Enable Javascript in your browser for an improved experience of regjeringen.no

Home/host issues for significant bank branches

Dear Mr Dombrovskis;

I refer to the pleasant meeting we had Tuesday 8th November, discussing developments in the area of financial market legislation. Among the issues we discussed were challenges which may arise for the host authorities if a large bank in one jurisdiction is transformed into a branch of a bank established in another jurisdiction. The new host authorities, who were originally the home authorities for the bank (subsidiary), would, following such “branchification”, become the host authorities for the branch. In order to safeguard financial stability and a national level playing field, the legal framework should, in my opinion, allow for sufficient continued application of host state regulations, either through a possiblity to deny systemically important banks to be transformed to branches of foreign banks, or by securing host state treatment of activities in the host state. In the following I will expand on some of the home/host challenges we have looked into.

It follows from the general EU and EEA framework that a financial service provider in one (home) state can provide financial services via a subsidiary in another member state/EEA-state, or via a branch in a host state. Branches are primarily subject to home state supervision. For significant branches (more than 2 pct. market share), there shall be a supervisory college (CRD IV Article 51 (3)) and a procedure for consultation (CRD IV Article 51 (2)). The regulations applicable to branches will primarily be the regulations of the home state. CRD IV Aricle 140 (2) (b) provides for mandatory recognition of a host state counter cyclical buffer rates of up to 2,5 pct. CRR Article 124 (5) provides that institutions shall apply the risk-weights and criteria to exposures secured by mortgages on commercial and residential immovable property that have been determined by the host state competent authorities. CRR Article 164 (7) provides that institutions shall apply the minimum LGD values for exposures secured by property as determined by the host state competent authorities if these are higher than the home state minimum LGD values. Apart from this, application of host state rules to branches operating in that host state is primarily voluntary for the home state authorities.

The ESRB has adopted two recommendations on voluntary reciprocity. In 2014 the ESRB recommended that counter cyclical buffer rates above 2,5 pct. are reciprocated.  In 2015 the ESRB adopted a more general recommendation on “(…) voluntary reciprocity for macroprudential policy measures”.  According to the recommendation, “macro prudential policy measure” means any measure that is adopted or activated by a relevant authority and addresses the prevention and mitigation of systemic risk. The ESRB regulation defines “systemic risk” as a risk of disruption in the financial system with the potential to have serious negative consequences for the internal market and the real economy. It follows from the preamble paragraph 12 that the recommendation also covers macroprudential measures that are not provided for in CRD IV/CRR. The ESRB recommends that the national authorities assess the cross-border effects (leakages and regulatory arbitrage) of measures, and that they request reciprocation from other states if this is deemed necessary to ensure the effective functioning of the relevant measures.

I have noted that the Commission, in the consultation on “Review of the EU Macro-prudential Policy Framework”, i.a. states that the effectiveness of national macro-prudential policies, and mitigation of distortion of competition by institutions that are not subject to the same measures, could be fostered by broadening the scope of the reciprocity framework provided in the CRD IV/CRR. I agree that reciprocity measures should be strengthened to make the application of the tools effective and to level the playing field for domestic banks and branches of foreign banks. The host authority is in the best position to assess the risks stemming from their markets and its measures should therefore also be applied to the relevant, local exposures of foreign financial institutions.

I would like to underline that this is even more important when branches of foreign banks are so large that they would have been systemically important if they had been incorporated in the jurisdiction (had been a subsidiary), and/or where the combined activities of branches of foreign banks constitute an important part of the national banking market. The importance of being able to impose host state regulations will increase with the size, measured through market share, of the branch or branches. The existing framework for division of competences between home and host authorities seems to be better suited to structures with small foreign branches, and not well suited for large branches in other jurisdictions.

Further, I would like to point out that the issue may be more important for those states that do not participate in the SSM. Participation in the SSM is, as you are aware, not an option for EEA EFTA States.

To use Norway as an example, foreign banks’ share of the Norwegian banking market is appr. 1/4. Most of these foreign banks operate in Norway through branches. The largest foreign bank operates through a subsidiary. This subsidiary presently has appr. 9 pct. of the Norwegian banking market, and it is designated as a systemically important bank in Norway. If this subsidiary is transformed to a branch, supervision will be transferred to the home state authorities, and most of the Norwegian banking regulation will no longer apply to the activities. We will then risk that the second largest bank in Norway will be a branch, and that only ¾ of the domestic banking market will be regulated and supervised nationally.

Our experience both from the banking crisis in 1987 to 1993 and from the financial crisis in 2008 to 2009 show that foreign banks act more procyclically than domestic banks, with more expansion of lending before the crisis, and more reduction of lending in the crisis. Further, if the home state regulation is weaker than the regulation in the host state, risk may be built up in the branch, and consequently in the host state real economy. A large bank with an international structure of large branches may also become both too big to fail, and too complicated to rescue.

One alternative for addressing the home/host issues could be to allow the competent authorities to deny systemically important banks incorporated in their jurisdiction to be transformed to branches of foreign banks.

An alternative could be to allow the competent authorities, which would become the host authorities, to set conditions, including conditions for application of host state regulation and host state supervision, in cases of “branchification”.

A third alternative could be to expand the scope of mandatory reciprocity to cover more host state provisions. This could include both measures that normally are deemed to be macroprudential, as the EBRD has argued, and host state measures which are not normally seen as macroprudential. Possible examples could be pillar I, II and III rules according to CRD IV and CRR and other general measures, hereunder risk-weight calculation rules for IRB banks for residential mortgage exposure, for mortgages on immovable property and for corporates, asset-class specific risk-weight floors, sectoral counter cyclical buffer requirements, liquidity requirements, LTI - and LTV-limits, and mandatory amortization.

The main objective should be to limit the risk of leakages and regulatory arbitrage, which potentially may jeopardise financial stability in the host country. A differentiated approach could be foreseen, with more powers for the host state, and more mandatory application of host state measures, for the larger branches, which pose larger risks to the financial system in the host state.

Please also allow me to mention that IMF staff, who this week visited Norwegian authorities in order to prepare the next Article IV consultation, included the following  in their concluding statement 17th November 2016:

“Regional cooperation on financial stability issues should be strengthened given the need to calibrate macroprudential measures to economic conditions of host countries. This is particularly important in view of the intended conversion of Nordea Bank Norge from a subsidiary into a branch. Enhanced supervisory cooperation among the Nordic finance ministries, central banks, and financial supervisors, which safeguards host country treatment and information exchange, would better ensure financial stability and a more level playing field for banks.”

Thank you again for the opportunity to discuss these issues.


Yours sincerely,
Siv Jensen