A new eReport: Excessive risk-taking by banks

Richard Baldwin 30 March 2012



For many, the global crisis was caused by the interlinked fragilities that arose in the banking and financial sectors; these themselves were created by mindless deregulation and permissive monetary policy. By the late 2000s, the system was so precarious that shocks from many directions could have triggered the economic conflagration we witnessed.

The actual trigger was the bursting in fall 2007 of the US housing bubble – a bubble that was created by banks’ excessive risk-taking. The hallmark of this risk-taking was the subprime mortgage – for example the infamous loans to NINJA customers (No Income No Job or Assets). But such excessive risk-taking by banks is far from rare. Ireland went bust in 2010 due to risky loans made by national banks; Iceland suffered a similar fate in 2008, and Spain is flirting with a comparable outcome due to excessive risk-taking in property lending by its banks.

CEPR eReport on bank risk-taking

A new eReport released today by CEPR – edited by Mathias Dewatripont and Xavier Freixas – explores the origins of excessive risk-taking by banks. The book comprises four substantial chapters (in addition to the introduction that nicely summarises these four and puts the analysis into a broader context).

  • Corporate governance, by Hamid Mehran, Alan Morrison, and Joel Shapiro

In principle, banks do what their managers decide and managers, in turn, are controlled by a board of directors. Excessive risk-taking must therefore involve a breakdown in control, or desire on the part of the board to encourage such activity. For example, strategic decisions by managers may be motived by consideration of their own bonuses, short-term stock price movements, or shareholders’ short-run interests (rather than stakeholders’ long-run ones). This line of reasoning directs attention to the structure of financial institutions’ corporate governance as one source of excessively risky behaviour.

Mehran, Morrison, and Shapiro argue that corporate governance may be especially weak due to the multiplicity of stakeholders (insured and uninsured depositors, the deposit insurance company, bond holders, subordinate debt-holders and hybrid securities holders), and the complexity of banks’ operations. Moreover the moral hazard created by the too-big-to-fail situation may have led boards to encourage risk-taking as they knew that big losses would be paid largely by taxpayers rather than stakeholders. The authors also look at banks’ executive compensation schemes and the composition of boards.

  • Procyclicality, by Rafael Repullo and Jesus Saurina

The boom-bust cycles in banking are at least in part caused by the procyclical availability of cheap funding and capital. In boom-times, funding is cheap and easy to get; in bust-times it is dear and scarce. This obviously procyclical nature of bank lending in the years before, during, and after the crisis has produced a consensus that banks should face anticyclical capital buffers to both reduce the size of the next boom and mitigate the damage during the next bust. The authors focus on one aspect of this, namely the question of whether and how much additional capital should be required during excessive credit growth phases, and how these excessive credit growth phases are to be identified. They study how the Basel III regulatory framework proposes to tackle the issue and the extent to which the rules accomplish their objectives.

The Basel III countercyclical provisions require higher capital-to-loan ratios when the credit-to-GDP ratio deviates from its trend. Their analysis, however, shows this works the wrong way for a majority of nations; the deviations are negatively correlated with GDP growth. In short, banks that follow the deviation from trend rule may actually be pursuing a procyclical rather than a countercyclical capital policy. The authors propose a simpler rule – the credit growth rate.


  • Disclosure, transparency and market discipline, by Xavier Freixas and Christian Laux

Prior to the crisis, market discipline was thought to be the perfect complement to supervision – channelling funds to sound institutions while penalising excessive risk-takers. The crisis has changed that view; most regulators and academics now see market discipline as a weak force. The authors of this chapter consider various theoretical aspects of how imperfections could gum up the information transmission – the key ingredient of market discipline. In addition to systemic problems, the situation worsens during a crisis because both firms and issuers have incentives to hide bad information.

The market’s main sources of information are firms’ financial reports and credit rating agencies and the authors address a number of reproaches levelled at both. On the financial reporting, the use of fair value analysis has come in for strong criticisms as it caused firms to write down asset falls as the markets collapsed with this leading to eroded capital and heightened uncertainty. The authors however argue that fair value is not much to blame as it only affects banks’ trading portfolios and there is substantial discretion for banks to suspend it if the losses are considered temporary. They are more critical when it comes to credit rating agencies, concluding that these profit-maximising firms are in an institutional setting that inadequately deals with conflicts of interests. They call for more regulation of credit rating agencies to redress this.

  • The banking resolution regime, by Xavier Freixas and Mathias Dewatripont

The last chapter addresses systems for taking banks into bankruptcy since beliefs about what happens when all goes wrong do affect risk-taking. If distressed bank are bailed out, risk-taking is never too risky for the bank. Banks are bailed out to avoid the high social costs of such failures. The first objective of regulation is therefore to reduce the cost of bankruptcies; this is the main focus of the last chapter.

Banking resolution, according to the authors, should be thought of as a bargaining game between shareholders and regulators. Shareholders want to maximise the value of their shares while regulatory authorities’ main objective is to preserve financial stability at the lowest possible cost. Given this, time plays against the regulatory authority. The authors thus argue for bankruptcy rules that are specially crafted for the banking sector (and different from those applying to non-financial corporations).

The authors also argue that time is of the essence, even with the perfectly efficient bankruptcy procedure. Banks in distress should be quickly closed or quickly bailed out. The chapter’s examination of banking crises in different countries shows great variety in the procedures followed and conclude that theory has no clear-cut recommendations to offer.

Plainly the design of the bank resolution mechanisms is critical. One proposal is to add a layer of capital to prevent future crises, but the authors defend the possibilities opened by contingent capital (like contingent convertibles and capital insurance). They argue that these types of mechanisms would preserve the best characteristics of debt and therefore limit moral hazard. The authors conclude by considering cross-country resolution and the challenges it implies and discuss the recent changes in the European banking resolution framework.


Dewatripont, Mathias and Xavier Freixas, eds (2012), The Crisis Aftermath: New Regulatory Paradigms, London: Centre for Economic Policy Research. 



Topics:  Global crisis Global economy Microeconomic regulation

Tags:  risk-taking, cross-border banking, macroprudential regulation


This book will help frame the debate about the role and the degree of regulation of banks in the aftermath of the sub-prime crisis which rolled into the Global Financial Crisis. I certainly agree that a bank's governance structure is vitally important in safeguarding the bank against errant internal behavior and external shocks. But banks all operate within a regulatory context which they game by opting in, and neutralize by opting out. And where the bank governance activity has become a box-ticking exercise and the tools employed focus on protecting the firm against past crises instead of being deployed to spot early warning signs of new problems well before a crisis emerges, then you have a systemic governance problem. That is compounded by conflicted external advisers who reap short term benefit by saying yes to everything and internally by the group think that sometimes prevails at board meetings, the absence of a normative framework (both ethical standards but also how things are done effectively in other countries), the absence of institutionalized  devil's advocates (e.g. through the audit committee of the board) and the fact that value-added information presented to the Board is often not focused and powerful enough so that legitimate challenges to current directions are not accepted as part of building company value.
At the macro level, where we are considering all banks vis-a-vis regulators, on the one hand, there is a negative synergy between the market and regulators (a combination of exploitive rent-seeking, territoriality, information control, fear of failure or perhaps even fear of lost employment opportunities with the regulated entities etc) where the regulations are badly designed  or absent and the lending culture is not conservative and, on the other hand, a positive synergy (intelligent limit testing, mutual rewards, system level protection (the commons)  etc) that arises when a conservative lending culture is supported interactively by a conservative regulatory regime.
For this reason, I was surprised that the book did not reference the success of the Canadian and Australian conservative bank regulatory regimes and their conservative banking cultures. The issues involved include: the level of Tier 1 capital before the GFC, sub-prime mortgages as a percentage of total mortgages, single bank regulator, presence or absence of universal banks, the extent of political interference in the housing market, the system favours borrowers over lenders or vice versa, the on-versus-off balance sheet issue for mortgages, mortgage interest deductibility, the ability to walk away from mortgage debt, LTV ratios etc etc.
The key point is that rule drivers (government) and market drivers (market players) are continuously shaping their respective behaviours. Outcomes are determined by the degree to which consensus provides value to both sides. The gamble is that rule drivers and market drivers are focused on creating long term value for the system. In the US, the short term consensus was to raise the homeownership rate by 5 percentage points. The cost was several trillion dollars. The homeownership rate is back to the low 60s where it was 10 years ago.
In the conservative regulatory regimes with conservative lending practices, this catastrophic failure did not arise, although the blowback from the US, UK and Euro-zone problems certainly created a drag on performance in Australia and Canada. Now banks in those countries are usually at the top of everyone's list for being good performers. Those banks and regulators are the real beacons of light in this debate.They kept on doing what they normally do and they did it well. They deserve a chapter at least.There is nothing wrong with experimenting with the banking system if it is done with intergenerational risk in mind, but ad hoc experimentation without fear of consequence leaves a door wide open for foolish behaviour by banks whose corporate governance antennae are potentially anesthetized  by internal group dynamics and incentive structures. The market driver behaviour and rule driver behaviour (and these two factors I have discussed are only a subset of eight key factors) that led to the GFC was mutually reinforcing. The control groups' (Australia and Canada) experience is that from an empirical perspective the sub-prime crisis and GFC did not need to happen, and from a normative perspective it should not have happened. Future regulatory experimentation need to take these facts into consideration.
These comments are my own and do not represent those of my firm.

Professor of International Economics, Graduate Institute, Geneva; Director of CEPR; Vox Editor-in-Chief

Vox Talks