As countries around the world begin to implement their countercyclical capital buffer (CCB) frameworks, the question arises for the authorities responsible for operating the buffer regime: Besides the domestic buffer, for which other countries should we care about setting appropriate countercyclical capital rates? This is a legit question since, according to the BCBS’s Guidance to national authorities operating the countercyclical capital buffer (p.5), “The home authorities will always be able to require that the banks they supervise maintain higher buffers if they judge the host authorities’ buffer to be insufficient”. Similar provisions are found in Art 139 of the Capital Requirement Directive (CRD IV), except that in the European Single Market context the responsible authorities can set (higher) buffer rates only for those countries which do not implement the CRD IV rules for the countercyclical buffers, i.e. non-EEA or third countries. Moreover, home authorities can set capital buffer rates for exposure located in those third countries, where the host authority does not set and publish any buffer rates at all.
Conditions to be fulfilled
In order to judge which countercyclical buffer rate for the third-country exposure would be sufficient to protect domestic banks, the authority in charge will have to monitor cyclical imbalances in the third country in question, possibly deploying a set of indicators including the credit-to-GDP gap. Obviously, this cannot be done for every third country to which domestic banks might be exposed. The authorities in charge should rather focus on third countries that satisfy the two conditions proposed below. For all other third countries, they should apply some standard procedure for recognition of the buffer rates.
The two conditions proposed here for identification of relevant third countries are:
- Excessive credit growth in the host country can jeopardise stability of the domestic banking system.
- This issue can be effectively addressed by setting a (higher) capital buffer rate for the exposure located in that country.
The first condition means that the cross-border exposure of the domestic banks represents a spill-over channel for the risks to financial stability. The second condition means that by setting a (higher) buffer rate for the third-country exposure, the home authority in charge will achieve some minimum incremental change in the domestic banks’ institution-specific buffer rates.
Each of these conditions is necessary and the two taken together are sufficient. The fact that the first condition is necessary results from the postulated objective of the countercyclical rates, namely to protect domestic banks from the risks associated with excessive credit growth. The second condition is necessary in as much as otherwise the instrument would have no measurable effect. In the latter case, any identified risks should be addressed by other, more appropriate instrument(s). Both conditions have to be fulfilled simultaneously. Otherwise the cross-border exposures either pose no threat to the stability of the domestic banking system or cannot be addressed by the capital buffer rates.
Criteria to verify the conditions
In order to verify whether the two conditions proposed above are fulfilled, responsible authorities need to operationalise them. For this purpose, following two criteria can be utilised accordingly:
- ‘Spill-over channel’: The total credit-risk exposure to the third country is large enough to serve as a channel for propagation of the risk to financial stability to the home country.
- ‘Addressability’: The share of own fund requirements for the relevant credit exposure to the third country is large enough to affect a fixed minimum incremental rise of the institution-specific capital buffer rates.
Variables and thresholds
In general, several concepts might be applied for measuring criterion 1 based on theoretical considerations, but only one approach really makes sense for criterion 2 based on pragmatic reasoning. Let us consider these concepts in turn.
For measuring criterion 1 (the spill-over channel), I propose the following concepts: Relevance to the national economy, asset concentration, and risk of contagion in domestic banking sector. The home authorities should select particular indicators and corresponding thresholds for each of these three concepts according to the specificities of their jurisdiction, possibly using econometric analysis. An example can go as follows:
- Relevance for the national economy: The amount of credit-risk exposure of all domestic banks to a third country exceeds 1% of the GDP.
- Asset concentration: The amount of credit-risk exposure of all domestic banks to a third country exceeds 0.5% of the aggregated total assets.
- Risk of contagion in domestic banking sector: The amount of credit-risk exposure of the systemically important banks to a third country exceeds 0.5% of their aggregated total assets.
The measuring concept based on (domestically and globally) systemically important banks accounts for the fact that, since these banks are per definition highly (nationally and internationally) interconnected, their exposure to third countries may represent a spill-over channel for risks to the financial stability.
For the purposes of criterion 1, the credit-risk exposure should be defined in a broad fashion as the original, non-risk-weighted exposure subject to credit risk, including on- and off-balance-sheet items. To the credit developments in third countries that breach at least one of the specified thresholds, special attention should be paid. In order to decide whether the potentially unsustainable credit growth abroad can be addressed by imposing (higher) countercyclical capital rates on the respective exposure of domestic banks, criterion 2 has to be verified.
For measuring criterion 2 (addressability), a pragmatic concept described below can be applied. Note that the institution-specific capital buffer rate is a weighted average of the domestic buffer rate and the rates applicable to the institution’s exposures located in other countries. The weights correspond to the relative volumes of the bank’s own fund requirements for the relevant credit exposure in each given country. Consider, for instance, a bank that holds relevant credit exposures in the home country A and in two foreign countries B and C. Assume that the bank’s institution-specific countercyclical capital buffer rate is as summarised in Table 1.
Table 1. An example of the calculation of the institution-specific CCB rate
The table illustrates that if a bank holds a larger part of its relevant credit exposure domestically, the countercyclical capital rates applicable to its foreign exposures have to be relatively high to have a sizeable effect on the resulting capital buffer requirement (see country B). For country C, even a capital buffer rate equal of 2.5% (the boundary for the mandatory reciprocity) results in an almost negligible contribution to the bank’s capital buffer requirement. If the weight of country C (here 2%) is small at the aggregated level of the domestic banking sector too, tightening the countercyclical buffer regime for this country would not increase the banks’ resilience by much. More targeted instruments should be applied to combat risk (if any identified) that may spill-over from country C to the domestic banking sector.
Following this logic, I suggest that the authorities in charge should choose some minimum incremental contribution to the institution-specific countercyclical capital rate that they want at least to achieve. They can translate this minimum contribution into the lower threshold for the share in banks’ own fund requirements. Only those third countries where the exposure share breaches the threshold will be the considered as ‘addressable’ by the capital buffer. As Table 2 illustrates, given a capital buffer rate of at most 2.5%, a minimum contribution to the institution-specific capital buffer rate of, say, 0.1% (second column) is only achieved if the country’s share in the bank’s own fund requirements hits a threshold of 4%. If the countercyclical capital rate for the third-country exposure can be raised to 5%, the minimum contribution of 0.1% is achieved if the country’s share hits the threshold of 2%. Note that halving the minimum desirable contribution to 0.05% corresponds to halving the lower threshold for a country’s share in a bank’s own fund requirements (third column).
Table 2. Lower threshold for the share in a bank’s own fund requirements depending on the fixed minimum contribution to the institution-specific CCB rate
In line of these considerations, the home authority may choose adhering to the following concepts and thresholds:
- Addressability at the system’s level: The share of the own fund requirements for the relevant credit exposure of the domestic banking system in a given third country exceeds 2%.
- Addressability at the banks’ level: The share of the own fund requirements for the relevant credit exposure of (a number of) the systemically important banks exceeds 2%.
The relatively low threshold of 2% chosen above insures that no potentially important third country is left unaccounted. The proposed measuring concepts for criterion 2 may be extended as to consider the relevant credit exposure not only to single third countries but also to groups of economically tightly related countries.
Of course, the measuring concepts and thresholds presented above serve in the first line illustrative purposes. Defining them in practice, the authorities in charge for operating the countercyclical buffer regime should duly regard specific characteristics of their domestic financial sector.
CCB-relevant countries and countries on the watch list
The third countries that pass both criterion 1 and 2 should be put on the list of ‘CCB-relevant’ countries. The risk to financial stability in those countries could spill-over to the domestic financial system through the credit exposure channel and can be addressed by means of the countercyclical capital buffer. Only for these countries a further analysis as to the appropriate capital buffer rate needs to be carried out. Other countries that pass criterion 1 but not criterion 2 should be put on a ‘watch list’ since they might become CCB-relevant in the future. For these countries, the risk possibly identified should be addressed with other macroprudential instruments deemed to be more appropriate than the countercyclical capital in the context of the respective risks. As to the countries that pass criterion 2 but not criterion 1, it is likely that risks, if any, stemming from the exposure located in these countries is better addressed with microprudential measures.
Finally, the list of the CCB-relevant countries should be regularly updated, e.g. once every year, in order to account for possible shifts in the domestic banks’ exposure allocation.
The EU context
The data for narrowing down the list of countries to be considered as CCB-relevant will become available in the EU as soon as the banks begin disclosing that information in accordance with the European Banking Authority’s Regulatory Technical Standards on identification and disclosure of the geographical location of the relevant credit exposures (EBA/RTS/2013/15 and EBA/RTS/2014/17). In the context of the Single Market, the analysis of vulnerabilities emanated from the CCB-relevant third countries and of the appropriate capital buffer rates should be done at the European level. This would promote a harmonised application of the third-countries CCB across the Union. For this purpose, Art 139 CRD IV envisages a possible role of the European Systemic Risk Board. The ESRB can, by the way of recommendations, help to achieve coherent application of the third-countries buffer settings across the EU.