VoxEU Column Financial Markets

Who should decide on emergency liquidity assistance?

What government agency should decide lender-of-last-resort policy? This column discusses the optimal allocation of decision-making authority, suggesting that the central bank decide emergency loans and the deposit insurance agency guarantee them. But providing greater liquidity assistance will also require punishment to deter moral hazard problems.

Many countries are revising their institutions to deal with troubled banks. In the UK, the Labour Party believes that the current arrangement – the Tripartite Standing Committee constituted by the HM Treasury, the Bank of England, and the Financial Services Authority – is the best framework for regulating and supervising financial institutions and wants to strengthen it. On the other hand, the Conservative Party has adopted as policy the abolition of the Tripartite system and the investiture of full responsibility to the Bank of England for the prudential regulation of all financial institutions, as was the case prior to the Financial Services and Markets Act of 2000.

The failure of the Tripartite system

When the crisis broke, the run on Northern Rock – the first run on a British bank in approximately 130 years – exposed that the Tripartite was a slow decision-making mechanism. Indeed, the Northern Rock bank run on the early morning of 14 September 2007 was an event in the aftermath of the liquidity crisis at the bank, rather than the event that triggered its liquidity problems. Northern Rock had been suffering from liquidity shortfalls since financial and interbank markets dried up in July 2007.

The members of the Tripartite faced inappropriate, somewhat conflicting, incentives that materialised in different opinions about a potential bank failure and delayed a final decision. The Financial Services Authority, which has supervisory responsibilities, strongly supported the illiquid bank. According to Boot and Thakor (1993), such an action would have been undertaken in order to protect the supervisor's reputation. Meanwhile, the Bank of England, which cares about systemic financial stability, refused to provide emergency assistance until the problem became systemic, i.e. until depositors ran on the bank. Moreover, there was a lack of coordination and information sharing between the Financial Services Authority and the Bank of England since the latter seems to have been unaware of the problems at Northern Rock until August 2007. According to Kane (1989, 1990), a bank supervisor who is interested in its long-run career earnings would have an incentive to delay disclosure of problems by following a “not in my watch” policy. Finally, the Memorandum of Understanding underpinning the Tripartite arrangement did not set out with sufficient clarity the duties and responsibilities of each agency.

The allocation of decision-making authority and its policy implications

The concept of lender of last resort goes back to Thornton (1802) and Bagehot (1873), who assert that the role of a lender of last resort is to provide emergency liquidity assistance to illiquid but solvent banks, against good collateral and at a penalty rate.1 But, who should decide on emergency liquidity assistance when several self-interested bank regulators could make such a decision?

A recent strand of the literature on the institutional allocation of bank regulatory powers aims to answer that question, and thereby to inform the current policy debate, particularly in the UK. For example, Repullo (2000) and Kahn and Santos (2005) study the optimal allocation of decision-making authority in a framework in which two self-interested regulators (i.e., a central banker and a deposit insurer) may decide whether or not to provide emergency liquidity assistance to a troubled bank. I extend that framework in two directions (Ponce 2009). First, the policymaker may instruct the unconditional provision of emergency liquidity loans, in addition to giving one of the two regulators the authority to decide whether or not to provide them. Second, bankers may play an active role manipulating the magnitude of their banks' liquidity shortfalls.

A first result is that the central banker should be responsible for conducting the lender-of-last-resort policy when the liquidity shortfall of the bank is small enough. In that case, the deposit insurer requires a level of bank’s solvency in order to provide a last-resort loan that is too large with respect to the first-best policy. Hence, it is highly probable that the deposit insurer refuses assistance to too many illiquid but still solvent banks. However, the central banker is softer than the deposit insurer, and its lender-of-last-resort policy is the closest to the first-best policy. If the size of the bank’s liquidity problem is large enough, then both the central banker and the deposit insurer are too tough. In that case, it is socially desirable to avoid too many bank closures through the provision of emergency loans that are not contingent on the level of solvency of the bank.2

An alternative way to implement this policy is for the central banker to always decide whether to provide an emergency loan but to require the deposit insurance agency to guarantee emergency loans over the liquidity shortfall that ought to trigger unconditional bailouts. This arrangement has some advantages. First, it reduces taxpayers’ exposure and makes transparent the way in which the costs of the policy are borne because the deposit insurance scheme is financed by the banking industry. Second, the deposit insurer has incentives to enhance the quality of the deposit insurance scheme; since the deposit insurance agency is economically liable for large emergency loans, the deposit insurer will attempt to avoid the occurrence of large shortfalls by improving depositors’ protection.

Similar reasoning suggests that the deposit insurer should conduct banking supervision (i.e., the responsibilities of gathering private information from banks and enforcing a set of regulatory rules and policies to them) because she will have an incentive to act promptly in order to avoid the occurrence of large liquidity shortfalls. However, this allocation may damage the capacity of the central banker to conduct the lender-of-last-resort policy because the deposit insurer may prefer not to truthfully reveal supervisory information to the central banker. Consequently, the central banker should have the authority to gather private information from banks in order to meet his responsibilities as lender of last resort.

Establishing these arrangements ex ante gives the banker incentives to increase the likely size of her bank’s liquidity shortfall, so as to increase the likelihood that it receives a bailout. Hence, the (second-best) optimal policy combines the optimal bailout with punishments for the banker when her bank’s shortfall is large enough for bailouts to be unconditional. Since bank regulators may find it hard to punish faulty bankers, an unconditional bailout by the central bank should trigger (in the spirit of the US’ FDICIA) the application of mandatory, punitive actions against the bank that receives it. Moreover, the severity of these actions should increase as the quality of the bank's assets decreases.

Disclaimer: The views expressed herein are those of the author and do not necessarily represent the views of the institutions to which he is affiliated.


1 More recently, Goodfriend and King (1988) (see also Bordo, 1990, and Kaufman, 1991) argued that, with sophisticated interbank markets, only aggregate liquidity provision (i.e., monetary policy) is necessary but individual interventions are not anymore. However, investors in the interbank market may face a “coordination failure” which implies that some solvent banks would not find liquidity in it. Indeed, it was the case during the hardest days of the current crisis. Freixas et al. (2004) and Rochet and Vives (2004) argued that public intervention in the form of an emergency liquidity assistance has its rationale in this market failure. See, Freixas et al. (1999), Freixas and Parigi (2008) and Santos (2006) for reviews of the literature on the lender of last resort.

2 If we assume that a large liquidity shortfall in an individual bank is caused by a “sudden financial arrest”, then this result provides another rationale for Caballero’s (2009a, b) proposal of using massive provision of credible public insurance as an effective “financial defibrillator” mechanism.


Bagehot, W. (1873), Lombard Street. A description of the money market. New York: Scribner, Armstrong & Co.

Boot, A.W.A. and Thakor, A.V. (1993), “Self-interested bank regulation”, The American Economic Review 83 (2), 206-212.

Bordo, M. (1990), “The lender of last resort: Alternative views and historical experience”, Federal Reserve Bank of Richmond Economic Review, 18-29.

Caballero, R. (2009a), “Sudden financial arrest”, paper presented at the 10th Jacques Polak Annual Research Conference at the International Monetary Fund.

Caballero, R. (2009b), “Sudden financial arrest”, VoxEU.org, 17 November.

Freixas, X., Giannini, C., Hoggarth, G. and Soussa, F. (1999), “Lender of last resort: A review of the literature”, Financial Stability Review 7, 151-167.

Freixas, X., Parigi, B. and Rochet, J.-C. (2004), “The lender of last resort: A 21st century approach”, Journal of the European Economic Association 6 (2), 1085-1115.

Freixas, X. and Parigi, B.M. (2008), “Lender of last resort and bank closure policy”, CESIFO Working Paper No 2286.

Goodfriend, M. and King, R.G. (1988), “Financial deregulation, monetary policy, and central banking”, in Haraf, W. and Kushmeider, R. (eds.), Restructuring Banking and Financial Services in America. AEI Studies, n.481, Lanham, MD: UPA.

Kahn, C.M. and Santos, J.A. (2005), “Allocating bank regulatory powers: Lender of last resort, deposit insurance and supervision”, European Economic Review 49, 2107-2136.

Kane, E.J. (1989), “Changing incentives facing financial services regulators”, Journal of Financial Services Research 2, 265-274.

Kane, E.J. (1990), “Principal-agent problems in the S&L salvage”, Journal of Finance 45 (3), 755-764.

Kaufman, G.G. (1991), “Lender of last resort: A contemporary perspective”, Journal of Financial Services Research 5, 95-110.

Ponce, J. (2009), “Lender of last resort policy: What reforms are necessary?Journal of Financial Intermediation.

Repullo, R. (2000), “Who should act as lender of last resort? An incomplete contracts model”, Journal of Money, Credit and Banking 32 (3), 580-605.

Rochet, J.-C. and Vives, X. (2004), “Coordination failures and the lender of last resort: Was Bagehot right after all? Journal of the European Economic Association 6(2), 1116-1147.

Santos, J.A. (2006), “Insuring banks against liquidity shocks: the role of deposit insurance and lending of last resort”, Journal of Economic Surveys 20(3), 459-482.

Thornton, J. (1802), An enquiry into the nature and effects of paper credit of Great Britain. London: Hatchard.

315 Reads