographical distribution as the credit risk exposures. This is definitely the easiest to implement and, arguably, there is indeed a degree of correlation between the location of credit and market risks exposures.

If however the EBA considers the credit risk" approach as too crude, in particular for large international institutions with a fair share of trading book exposures, we suggest that internal model institutions weight their global trading book IRC with the standalone IRC calculated on a country by country basis."
Yes, we agree with the inclusion of a threshold for trading book exposures and would like to suggest a higher threshold than the one proposed in the consultation paper, as 2% is too proportionately too low.
Yes for securitization exposures, consistency with the “COREP SEC Detail” reporting template should be achieved. When the look-through approach is not possible, a fall-back rule should be applied.

No in relation to funds/CIU: the Look through approach is burdensome and will not bring much added value to compute the counter-cyclical buffer.

The “engagements internationaux” template, an “Individual to Aggregate” type of reporting currently produced by French banks and covered by the Financial Stability Board’s “Datagaps” initiative specifies that when reporting banks are unable to apply the Look-through approach, exposure on securitization can be declared on the country of residence of the relevant Special Purpose Vehicle.
Accordingly, reporting institutions might only be able to provide estimates for the allocation of claims to the country where the debtor of the underlying resides or to allocate the claims to the country of the immediate risk, which is the country where the issuer of the securities resides.
We would like to suggest this as a fallback rule complementing the “COREP SEC Detail” reporting template approach.
Yes, we agree.
The statistical annex, based on the BIS’ foreign claims database aims at assessing the impact of the retained principle (the immediate risk or the ultimate risk) on the outstanding exposures amount and hence on the amount of countercyclical capital buffer. However, as the EBA mentions, the comparison of data based on immediate risk principle and ultimate risk principle is a challenging exercise due to differences in terms of scope (perimeters) and consolidation. Regarding the data based on the immediate risk principle, they are determined on a consolidated basis and therefore include foreign subsidiaries perimeters. However, the data based on ultimate risk principle take into account only resident entities. However, the paucity of data can lead to privilege the existing BIS statistics and ignore the caveats associated with these data.
According to calculations made by the EBA, the median of the relative deviation of the differences on foreign claims held by banks in countries reporting to the BIS towards the 28 member states of the EU on immediate risk and ultimate risk is low and remains below 5 % for the period Q1 2005 - Q4 2012. In other words, based on the available data, the choice of one or other of the two approaches will generally lead to similar results (i.e. modest impact in quantitative terms). For the purposes of simplification, it seems preferable, as suggested by the EBA, to retain the option A1 (i.e. the country where the obligor resides), or the immediate risk principle.
The EBA underscores that an exception to this general immediate risk principle is that where the guarantor plays an important role that the guarantor or collateral’s economic environment would represent a better reference for the purposes of the counter-cyclical buffer. The exception concerns the specialized financing (i.e. project finance), where the country of the source of income represents the country holding the ultimate risk. The exception to the immediate risk principle general seems problematic because the collection and reporting necessary to assess the ultimate risk for specialized lending could induce a significant additional cost significant ultimately borne by the credit institutions. Moreover, from a conceptual point of view, the guarantor's economic environment is not limited to the credit-to- GDP ratio.
Sarah Quemon
F