Confusion about the lender of last resort

Ulrich Bindseil, Luc Laeven

13 January 2017



In response to severe liquidity squeezes and market malfunctions during the Global Crisis, major central banks around the world, including the ECB, engaged in market operations that injected significant amounts of liquidity in banks and asset markets. This ensured their proper functioning and, with it, the transmission of monetary policy.

Central bank operations like this are not new or unprecedented. Central banks were originally created to provide credit to other banks in case of financial instability and liquidity squeezes (Bignon et al. Ugolini 2012, Calomiris et al.  2016). Central bank lender of last resort (LOLR) measures respond to moments when solvent individual banks, or entire banking systems, are unable to maintain depositor confidence (Garcia-de-Andoain et al. Manganelli 2016). They allow the banking system to continue to provide transaction services through the payment system, and to provide credit to bank-dependent borrowers. The Global Crisis prompted major central banks to conduct LOLR operations on a large scale, and reinvigorated the debate on the constraints and boundaries of LOLR policies.

Against this background, some have criticised the collateral framework of major central banks. For instance, Nyborg (2016) criticises the Eurosystem collateral framework, arguing that central bank interventions are based on terms determined by the central bank instead of the market. He claims that this approach undermines market discipline. Nyborg also argues that these policies encourage banks to offload illiquid assets to the central bank in a time of crisis, and that “favouring illiquid collateral in the collateral framework may then lead to an overproduction of illiquid real assets” (pp 20-21).  This criticism is misguided, and misses the bigger picture of the role of the lender of last resort.

First, central banks are never liquidity-constrained and can never default in the currency they issue. This is key to understanding the central bank’s LOLR role. Not being liquidity-constrained implies that liquidity risk is priced differently by central banks, particularly in a liquidity crisis when the price of liquidity skyrockets. From the perspective of its borrowers,  never defaulting on obligations in the domestic currency gives the central bank a risk-free counterparty status, and makes borrowers accept deep haircuts when providing collateral. In contrast, in a repo transaction between two similarly risky parties, setting the haircut level is a zero-sum game. Deep haircuts protect the cash provider, but equally expose the collateral provider to credit risk (Ewerhart and Tapking 2008, Bindseil et al. 2016).

Second, the LOLR should lend against good collateral and at penalty rates relative to normal times. This is the famous 'Bagehot principle' (1873). It is often misunderstood as implying that the central bank should lend at terms that are less favourable than the market, even in a crisis. In fact, the opposite is true. Loan terms should be less favourable compared to normal times, but precisely because of market malfunction during a crisis they should be offered at below-market rates during a crisis. Mirroring such pricing, central bank collateral frameworks tend to apply haircuts on collateral that are more stable through the financial cycle than those observed in the market. In this way the central bank reduces the procyclicality of the financial system (Kiyotaki and Moore 1997).

Banks are, by nature, particularly susceptible to liquidity risk. They rely on risk-intolerant money market funding, and the opacity of bank lending implies that the market has only imperfect information on the quality of the bank’s assets, and therefore the ability of the bank to meet its obligations. While capital and liquidity regulation can contribute to improvement in the funding stability of banks, vulnerabilities will remain. The market freezes that result can give rise to substantial disruptions of the payments system and the supply of bank credit. LOLR interventions can avoid disruptions to payments and credit intermediation that result from liquidity risk. As a result, they contribute to an efficient financial system that supports growth in the real economy.

The LOLR function has risks as well as benefits. First, there is the risk of lending to institutions that turn out to be insolvent. In this case, the run would not be stopped, and the loss-given default of the remaining creditors would have increased. Therefore, solvency, conditional on liquidity, always must remain a key requirement for any bank that benefits from the LOLR.  Second, there is the risk that LOLR measures lead to financial losses for the central bank. Therefore, all lending must be adequately collateralised, and sufficiently deep haircuts need to apply. Third, there is the risk that LOLR will reduce incentives for banks to make their own provisions against liquidity risks, relying on the LOLR to provide this insurance. These risks can be mitigated by liquidity regulation, by associating disincentives (penalty rates, additional reporting duties or central bank monitoring) to LOLR credit, and by requiring adequate collateral with sufficiently deep haircuts. The optimal collateral framework must consider these trade-offs to ensure that it is not overly generous (Bindseil 2016).

The large scale and scope of central bank LOLR operations during the global financial crisis helped to prevent a meltdown of the global financial system. While it has long been recognised that central bank liquidity support is unavoidable in certain situations, the intensity of LOLR operations during the crisis raised concerns that central banks were taking excessive risks, and that they were encouraging moral hazard. Nevertheless, despite the severity of the crisis and the scale and scope of LOLR operations, central banks have not made any financial losses. This illustrates the powerful economic logic underlying the LOLR, and shows that central banks applied this logic with prudence. Liquidity regulation, combined with an adequate central bank operational framework, can discourage excessive ex ante reliance on the LOLR, and can guard against moral hazard.

Authors’ note: The views expressed here are our own and should not be interpreted to reflect those of the ECB or the Eurosystem.


Bagehot, W (1873), Lombard Street: A Description of the Money Market, Richard D. Irwin, Homewood, IL (reprinted in 1962).

Bignon, V, M Flandreau, and S Ugolini (2012), “Bagehot for beginners: the making of lender of last resort operations in the mid-nineteenth century”, Economic History Review, 65 (2), pp. 580–608.

Bindseil, U (2016), “Evaluating monetary policy operational frameworks”, prepared for Jackson Hole conference organised by the Kansas Federal Reserve Bank.

Bindseil, U, M Corsi, B Sahel and A Visser (2016), “The Eurosystem collateral framework: a review of recent comments”, forthcoming.

Calomiris, C W, M Flandreau, and L Laeven (2016), “Political foundations of the lender of last resort: A global historical narrative”, Journal of Financial Intermediation, forthcoming. Also available as CEPR DP No. 11448

Ewerhart, C, and J Tapking (2008), “Repo markets, counterparty risk, and the 2007/2008 liquidity crisis”, ECB Working Paper No. 909.

Garcia-de-Andoain, C, F Heider, M Hoerova and S Manganelli (2016), “Lending-of-last-resort is as lending-of-last-resort-does: Central bank liquidity provision and interbank market functioning in the euro area”, Journal of Financial Intermediation, 28, pp. 32-47.

Kiyotaki, N and J Moore (1997),”Credit cycles”, Journal of Political Economy, 105, pp.  211–248.

Nyborg, K G (2016), Collateral Frameworks: The Open Secret of Central Banks, Cambridge University Press.



Topics:  Global crisis

Tags:  lender of last resort, LOLR, liquidity, central bank

Director General of Market Operations, ECB

Director-General of the Directorate General Research, European Central Bank and CEPR Research Fellow