Bank run

A bank run expresses a sudden loss of confidence from depositors towards their bank's ability to provide liquidity and redeem its customers; it results in a massive and sudden withdrawal of bank deposits and a contraction of available cash stock, which may lead to bankruptcy and a banking crisis. It results from depositors' fear of not being able to recover their funds. It may involve one or more banks, while banking panic involves several institutions and implies a contagion phenomenon (De Bandt and Hartmann, 2000, p. 263).

Bank runs usually raise questions about the mechanisms of financial crises and the solutions to be implemented. These explanations deal with banks' internal and external factors (Bordo, 2000, p. 110), such as the structure of banks' balance sheets, information asymmetries, deposit contracts and the existence of mimetic behaviour (Boyer et al., 2004, p. 102). Thus, a bank run is often discussed in parallel with bank liquidity, banking crises (concerning banks' assets) and systemic crises. It may be considered as the origin of these crises and a result of financial dysfunctions requiring central banks' interven- tion. Seemingly a simple problem of illiquidity, it may become more problematic when it involves a bank failure contagion effect (cascade weakening) and a domino effect (cas- caded bank bankruptcies).

Financial economists have been studying the occurrence of bank runs, albeit only from a microeconomic point of view. Diamond and Dybvig (1983) build their financial theory explicitly on bank runs to justify the " raison d'être" of banks. Banks exist because they solve the issue of customers' liquidity; then they expose themselves to illiquidity and to banking panics, because contracts (deposits) can be challenged and then panic may settle following a self-fulfilling prophecy.

Most of the relevant microeconomic models (single bank or multibank models) attempt to explain the nature of initial shocks (see Gorton, 1985, regarding informa- tional shocks; Mishkin, 1991 and Flannery, 1996, regarding adverse selection shocks), of concerned agents (through the interbank market, see Rochet and Tirole, 1996, or the payment system), and of information involved (see Bhattacharya and Jacklin, 1988). Nevertheless, they hardly explain systemic crises.

The complexity of bank runs is at last considered in some macroeconomic models (Minsky, 1996; Rochet, 2010). It actually involves financial stability and monetary policy, as a part of a broader and systemic phenomenon. This approach pertains to the long- standing tradition of the lender of last resort (LLR). It has showed its relevance again several times since 2008 with Northern Rock or in Cyprus.

Generally speaking, there are four lines of action in a case of panic (Bordo, 2000, p. 111). For Thornton and Bagehot, the LLR provides the necessary cash to solvent banks, it restores confidence by lending cash at a penalty rate of interest and it announces its interventions. According to Goodfriend and King (1988), the LLR intervenes only by increasing the monetary base. Goodhart and Huang (2005) note, however, that the dis- tinction between banks' insolvency and their illiquidity is difficult to establish and that it is also worth supporting insolvent banks. Finally, for the free-banking school, the market is the only correct answer in a case of panic.

Traditionally, interventions consist in guaranteeing bank deposits and suspending convertibility. These purely microprudential regulatory perspectives have been widely adopted and have shown their limits. Thus, a macroeconomic approach and an advanced theory of money and bank are necessary. As pointed out by De Carvalho (2009, p. 278), the "evidence from the current crisis so far seems to confirm the Post Keynesian view. There was no large shock impacting the financial system." Central banks are more than ever called upon to provide liquidity both for illiquidity issues and insolvency issues for individual banks and the banking sector as a whole. In order to prevent bank runs and to stabilize the financial system, it seems necessary to control leverage effects, to extend regulation to other financial institutions, to get back to regulation through self-discipline, and finally to regulate financial innovations (De Carvalho, 2009).

See also:

Asymmetric information; Bagehot, Walter; Bank deposits; Cash; Contagion; Financial crisis; Financial innovation; Financial instability; Free banking; Lender of last resort; Sudden stops; Thornton, Henry.