We do not think that non-EEA banks are a part of the shadow banking sector. This would be a contradiction in itself. Banks are by definition not shadow banks. Also, they do not pose the same risks to the financial sector.
The CP correctly points out the points of concern with regard to shadow banking in 3.1. item 2 and in 3.1.1. However, the issues pointed out are generally not empirically observable in the case of foreign non-EEA banks.
The CP points out that the shadow banking system heavily relies on short-term wholesale funding. But this is not generally the case with banks, no matter in which jurisdiction they are localized, for they also have access to retail client deposits and central bank lending.
The shadow banking system is also understood to generally lack transparency. This is another issue that is not generalizable for foreign banks. The amount of leverage, maturity and liquidity transformation are known at least to home supervisory authorities, for this is common sense for all supervisory authorities around the world (no matter which stage of the Basel framework they apply).
Furthermore, the shadow bank system is understood to be vulnerable to runs, because they are dependent on refinancing in the market. This is not true for banks, as they have a liquidity backstop at their central banks, no matter in which country they are localized.
Moreover, the CP assumes that fire sales from the shadow banking sector have spread the stress during the last crisis. This is yet another issue that has not generally been observable with foreign non-EEA banks, at least not in a different manner than EU credit institutions. It would be inappropriate to blame foreign non-EEA banks for balance sheet reduction albeit EU credit institutions were doing the same.
In addition, it must be pointed out that foreign non-EEA banks are leveraged, but generally not higher (and often much lower) than EU credit institutions. This holds true no matter if the regulation applied in the home country is in line with Basel I, II or III. Foreign non-EEA banks are clearly not “highly leveraged structures” with “inadequate loss-absorbing capacities”. That is another very important difference to the shadow banking sector which is not limited in leverage by any regulation or supervision at all.
Another point of concern the CP raises is interconnectivity. It states that the shadow banking sector is highly correlated and interconnected to the regulated banking sector (i.e. in the EU) due to ownership linkages, credit commitments and as counterparties. We cannot find this to be true to the same extent for foreign non-EEA banks. They are clearly not linked to the EU banking sector by ownership and, unlike SPV’s, they do not rely on other bank’s credit commitments for refinancing.
Finally, the governance of non-EEA banks is generally very similar to EU credit institutions. There is no evidence for opaqueness and complexity to the same extent as alleged to the shadow banking sector.
II. Treatment of Union branches of non-EEA banks
One remarkable effect of the shadow bank definition as proposed in the draft Guidelines is that it would cover Union branches of non-EEA banks. The reason for this is that these are not credit institutions in the meaning of CRR and they are not supervised according to rules agreed upon on a pan-European level.
However, it must be noted that Union branches of non-EEA banks are covered by national supervisory law and supervision in the Member States. E.g. in Germany, they are subject to the analogous application of CRR on a stand-alone basis. As a consequence we clearly oppose to the idea that Union branches of non-EEA banks be treated as shadow banks.
In addition, as far as we know, all EU national supervisory systems allowing for establishment of Union branches under national law require due diligence to be carried out by national supervisors as to the quality of supervision of the respective head quarter in its home country. Furthermore, the applicant has to demonstrate stability, viability, reliability and professional expertise of the head quarter, including as regards AML issues.
We would therefore propose to
• exclude Union branches of non-EEA banks from the definition of shadow banks, and
• exclude from the definition of shadow banks every non-EEA bank to whom the right to establish a Union branch in the EU was granted by national EU authorities.
III. The criterion of “equivalence”
The draft Guidelines include foreign non-EEA banks in the shadow banking sector if their home jurisdiction fails to meet the “equivalence” test. This means that if a non-EEA bank is subject to prudential and supervisory requirements that are at least equivalent to those applied in the EU, it is not regarded as a shadow bank, whereas if equivalence has not been determined yet, it would be treated as a shadow bank.
We highly recommend to abandon this approach and replace it. We would like to propose a workable and adequate alternative, which we outline below, for the following reasons:
A) The equivalence test is too complicated.
According to the Note to the Members of the European Banking Committee Accompanying the draft Commission Implementing Decision on the equivalence of the supervisory and regulatory requirements of certain third countries and territories for the purposes of the treatment of exposures according to Regulation (EU) No 575/2013 (EBC/05/2014), the operational standards for the assessment of the equivalence of supervisory and regulatory arrangements applicable to credit institutions comprises no less than 24 criteria distilled from 12 Directives and Regulations which are complicated to assess.
According to this Note, the equivalence test is designed to determine whether the relevant third country is able to achieve the same final objectives as the EU in terms of:
1) ensuring the stability and integrity of both the domestic and global financial system in its entirety;
2) ensuring effective and adequate protection of depositors and other consumers of financial services;
3) promoting cooperation between different actors of the financial system, including regulators and supervisors;
4) ensuring independent and effective supervision;
5) ensuring proper implementation and enforcement of relevant internationally agreed standards.
B) The equivalence test cannot be performed by EU credit institutions themselves.
This assessment framework tries to establish the quality of a financial system. It is clearly too much to handle the evaluation of a given bank on a stand-alone level. A full equivalence test of this kind cannot be performed by an EU credit institution for a given debtor/exposure. Either EBA or the Commission would have to commit to perform it. Given that there are 193 jurisdictions to assess, the task is largely overdesigned. Also, it is highly politically sensitive.
C) The equivalence test is interconnected.
The equivalence test is interconnected with a number of other issues regulated by CRR, so it cannot be done independently for the purposes of defining a shadow bank. Whereas it would be justified to apply stricter approaches in one area (e.g. the definition of an “institution” in the meaning of Art. 107 (3) CRR or Art. 391 CRR), this would spill over to overly burdensome requirements for exposures to banks which are sound and solid but then deemed to be shadow banks.
D) Finally (and most importantly), the equivalence test leads to wrong results.
It rules out too many foreign non-EEA banks which cannot be said to be shadow banks. Even a Basel I supervisory regime effectively rules out the possibility that the bank is an unregulated entity running up massive leverage. So we think that a “Shadow bank test” should be designed to identify shadow banks in third countries with no overshooting.
E) We would like to propose the following alternative:
According to the CP, shadow banks can be said to pose specific risks. If we match these risks with measures in place to prevent them in case of a bank, we find criteria which can easily be assessed by credit institutions themselves. A full equivalence test for national supervisory and regulatory systems does not need to be conducted.
These risks are liquidity problems (which are closely connected to run risks), excessive leverage, interconnectivity and complexity.
We think that these risks are addressed and prevented in the banking sector by basically two characteristics that are non-existent for shadow banks and which are therefore key criteria for defining the difference between both of them. These characteristics are
• the bank has access to a central bank as a lender of last resort,
• the bank is subject to prudential banking supervision by public authorities.
A “shadow bank test” should therefore consist of two criteria which effectively rule out the possibility of accidentally including too many non-EEA banks:
1. Does the entity concerned have access to central bank lending?
2. Is the entity concerned subject to prudential supervision?
If these two criteria are met, the possibility of a shadow bank can efficiently be ruled out. Such a shadow bank test would be meaningful, specifically designed to meet the needs of Art. 395 (2) CRR and easy to perform on the basis of readily obtainable information. It could be conducted by EU credit institutions itself and would not require an EU-wide assessment exercise which has the disadvantages shown above.
In addition, credit institutions might assess whether a non-EEA bank is a member of a deposit guarantee scheme and/or whether it is a public sector entity in order to define appropriate exposure limits.
IV. Treatment of EU entities performing services listed in Annex I CRD
According to the Definitions section of the draft Guidelines, every entity which carries out one or more credit intermediation activities as listed in Annex 1 points 1 to 3, 6 to 8, 10 and 11 and which is not an excluded undertaking is to come under the definition of a shadow banking entity. This definition is too broad even for EU-based entities. Any definition must take into account the national regulation and supervision in place in the respective EU Member States. In the light of the situation in Germany, which is as follows, simplistic approaches would lead to severe problems:
First, undertakings that carry out
• deposit taking in the meaning of Annex 1 point 1 CRD IV (but do not carry out lending under point 2.),
• lending in the meaning of Annex 1 point 2 CRD IV (but do take deposits under point 1.), and
• services under Annex 1 point 6 CRD IV (guarantees and commitments)
are covered by § 1 para. 1 sentence 2 no. 1, 2 and 8 of the German Banking Act, respectively. They come under the German definition of a credit institution and are treated and supervised completely in line with CRD IV and CRR. It seems inaccurate to include them under the term of shadow banks.
Second, undertakings which carry out financial leasing in the meaning of Annex 1 point 3 CRD IV are covered by § 1 para. 1a sentence 2 no. 10 of the German Banking Act. They are supervised according to the German Banking Act. In the light of the supervision these undertakings are subject to, it would be inappropriate to include them in the definition of shadow banks, too.
To address this issue, we would like to propose the following:
If the basic approach to defining shadow banking undertakings was retained, we propose to at least introduce a counterbalance by a corrective criterion. Undertakings in an EEA Member State should not be deemed shadow banks if they are subject to a national regulatory regime and supervision which is in line with the basic principles of CRR.
We broadly agree with EBA’s approach. However, we think that a reference to the principle of proportionality should be introduced because the shadow bank entities are very different in nature. Some exposures warrant very high levels of due diligence, whereas other exposures could easily be demonstrated to be less risky and less complex. The intensity and frequency of monitoring carried out should vary accordingly.
But again, we think that the definition of shadow banking entities is not yet accurate and practicable (cf. our answer to Q1). If unchanged, this would lead to too many exposures being subject to the control mechanisms under Title II point 2. of the draft Guidelines.
Loans to shadow banking entities carry different degrees of shadow bank-specific risks. There is a difference between monitoring the management of exposures to, e.g., an unregulated SPV conduit on the one hand and a big retail AIF under EU and national legislation and supervision on the other hand.
So we would like to propose not to force credit institutions to use a “one size fits all”-approach, which could lead to underestimating risk for the risky and over-allocation of risk management resources to the less risky exposures, but to introduce a reference to proportionality as regards the size and nature of the exposures concerned, relative to the credit institution’s overall exposures and business model.
Furthermore, as long as the definition of shadow banking entities is too broad (cf. our answer to Q1), the oversight arrangements will cover too many exposures, distracting the risk management’s resources from the risky ones.
We agree with the approaches to setting aggregate and individual limits on exposures to shadow banking entities. Especially, we welcome the principle of proportionality being reflected in this approach.
However, at the current stage the definition of shadow banking entities is too broad (cf. our answer to Q1). This results in unnecessary additional burden in establishing aggregate and individual limits for exposures which could more suitably and proportionately be treated within the regular risk management process.
Prior to discussing which of both options would be more conservative, it has to be made sure that both are in line with CRR. We think that this is not the case.
Option 1 does not reflect the basic presumption of the large exposure regime that limits can be set both on the aggregate level of all exposures concerned and/or on the individual level of each exposure. There is no way of interpreting the CRR as to allow for one exposure due to its characteristics “infect” another exposure so that the credit institution would be punished for having one “bad” exposure in the portfolio which then leads to a penalty for some or all other exposures.
This is also clearly not in line with the principle of proportionality, for a very small exposure could lead to a massive disadvantage with regard to large exposures which do not pose any problems.
Therefore, Option 1 has to be rejected and Option 2 chosen.
Unfortunately we are in doubt whether the 25% limit is consistent with the current large exposure framework. Under Art. 395 (1) CRR credit institutions are allowed to allocate exposures up to 25% of eligible capital to one client or a group of connected clients, irrespective of the client’s nature or business. This is an unalterable decision by the legislator. There is no way of setting a quantitative limit to exposures to shadow banking entities below that 25%-limit without infringing the CRR. As the limit of 25% for shadow banking exposures is planned to cover all relevant exposures altogether, this means that on a single exposure level the limit would be decreased below the threshold set forth in Art. 395 (1) CRR.
If the definition of shadow banking entities is not changed, the 25% limit does not take into account the current situation of foreign banks in the EU and their exposures to other group institutions located in non-EEA countries. Foreign banks in the EU (and especially in Germany) have an important role in e.g. trade financing and financing cross-border investment activities as well as in cross-border payments. Their activities are important for credit to the EU real economy. Any further limitation to their exposures to their respective parent and/or group entities will have a detrimental impact.
Furthermore, if the 25% limit was chosen, exposures to non-EEA banks would be subject to stricter limits than exposures to any other “real economy”-undertaking, albeit the non-EEA banks are regulated and supervised entities. This is inappropriate and not acceptable.
As a consequence, we propose to set a fallback limit for overall exposure to shadow banking entities of between 50 and 100 % of eligible capital, so as to respect Art. 395 (1) CRR.
