Guest post: Why regulators should focus on bankers’ incentives

Charles Goodhart

Last autumn, Charles Goodhart gave a special lecture at the Bank.  In this guest post he argues that regulators should focus more on the incentives of individual decision makers.

The incentive for those in any institution is to justify and extol the virtues of the decisions that they have taken.   Criticisms of current regulatory measures are more likely to come from outsiders, perhaps especially from academics, (with tenure), who can play the fool to the regulatory king.   I offer some thoughts here from that perspective.  I contend that the regulatory failures that led to the crisis and the shortcomings of regulation since are largely derived from a failure to identify the persons responsible for bad decisions.   Banks cannot take decisions, exhibit behaviour, or have feelings – but individuals can.  The solution lies in reforming  the governance set-up  and realigning incentives faced by banks’ management.

Recent regulatory problems have been greatly reinforced by a widespread tendency to apply human characteristics, i.e. to anthropomorphise, to an inanimate institute, in this case a bank.  We tend to talk about Bank X having assumed too much leverage, or having behaved in an improper fashion—rather than management of Bank X did such and such; We say that Bank X got bailed out rather than the creditors and clients of Bank X got bailed out.

The outcome has been a regulatory system primarily based on imposing a structure of regulations on banks, with insufficient concern about the incentives on bankers to adjust to, and to ‘game’, that system.  By the same token there has been insufficient concern for reform of the incentive and control system facing bank managers.  Some reforms of the governance structure for bankers have been introduced in the UK, for example in the guise of claw-back rules and the senior managers’ regime.  But even while the Bank of England has been in the vanguard on this, I believe that much more could, and should, be done in this respect.

There are two questions that need reconsideration.  The first relates to the scope of responsibility for outcomes in a hierarchical institution; the second relates to the downside that those responsible should face when failure or bad behaviour occurs.  When bad behaviour occurs at a lower level, e.g. traders conspiring to rig Libor, or junior employees mis-selling products onto uninformed customers, should managers, directors and CEOs be able to shelter behind the fact that they did not know what was going on?  Should they, could they, have taken steps to find out?  There is surely a case for reversing the burden of proof.  Each manager should be held responsible, and subject to suit, unless they can demonstrate that they took all reasonable measures to oversee and to constrain the actions of their juniors.  The New Zealand procedure of requiring more self-discipline amongst bank directors could be adopted.

Similarly the personal liability of shareholders could be related to their informational advantage and capacity to influence decisions.  Junior employees and outside shareholders, up to a holding of X% of market value, would keep limited liability.  Junior managers, and large shareholders, could have double liability; senior managers, perhaps, treble liability, and CEOs perhaps unlimited liability.

We would be told that such measures would make banks unduly risk averse and prevent qualified bank managers going into the profession.  Where is the evidence for that?  Banking worked on the basis of unlimited liability up to the second half of the 19th Century, and banks then took plenty of risks.

Having argued that the desideratum for financial regulation should have been to reform the incentive structure, I consider four areas of current and recent banking reforms where failure to consider the incentive and informational structure adequately has weakened their efficiency.

  1. Higher Required Capital Ratios

Left to themselves, bankers will meet higher Capital Adequacy Ratios by deleveraging and withdrawing from low profit, possibly safer but capital intensive business.  This will have adverse effects on economic recovery and capital markets.

The need is to impose incentives on bankers, and shareholders, to restore the level of equity (or to redeem a prior public sector injection of equity) to a desired quantum, by preventing any share buy-backs, dividend payment or increase in managerial compensation until that quantum (n.b. not ratio) was reached.  This was done in the USA with the TARP.  On the whole, it was not done in continental Europe.  So banking recovery has been weaker in the latter.

  1. Bail-in

Outsiders, even large bond-holders, remain at a serious informational disadvantage and so they are unlikely to be able to constrain the riskier actions of management to a significant degree.  If such informational asymmetry is recognized by providing protection for all small bond-holders, with a holding of less than X in value, then all large institutional investors will maintain a holding of precisely X-1 in all such bail-inable bonds.

Whereas the principle that bond-holders should not be (completely) rescued by a transfer from taxpayers is sound, the idea that such bail-in can fully replace a public-sector bail-out in the face of a severe systemic shock is unduly optimistic.  Avgouleas and Goodhart argue that history shows that when a bank’s difficulties reflect broader macroeconomic problems rather than bank-specific issues, imposing haircuts on creditors of one bank is likely to accelerate panic and risks contagion to other institutions, which may require public funds to shore up the system as a whole. Indeed, the extent of penalty to be imposed on bail-inable bond holders in the European Union (8% of a bank’s liabilities, including own funds) before any public support becomes available, is so great that the failure of one bank, unless for totally idiosyncratic reasons, is likely to lead to a widespread collapse of the bank bond market as a whole, at least for a time, with implications for contagion.

Moreover, a bank resolved by bail-in may have had its equity base restored, but will nevertheless be clearly weaker than it will have appeared before,  with the likely consequence that it will face a major outflow of deposits, such as occurred in the aftermath of the rescue of Continental Illinois in 1984.  The injection of public sector capital under bail-out will be replaced by the injection of public sector liquidity from the Central Bank under bail-in, which could put the latter in the political firing line.

  1. Structural Reform

Although losses for retail depositors and failures of payment mechanisms were conspicuous by their absence in the recent crisis, this was largely due to the policy of bail-out.  If bail-out is to be cut back, perhaps further safeguards need to be introduced to maintain ‘essential’ banking activities.  One set of suggestions is to impose constraints on the structure of banking, to make banks smaller, or less complex. The proposals of Vickers, Volcker and Liikanen come to mind.

But so long as bankers have an incentive to assume more risk, then they will try to work around such structural changes.  If, instead, incentive structures were changed to impose appropriate penalties for failure on the banker, then bankers would themselves choose whatever structure, perhaps smaller and simpler, would provide them with an acceptably reduced chance of failure.  Historically, unlimited liability for bank shareholders was abolished principally to allow larger banks capable of financing big business (and big government) to emerge.  If the current concern is that banks have become too big and too complex to manage, or to allow to fail, why not just raise the managerial penalty for failure?

Managers on the spot will have a better idea of what they can safely undertake, than illustrious independent outsiders. Leaving the current incentive system unchanged, while seeking to enforce structural change by diktat will have bank management, and their skilled professional advisers, seeking loop-holes.

  1. Fines for Bad Behaviour

The current practice of imposing massive fines on banks for the prior misbehaviour of some of their (prior) employees is monstrously unfair and inefficient.  It is unfair because it primarily hits shareholders, who had no direct responsibility, bank clients, since the bank involved has to cut back and raise profit margins to meet impaired capital adequacy, and the economy as a whole because such fines represent a massive head-wind against bank participation in the economy.  It is inefficient because shareholders can apply little effective pressure on management, and current management can generally claim that the misbehaviour occurred under past management, all the while secure in the likelihood that by the time current malpractice comes to light, and to adjudication, they too will have long been retired from the job and able to claim ignorance of such events, as a splendid letter to the FT recently pointed out most lucidly.

Regulators should not levy fines on banks, despite them having the legal status of a person, and instead apply the fines to those responsible at the time of the offense, whether subsequently retired or not.  As earlier noted, the onus of proof should be on the managers of those who committed the misdeed, and so on up the hierarchy, to convince (a jury) that they took all reasonable steps to prevent the misdeeds of their subordinates.  The size of fine could be related positively both to the extent of negligence and to rank.

Of course, the threat of personal liability and loss could make bankers overly cautious, as is often said of US medical practice, with unnecessary and costly tests of patients to reduce the threat of suits.  As usual, there would be an optimum internal degree of liability, but we are currently well below it.


If a bank CEO knew that his own family’s fortunes would remain at risk throughout his subsequent lifetime for any failure of an employee’s behaviour during his period in office, it would do more to improve banking ‘culture’ than any set of sermons and required oaths of good behaviour. The root of the problem is the bad behaviour of bankers, not of banks, who are incapable of behaviour, for good or ill.   The regulatory framework should be refocused towards the latter, with a focus on reforming incentives.

Charles Goodhart is Emeritus Professor at the London School of Economics.  

If you want to get in touch, please email us at or leave a comment below.

Comments will only appear once approved by a moderator, and are only published where a full name is supplied.

Bank Underground is a blog for Bank of England staff to share views that challenge – or support – prevailing policy orthodoxies. The views expressed here are those of the authors, and are not necessarily those of the Bank of England, or its policy committees.


Filed under Banking, Financial Stability, Microprudential Regulation, Resolution

3 responses to “Guest post: Why regulators should focus on bankers’ incentives

  1. Michael Bruce

    Excellent paper. It’s an axiom of human behaviour that people do what they are rewarded for and they will game any system to get its rewards. Under the present dispensation fining the Banks for bad behaviour has simply become a cost of doing business. Until and unless individual managers are personally visited for the consequences of their behaviour things will go on as before. At the moment we have a situation where top management create a compelling context in which their juniors behave badly, then deny all knowledge. If they ask well as their juniors went to jail their behaviour would change. Transparency and consequences are what we need here.

  2. Michael Hochgürtel

    What about state owned banks ?

  3. Lachlan Riddell

    A very interesting read, it is an obvious conclusion that we need more personal accountability from bank failures. Two areas need to be further debated, however. Firstly, what happens when every single bank CEO in a country shuts up shop & stops lending at the first sign of a turn in the credit cycle? Are we prepared to accept that any shift towards personal accountability is likely to make banking even more pro-cyclical? Second, for how long into the future should we judge a CEO’s actions? If a loan goes sour after 7 years, but, say, 6 years after a CEO’s departure – is that loss still his responsibility? What is the right statute of limitations for a bank CEO? Lastly, I disagree with your point about protecting shareholders because they have no direct responsibility for bank failure: they need to understand that they absolutely DO have the responsibility of holding management accountable (its their capital) and there are plenty of ways to do so (eg AGM’s).