Cross-border supervisory cooperation: opportunities and challenges

10th July 2025

This article by Thorsten Beck discusses the factors that drive to a higher or lower intensity of supervisory cooperation, or even to supranational arrangements, in dealing with cross-border banks, and the impact this has on banks' commercial strategies.

Key takeaways 

• National supervision and the global footprint of banks and capital markets create a tension that is not easy to overcome.
• One size does not fit all: for some country pairs or regions supranational supervision might be best, for others, lighter forms of cross-border cooperation.
• Cross-border supervisory cooperation can have positive effects on bank stability, though not necessarily for the largest global banks.
• Closer supervisory cooperation can also create arbitrage reactions by banks, shifting lending and risks into third countries. While this might have negative implications for financial stability, it can also be beneficial for recipient countries.

Regulation and supervision of banks is national, while the footprint of the world’s largest banks is global and capital markets are closely integrated. This contrast implies informational and incentive frictions. Regulators and supervisors primarily collect information about financial service activities in their respective jurisdiction, while their mandate is to safeguard national financial stability.  Decisions taken by national regulators and supervisors, however, have implications for stakeholders outside their jurisdiction. The decision to intervene in a failing bank has implication for foreign depositors, owners and borrowers. The decision to intervene might affect banks in other countries through market linkages, including fire sale externalities, informational contagion, or panics. National supervisory decisions thus have cross-border implications that are not taken into account by national supervisors.

Cooperation between regulators and supervisors has been happening for the past 50 years. Regulators have agreed on minimum standards for capital (and recently liquidity) requirements across major jurisdictions under the Basel I, II and III agreements. Supervisors have signed Memorandums of Understanding  (MoU) to exchange information on material changes in cross-border banks.  Supervisors of parent banks (also known as home supervisors) have convened colleges of supervisors with (host) supervisors of the most important subsidiaries of their own cross-border banks. 

The Global Financial Crisis as wake-up call

The Global Financial Crisis has shown the limitations of this cooperation. Cooperation between supervisors in the form of MoUs and colleges did not extend to the stage of managing failure of these banks and banks had to be resolved on the national level. As the former governor of the Bank of England Mervyn King said, banks are global during life and national in death.

In Beck et al. (2013), we show how the cross-border footprint of banks biased intervention decisions of national supervisors. Taking banks’ CDS spread at the time of intervention as a measure of regulatory lenience (with a higher spread indicating intervention at a later stage), we find that higher foreign asset and deposit shares and a lower foreign equity share are associated with later intervention. The intuition for this result is that the gains from letting a weak bank continue mainly accrue to equity, while the costs accrue to debt holders and other stakeholders in the economy; for example borrowers of foreign subsidiaries and branches lose access to external funding. A high share of foreign depositors and borrowers implies that a higher cost of bank failure accrues to stakeholders outside the supervisor’s jurisdiction, while a higher share of equity holders implies that a higher benefit of allowing the bank to continue operating accrues outside the supervisor’s jurisdiction.

Would national bias in supervisory decisions imply that a global supervisor would be the optimal institutional solution as suggested by some observers after the Global Financial Crisis? In Beck and Wagner (2016) we show that this is not necessarily the case, as there are two off-setting factors.  On the one hand, there are negative cross-border externalities from bank failure (including ownership links described above, fire sale externalities, informational contagion, or panics when countries’ capital markets are integrated, regulatory arbitrage and a common currency), which raise the benefit of delegating the intervention decision to the supranational level. On the other hand, there is a cost from cooperation, as common policies may not be optimal for either country (or both of them).  This could be due to countries perceiving different costs to letting banks fail or having different banking market structures so that the failure of a bank imposes different costs on either economy.  Beck and Wagner (2016) show with a simple theoretical model, that supranational supervision is more likely to be welfare enhancing when externalities are high and country heterogeneity is low. This suggests that different sets of countries (or regions) should differ in the extent to which their regulators cooperate across borders.  This might explain, for example, why there has been such a strong push after the Eurodebt crisis for a banking union in Europe, but non-euro area countries of the European Union have been reluctant to join. 

Beyond anecdotal evidence from specific countries or regions, do we observe systematic differences in cooperation patterns across the globe in line with different externalities and heterogeneity? Beck, Silva-Buston and Wagner (2023) show that cooperation pattern observed in data for 93 countries between 1995 and 2013 vary consistently with (net) cooperation gains arising from externalities and heterogeneities. Specifically, we show that country pairs with higher bilateral externalities from cross-border banking and capital market integration are more likely to have a supervisory cooperation agreement in place, while country pairs with higher bilateral heterogeneity are less likely to do so.   Higher externalities and lower heterogeneity between countries also make it more likely that countries cooperate in more intense forms (e.g., having a common supervisor instead of only exchanging information), and that they also accelerate cooperation.

Cross-border cooperation and financial stability

Does closer supervisory cooperation between countries also help make cross-border banks more stable and less likely to fail?  While failure of cross-border banks is rare to observe, one can focus on measures of bank’s stability, such as the accounting-based measure - Z-score indicating distance to default – or a market-based gauge - the bank’s Marginal Expected Shortfall, which measures a bank’s average return when the market experiences stress, thus capturing systemic risk exposure (Acharya, Pedersen, Philippon, and Richardson, 2017).

Beck, Silva-Buston and Wagner (2023) find that a higher incidence of supervisory cooperation is associated with higher bank stability, but only for relatively smaller cross-border banks, consistent with supervision at larger banks being less effective due to too-big-to-fail and higher complexity.  The link between cooperation and bank stability runs through asset risk rather than capital levels, consistent with the notion that asset risk is difficult to observe and control at arms-length; intensive cooperation and information exchange should hence have a pronounced effect. Effectiveness of cooperation increases both with the stringency of home and host supervision, as well as the quality of information that is available to supervisors.

These more general result in a broad cross-country sample are also confirmed by studies focusing specifically on the introduction of the Single Supervisory Mechanism in the euro area in 2014.  Fiordelisi, Ricci, and Stentella Lopes (2017) show that banks that expected to come under the supervision of the SSM reduced their lending activities and increased their capital ratios in comparison with banks below the asset threshold for supervision by the SSM. Similarly, Eber and Minoiu (2016) show that SSM-supervised banks reduced their asset size and reliance on wholesale debt over the period 2012-15, compared with banks that did not fall under the supervision of the SSM. Finally, Altavilla, Boucinha, Peydro, and Smets (2020) use loan-level data and show that banks under supranational supervision reduced credit supply to firms with very high ex-ante and ex-post credit risk, while stimulating credit supply to firms without loan delinquencies. Overall, the introduction of a supranational supervisor thus resulted in less risk-taking by banks subject to its mandate.

Regulatory arbitrage undermining supervisory cooperation

However, banks react not only to changes in regulation (often referred to as regulatory arbitrage) but also changes in supervisory structures. An extensive literature has shown that banks react to regulatory reforms or regulatory differences between countries and Beck, Silva-Buston and Wagner (2025a) show that this is also the case when countries start cooperating with each other. 

Specifically, Beck, Silva-Buston and Wagner (2025a) use data on  banking groups in 47 home and 116 host countries during 1995-2019, and show that a subsidiary's lending (as a proxy for the amount of risk-taking) increases when supervisory cooperation between the country of the parent bank and the other host countries of the group increases. This lending increase goes hand in hand with a higher riskiness of the subsidiary in general: the balance sheet becomes more leveraged, and default risk (as proxied by the Z-score) increases. These findings are also confirmed on the loan-level: we find that the probability of allocating a specific syndicated loan to a particular subsidiary increases in the supervisory cooperation coverage of the rest of the banking group.

This finding of regulatory arbitrage following supervisory cooperation agreements might also explain why Beck et al. (2023) did not find stability-enhancing effects from cooperation for the largest cross-border banks as these banks have more opportunities to shift risks into third countries as long as supervisory cooperation is not complete. Risk-shifting into third countries can be mitigated (that is, lending responds less to cooperation between the home supervisor and the supervisors of other host countries) when the subsidiary country has stricter supervision and better market discipline  relative to other countries in which the group has foreign operations. Risk-shifting is also mitigated when the subsidiary country cooperates with the home country, as well as when it cooperates with the other host countries of the group.

Is there also a bright side to such regulatory arbitrage, i.e., banks shifting risks to third countries in reaction to increased cross-border supervisory cooperation? Beck, Silva-Buston and Wagner (2025b) show that subsidiaries of banking groups improve loan conditions for firms when the group’s opportunities to take risks in other countries are reduced due to cross-border supervisory cooperation. The expanded lending is targeted toward higher-quality firms and those with which the group is already familiar. The improved lending conditions have positive real effects, allowing recipient firms to increase capital spending and profits. 

These results suggest that there can be benefits for firms in countries that receive lending inflows due to regulatory arbitrage. The fact that the additional lending seems targeted towards high-quality and safe firms, coupled with the fact that financial constraints are likely to be tighter in countries with weaker supervisory standards (and hence the ones receiving the inflows), allows for the possibility that a reallocation of lending between countries on net alleviates financial constraints, resulting in more desirable projects being funded.

The small host country conundrum

There are asymmetric incentives to cooperate between parent supervisors of large cross-border banks and host supervisors of their subsidiaries in smaller economies.  Take the example of Bosnia-Herzegovina.  In 2017, Unicredit had a market share of 24%, while the subsidiary made up 0.4% of Unicredit’s total balance sheet. (Ahmad Fontan et al., 2019).  More generally, the subsidiaries of many Western European countries have important market shares in many smaller Central, Eastern and South-Eastern European countries, but these subsidiaries make up a miniscule part of the parent banks’ consolidated balance sheets. For the home country, cooperating with a small host country is an afterthought, especially when convening a college of supervisors. For the host country, on the other hand, such cooperation is rather critical. The reaction of small host countries is often to ring-fence, which, however, undermines the benefits from cross-border banking. At the same time and as discussed above, such asymmetry and regulatory arbitrage reactions by cross-border banks might have some benefits for recipient countries if more lending reduces financing constraints.

Conclusions and policy implications

National supervision and the global footprint of banks and capital markets create a tension that is not easy to overcome. Where countries’ preferences and interests are aligned, the move to a supranational supervisor can be optimal if cross-border externalities from bank failure are large.  In other cases, lighter forms of cross-border supervisory cooperation might be more appropriate.  While supervisory cooperation across border can enhance financial stability, cross-border banks also react by shifting risks into third countries. 

In sum, cross-border supervisory cooperation can help enhance financial stability but absent a move towards supranational supervision, the prime responsibility still stays with national authorities. Stronger national supervisory and resolution frameworks can enhance stability benefits from cross-border supervisory cooperation or – in their absence – reduce risk shifting of cross-border banks into this jurisdiction. 

References

Ahmad Fontan, I., T. Beck; K. d’Hulster, P. Lintner, and F. D. Unsal (2019). Banking Supervision and Resolution in the EU: Effects on Small Host Countries in Central, Eastern and South Eastern Europe. World Bank, FinSAC

Acharya, V.; L. Pedersen; T. Philippon; and M. Richardson. 2017.  Measuring Systemic Risk. Review of Financial Studies 30, 2–47.

Altavilla, C., Boucinha, M., Peydró, J.-L., & Smets, F. R. (2020). Banking supervision, monetary policy and risk-taking: big data evidence from 15 credit registers. ECB Working Paper No 2349.

Beck, T., Todorov, R. and Wagner, W. (2013), Bank Supervision Going Global? A Cost-Benefit Analysis, Economic Policy 28, 285–44.

Beck, T., and Wagner, W.. Supranational Supervision: How Much and for Whom? International Journal of Central Banking, 12 (2016), 221–268.

Beck, T, Silva-Buston, C. and Wagner, W. (2023), The Economics of Supranational Bank Supervision, Journal of Financial and Quantitative Analysis 58, 324-51.

Beck, T, Silva-Buston, C. and Wagner, W. (2025a), Supervisory cooperation and regulatory arbitrage, Review of Finance 29, 381–413.

Beck, T, Silva-Buston, C. and Wagner, W. (2025b), Regulatory arbitrage and real effects. Working paper.

Eber, M. and C. Minoiu (2016). How do banks adjust to stricter supervision? Mimeo.

Fiordelisi, F., Ricci, O., & Stentella Lopes, F. S. (2017). The unintended consequences of the launch of the single supervisory mechanism in Europe. Journal of Financial and Quantitative Analysis 52, 2809–2836.