Large scale bank bailouts can be very wasteful for society in terms of resources lost and increased public debt [1]. The doom loop means that the public cost of funding a bailout can place a serious strain on public finance and undermine country solvency. The fiscal cost of the recapitalisation and asset relief of 22 large European banks and 13 large US banks amounted to €298 billion and $US205 billion respectively, and the cash injections to UK banks were up to £133 billion [2]. More specifically, the UK’s fiscal cost amounted to 8.8% of GDP, whereas in Ireland it was 41% of GDP [3]. In both cases public debt spiked and in the case of Ireland the country had to be placed under troika supervision in the context of an EU rescue package. In fact, it is the size of its public debt and of its economy that saved Italy from a similar fate during the great turmoil that its banking sector underwent during 2015-2017.

In response to the 2008 financial crisis, policy makers introduced measures to increase financial stability and reduce the likelihood of large-scale bank bailouts. These measures may be classified into activist policy and other structural reforms. The former tend to perform countercyclical stabilization functions whereas the latter are meant to permanently increase financial stability by altering participants’ behaviour. 

Structural measures, on the other hand, include separation of banking activities in the form of the UK’s ring-fencing regime or the less draconian Volcker Rule, mandatory recovery and resolution plans (living wills), restrictions on executive pay, a Senior Managers Regime like the UK’s, and special resolution regimes. The latter are special legal and regulatory regimes that deal with bank failures both ex ante and ex post and facilitate orderly exit while minimising systemic disruption to minimize contagion and preserve critical functions [4].

The fundamental principles of special resolution regimes in all three jurisdictions are premised on the assumption that the costs of bank failure should be internalised falling on bank shareholders and bank creditors rather than the taxpayer [5]. They have thus incorporated a bail-in mechanism under which bank creditors take a haircut on their debt holdings or see their debt converted into equity to absorb the losses of a bank in resolution regardless of whether the bank is thus recapitalised to carry on as a going concern or it becomes a gone concern.

As bailing in general bank creditors could be disruptive, regulators have introduced increased capital buffers and the concept of eligible liabilities to be bailed-in such as the Total Loss Absorption Capital requirements introduced by the Financial Stability Board and endorsed by regulators worldwide and the BBRD’s Minimum Requirements of own funds and Eligible Liabilities. Furthermore, special resolution funds like the Eurozone’s Single Resolution Fund can absorb the balance of losses.

But bail-in action within resolution supported by the increased buffers has not been tested yet in action, especially in the context of the failure of a big cross-border or nationally systemically important bank. Thus, it is a matter of considerable speculation whether the buffers would prove sufficient to prevent a bailout in the event of a major bank failure. If the buffers do not prove sufficient then regulators will probably still conduct a bailout since the bailing-in of liabilities beyond the specially designated ones could just spread contagion [6]. Therefore, the best way to minimize the possibility of a string of bailouts in the future is to revisit the incentives of key bank constituents such as shareholders and senior management.

The problem of management incentives

Redressing the incentives of those who make the actual business decisions in a bank, namely senior management and the shareholder appointed board, is one of the biggest challenges facing contemporary financial stability regimes. Rent-seeking behaviour or recklessness by those two key constituencies is always behind risky lending that ultimately leads to bank failures [7]. By comparison post-2008 resolution regimes seem to reinforce market discipline by just focusing on bank creditors. But there are many reasons to explain why creditors are not the best bank management monitors or the more stable source of loss absorption.

First, the agents of large shareholders who are the actual decision-makers, have no reason to make decisions that protect creditors’ interests [8]. Second, by its very nature bank business is quite impenetrable to outside scrutiny especially as regards the bank’s stability and the quality of its assets. Third, banks do not just fail because they have engaged in bad lending but also because of interconnectedness, especially when, under conditions of generalised crisis, the value of bank assets depreciates due to fire sales. Finally, some creditors may not be much more than passive investors holding bank debt only because interest rates are low, and they find it very hard to locate any acceptable returns in other investments.

All this indicates that the absence of the threat of punitive sanctions imposed on bank management and shareholders, which go beyond clawbacks, leaves a gaping hole in contemporary financial stability regimes. Any super-imposed mechanism to redress shareholders and senior managers’ incentives over what is provided by today’s prudential regimes may only be an ex post sanction. But for the deterrence effect of such a sanction to be credible the sanction must be automated offering 100% certainty of imposition of the sanction ex post. It must also be a sanction that is unbiased and objectively calibrated. For example, one of the biggest shortcomings of the current regime on management compensation clawbacks in the event of bank failures and the imposition of regulatory sanctions or the initiation of criminal proceedings is that the respective regimes can appear random and improbable in their ex-post application, lowering the degree of deterrence associated with the threat of the sanction.

Proposal: Optimal financial penalties on bankers

Nolan, Sakellaris and Tsoukalas have argued in an earlier paper that bank bailouts are unlikely to have ended as a result of the recent prudential reforms [9]. They argue that policymakers should publicly acknowledge that bailouts may yet occur but commit to penalising senior bank management if any bailout does in fact occur. Nolan, Sakellaris and Tsoukalas propose that a tax on bankers’ pay (following a bailout) which is made conditional on government’s fiscal capacity could limit bankers’ risk-taking activity and, consequently, the likelihood of bail-outs. Amongst other things, such a framework would cause bankers and shareholders to confront the implications of their choices to run a systemically important institution.

The penalty that Nolan, Sakellaris and Tsoukalas propose is modelled, for simplicity, as a consumption tax. They argue that the penalty is attractive for a number of reasons [10]. First, it recognises that bail-outs may be an effective way to stabilize the financial system. Second, it attempts explicitly to locate a remedy to the moral hazard problem at its source. Third, it is time consistent. That is, it recognises that any future bailouts may be undertaken, if needed. No implausible threats will be made by regulators about the impossibility of any future bank bail outs. Yet those who make the key decisions: senior managers and shareholders, will have to do their utmost to avert such a possibility as it will be them who stand to lose the most.

Implementing the penalty

Following most bailouts, shareholders are typically ‘wiped out’. Nolan, Sakellaris and Tsoukalas see the penalties falling on key individuals in designated systemically important financial institutions. That would be consistent with recent supervisory developments. For example, in some supervisory regimes certain regulated institutions already have to identify key individuals. In addition, the Financial Stability Board designates systemically important financial institutions. And supervisory agencies in many countries increasingly take a view on which institutions they regard as of national systemic importance and which roles and named individuals are key. The penalty need not be levied on key individuals at all institutions that are bailed out. Smaller institutions might not face such penalties. If that turned out to lead to gaming of the system then the scope of the penalties could be widened. Penalties could be substantial financial fines enforceable by courts. Or, they could be deductions from pension rights.

According to several economists, many of the challenges society faces from the financial sector reflect political preferences; they argue that banks are ‘fragile by design’ or, similarly, that financial regulation is feeble by design, because they are ridden with perverse incentives due to the conflicting interests of key interest groups. Why might then the suggested penalties be more robust than other regulatory measures? First, these penalties should be transparent and their form known to all, since they will be agreed ex ante. Second, the levying of these penalties following a bailout would be giving satisfaction to the public’s feeling of justice. Provided that the sanction is modelled in such a way as to be sufficiently severe it should have an instant ex ante impact on the behaviour of banks and bankers who would be aware that the possibility of imposition of the sanction is reaching absolute certainty.

Penalties should be conditioned on fiscal conditions for two reasons. First, higher fiscal capacity for bailouts implies higher risk taking from banks. Therefore, for the penalty to tackle moral hazard effectively it has to increase in some proportion with the size of an anticipated bailout, which, as the authors of this post argue in a joint forthcoming paper, should in turn relate to the overall levels of bank riskiness—this is the ex-ante aspect of the penalty. Second, at the time of the bailout, the implementation of the penalty is calibrated to reflect wider market conditions and the fiscal position of the country in question.  If it is more expensive for the government to raise funds, a heavier penalty will be levied—this is the ex-post aspect of the penalty.  These aspects constitute a key difference compared to claw-back bonus rules which often are a foregone gain rather than a personal loss and are not guaranteed to reduce excessive risk taking.

As already mentioned, one of the principal objectives of post-2008 bank regulation is to minimize taxpayer intervention which, of course, will also offer relief to countries with low fiscal capacity due to either the size of the economy or high levels of public debt. Clearly an elimination of bankers’ and shareholders’ incentives to induce a bailout will boost financial stability.


As we argue in our joint forthcoming paper the prescribed sanction on management can lead to multiple benefits. Where a bailout is necessary to restore confidence and maintain macroeconomic and financial stability, penalties can still be levied on individuals without exacerbating systemic risk. Moreover, the sanction can provide powerful incentives for bank management to take prompt action and prevent the piling up of non-performing loans, whereas bail-ins may act as a disincentive on management trying to avert that possibility for as long as possible [11]. Finally, the modelling of the possible ex ante impact of such a unified and draconian ex post regime on risky bank loans could act as a major counter-balance to rational herding. This could lead to balance sheet diversification creating a more resilient financial ecosystem.

Emilios Avgouleas is the Chair in International Banking Law and Finance at the School of Law, University of Edinburgh.

Charles Nolan is the Bonar MacFie Chair in Economics at the Adam Smith Business School, University of Glasgow.

John Tsoukalas is Professor of Economics at the Adam Smith Business School, University of Glasgow.


[1] Laeven, L. and Valencia, F. (2013). Systemic banking crises database. IMF Economic Review, 61.

[2] Schoenmaker, D. (2016). Resolution of international banks: Can smaller countries cope?”, CEPR Discussion Paper No. 11600.

[3] Cunliffe, J. (2016). Ending too-big-to-fail: How best to deal with failed large banks. European Economy. Banks regulation and the Real Sector, 2.

[4] E.g., Dodd Frank Act’s Orderly Liquidation Authority (OLA) in the US, the Bank Recovery and Resolution Directive (BRRD) in Europe, and the UK’s Special Resolution Regime (SRR) under the Banking Act 2009 and Financial Services (Banking Reform) Act 2013.

[5] Zhou, J. et al. (2012). From Bail-out to Bail-in: Mandatory Debt Restructuring of Systemic Financial Institutions. IMF Staff Discussion Note 12/03.

[6] Avgouleas, E, Goodhart, C. (2015). Critical reflections on bank bail-ins. Journal of Financial Regulation 1, 3; Dell’Ariccia et al. (2018). Trade-offs in Bank Resolution. IMF Staff Discussion Note 18/02.

[7] Admati et al. (2013). Fallacies, Irrelevant Facts, and Myths in the Discussion of Capital Regulation: Why Bank Equity is Not Expensive. Stanford GSB Research Paper13-7; Avgouleas, E., Cullen, J., (2015). Excessive Leverage and Bankers’ Pay: Governance and Financial Stability Costs of a Symbiotic Relationship. Columbia Journal of European Law 22/1.

[8] Admati, A.R., DeMarzo, P.M., Hellwig, M. and Paul Pfleiderer (2018). The Leverage Ratchet Effect. Journal of Finance 73/1, 145.

[9] Nolan, C., Sakellaris, P. and Tsoukalas, J. (2016). Optimal Bailout of Systemic Banks. University of Glasgow Working paper.

[10] Ibid.

[11] Avgouleas, E., Goodhart, C. (2019). Bank Resolution a Decade after the Global Financial Crisis: A Systematic Reappraisal. In Arner, D., Avgouleas, E. et al. (eds), Systemic Risk in the Financial Sector: Ten Years after the Great Crash; Avgouleas, E., Goodhart, C. (2016). An anatomy of bank bail-ins. European Economy. Banks regulation and the Real Sector, 2.


With the support of