AdobeStock_281930915.jpeg
VoxEU Column EU institutions Financial Regulation and Banking

The new supervisory architecture in Europe

The decade since the Global Crisis has seen notable changes in the architecture of supervision, with separation of responsibility for monetary and financial stability having been reversed in many countries on the one hand, and a move towards more cross-border cooperation between supervisors on the other. This column discusses these two trends in Europe, where responsibility for supervision of the largest banks is housed in the same authority with responsibility for monetary policy, the ECB. It argues that the Single Supervisory Mechanism is a good reflection of the subtle economics of supervisory architecture and the many trade-offs that have to be taken into account.

The decade since the Global Crisis has seen notable changes in the architecture of supervision across Europe.  On the one hand, the tendency to separate the responsibility for monetary and financial stability into two different authorities has been reversed in many countries (Melecky and Podpiera, 2013). One prominent example is the UK, where the Financial Supervisory Authority was dissolved and the responsibility for bank supervision returned to the Bank of England.  On the other hand, there has been a move towards more cross-border cooperation between supervisors, which in the case of the euro area culminated in the creation of a Single Supervisory Mechanism (SSM) and a Single Resolution Mechanism (Beck et al. 2018). These two trends interact with each other in the euro area, as the direct responsibility for the supervision of the largest banks is housed in the same authority with responsibility for monetary policy, the ECB. 

In a recent paper, we discuss these two trends in Europe, in light of a growing stream of research and ongoing policy debate on supervisory architecture (Ampudia et al. 2019). 

In or out: Supervisors and central banks

The pre-crisis paradigm saw monetary stability and financial stability as separable objectives, with central banks being responsible for monetary and prudential authorities for financial stability. The Global Crisis has shown that monetary and prudential policy interact with each other and both affect monetary and financial stability through different channels (Figure 1). The central bank influences financial stability for example through the risk-taking channel of monetary policy (e.g., Dell’Arriccia et al. 2014, Jimenez et al. 2014) while being directly involved in the financial safety net as lender of last resort. At the same time, supervisors have an indirect impact on monetary policy, as bank stability is an important pre-condition for effective monetary policy transmission. 

Figure 1 Interactions between prudential and monetary policy and their respective objectives

But does this imply that the same authority should be responsible for both? The literature has provided arguments in favour and against consolidating these responsibilities into one institution. Arguments in favour include the following: 

  • Consolidating responsibilities can help avoid coordination failure and account for the interdependence of the policies discussed above (Carrillo et al. 2017). 
  • Central banks can benefit from supervisory information when assessing monetary policy decisions.  Having supervisors in the same institution implies lower frictions in transferring information (Peek et al. 1999)
  • A unified structure of central banking and supervision makes it possible for the supervisor to benefit from the independence and the reputation of the central bank, thus limiting the risks for political and regulatory capture. 

Arguments against include the following:

  • The reputation of the central bank and, in turn, its credibility and effectiveness could be negatively impacted by damages to the reputation of the supervisor following a bank failure.
  • There could be distortions in decision making, such as deviating from the optimal conduct of monetary policy in an attempt to preserve the stability of the financial institutions or being too lenient as a supervisor to reduce losses for the central bank from lender of last resort operations.

Assessing these arguments in a cross-country setting is difficult.  We add to the existing literature with simple cross-country comparisons, without implying causality. 

  • Based on different fixed and random effects models, we do not find evidence that GDP growth is lower in countries and years where bank supervision is integrated into the central bank 
  • We also find no evidence of higher deviations from the inflation target if bank supervision is integrated into the central bank. 
  • We find a higher probability of a credit boom turning into a banking crisis in countries where bank supervision is outside the central bank. 
  • We find that in countries and years where bank supervision is in the central bank, there is a higher likelihood that loan-to-value ratios are used as macroprudential tools during credit booms and that credit booms are less likely to turn into crises. 

This suggestive evidence is not consistent with arguments against unifying responsibilities for monetary and financial stability into one institution.  If at all, it points to potential financial stability gains from integrating supervisory responsibilities into the central bank. 

The choice whether to separate bank supervision and monetary policy involves a complicated trade-off between different objectives. The design chosen in the euro area is a compromise between separation and integration. In the case of the euro area, the SSM is part of the ECB, but a strict separation principle applies: supervisory and monetary policy objectives are clearly separated, as well as the decision bodies and operational units. 

Centralised versus decentralised supervision

The debate on supervisory architecture has also a second dimension to it, which is the degree to which banks should be supervised by local/national supervisors or centralised supervisors. This debate is particularly relevant for the euro area, where a hybrid structure has emerged, but also for the US, where banks can be supervised by state or federal supervisors. 

Arguments for decentralised supervision include:

  • better information availability of geographically and culturally closer supervisors (see Gopalan et al. 2017, for evidence based on the US), and
  • better knowledge of specificities of local credit markets

Arguments for centralised supervision include:

  • economies of scale in supervision and thus efficiency gains (Eisenbach et al. 2016),
  • centralised supervisors taking into account externalities of bank failures outside local jurisdictions (see Beck et al. 2013 for evidence during the Global Crisis), and
  • centralised supervisors being less lenient with failing banks compared to local supervisors (see Agarwal et al. 2014 for evidence in the US).

The recent literature has also addressed how a two-tier supervisory architecture that combines local and centralised supervisors can optimally distribute responsibilities. The local supervisor is more effective in acquiring information, while the centralised supervisor faces a lower cost in taking an intervention and liquidation decision (Repullo 2018). However, if there is a conflict in objectives, local supervisors might reduce their effort in collecting and transmitting information (Carletti et al. 2016). 

Moving to centralised supervision might also impact the behaviour of banks and other market participants. If local supervisors are too lenient with banks that have too many foreign creditors, introducing a central supervisor should allow banks to borrow more easily internationally, and at lower rates. Central supervision would, as a result, indirectly contribute to financial integration (Colliard 2019). 

If one views the design of the SSM through the lens of the recent literature, the chosen architecture seems efficient under the assumption that information acquisition for supervising large banks entails only a small cost advantage for a local supervisor relative to a central supervisor, while the risk of being exposed to regulatory capture is potentially very large, this is consistent with the arguments above. The ECB (i.e. the central supervisor) is responsible for the supervision of significant banks in cooperation with local supervisors through the Joint Supervisory Teams (JSTs), while less significant banks are supervised only by local supervisors. 

While the SSM started operating only a few years ago, some initial evidence is consistent with the efficiency of this structure. Specifically, Fiordelisi et al. (2017) show that the most significant banks (i.e. those banks affected by the change in supervisory authority) reduced their lending activities and increased their capital ratios in comparison with less significant banks (i.e. those banks below the asset threshold for supervision by the SSM) in the run-up to the SSM. Similarly, Eber and Minoiu (2016) show that SSM banks reduced their asset size and reliance on wholesale debt. 

The introduction of the SSM is probably the largest recent change in supervisory architecture in developed countries. It reflects well the subtle economics of supervisory architecture and the many trade-offs that have to be taken into account. Indeed, the chosen design tries to find a middle ground between (i) integration versus separation with monetary policy; (ii) local versus central supervision; (iii) centralisation versus delegation of information collection versus decision-making and rule-setting

References

Agarwal, S, D Lucca, A Seru, and F Trebbi, (2014), “Inconsistent regulators: Evidence from banking”, Quarterly Journal of Economics, 129(2). 

Ampudia, M, T Beck, A Beyer, J-E Colliard, A Leonello, A Maddaloni and D Marqués-Ibanez (2019), “The architecture of supervision." European Central Bank Working Paper Series 2287. 

Beck, T, R Todorov, and W Wagner (2013), “Supervising cross-border banks: theory, evidence and policy,” Economic Policy, 28(73).

Beck, T, C Silva-Buston, and W Wagner (2018), “The economics of supranational bank supervision.” CEPR Discussion Paper.

Carletti, E, G Dell'Ariccia, and R Marquez (2016), “Supervisory incentives in a banking union,” International Monetary Fund.

Carrillo, J A, E G Mendoza, V Nuguer, and J Roldán-Peña (2017), “Tight money-tight credit: Coordination failure in the conduct of monetary and financial policies,” NBER working paper 23151.

Colliard, J-E (2019), “Optimal supervisory architecture and financial integration in a banking union,” Review of Finance, forthcoming.

Dell'Ariccia, G, L.Laeven and R Marquez (2014), “Real interest rates, leverage, and bank risk-taking,” Journal of Economic Theory, 149, 65-99.

Eber, M and C Minoiu (2016), “How do banks adjust to stricter supervision?,” Working paper.   

Eisenbach, T M, D O Lucca, and R M Townsend (2016), “The economics of bank supervision,” NBER Working Papers 22201.

Fiordelisi, F, O Ricci, and S Lopes (2017), “The unintended consequences of the launch of the single supervisory mechanism in Europe,” Journal of Financial and Quantitative Analysis, 52(6), 2809-2836.

Gopalan, Y, A Kalda and A Manela (2017), “Hub-and-spoke regulation and bank leverage,” Discussion paper.

Jimenez, G, S Ongena, J L Peydró and J Saurina (2014), “Hazardous times for monetary policy: What do twenty‐three million bank loans say about the effects of monetary policy on credit risk‐taking?,” Econometrica, 82(2), 463-505

Melecky, M and A M Podpiera (2013), "Institutional structures of financial sector supervision, their drivers and historical benchmarks," Journal of Financial Stability, 9(3), 428-444.

Peek, J, E S Rosengren and G M B Tootell (1999), “Is Bank Supervision Central to Central Banking?”, Quarterly Journal of Economics, 114(2), 629-653.

Repullo, R (2000), “Who should act as lender of last resort? An incomplete contracts model,” Journal of Money, Credit and Banking, 32, 580-605.

Repullo, R (2018), “Hierarchical Bank Supervision”, SERIEs – Journal of the Spanish Economic Association, 9, 1-26.

4,094 Reads