Central Bank Tools and Liquidity Shortages
Stephen G. Cecchetti and Piti Disyatat*
9 February 2009
Abstract
The current financial crisis has brought into sharper relief the fundamental role of central banks as lender of last resort and raises the question of whether the tools they have available are sufficient for confronting the challenges posed by modern liquidity crises. This paper attempts to provide a practical conceptual overview of these issues highlighting, in particular, the need to distinguish between different types of liquidity shortages before any general set of principles can be established with respect to how the lender of last resort function should be conducted.
JEL Classification: E50, E52, E58, E60
Keywords: Monetary policy, lender of last resort, liquidity operation, systemic crisis, liquidity shortage
* Cecchetti is Economic Adviser and Head of the Monetary and Economic Department, Bank for International Settlements; Research Associate, National Bureau of Economic Research; and Research Fellow, Centre for Economic Policy Research. Disyatat is Senior Economist, Monetary Policy and Exchange Rate Unit, Bank for International Settlements. This paper was prepared for the Federal Reserve Bank of New York Conference on “Central Bank Liquidity Tools,” 19-20 February 2009. We are grateful to Claudio Borio, François-Louis Michaud, and Christian Upper for comments. All remaining errors are ours. Views expressed are those of the authors and not necessarily those of the BIS.
1. Introduction
The global financial crisis that began in mid-2007 has renewed concerns about financial instability and focused attention on the fundamental role of central banks in preventing and managing systemic crisis. In response to the turmoil, central banks have made extensive use of both new and existing tools for supplying central bank money to financial institutions and markets. Against this backdrop, there has been intense interest in the implications that recent financial developments may have on the fundamental nature of central banks’ lender of last resort (LOLR) function and whether the traditional tools that have been at policymakers’ disposal remain adequate in the face of modern liquidity crises. This paper attempts to address these issues and in doing so, provides a perspective of recent central bank liquidity operations that is tied more closely to their underlying purpose from a LOLR perspective.
We begin by defining three types of liquidity shortages that central banks may need to address. By doing this, we emphasize the fact that the conditions under which central bank liquidity – reserves or central bank money – are made available should and do differ depending on the underlying nature of the crisis officials are trying to mitigate. Importantly, this means that there may not be a single set of principles for central banks’ LOLR function. Rather, actions of the central bank in its capacity as the lender of last resort will depend on the nature of the crisis it faces. Recognizing this goes some way toward reconciling the debate surrounding the appropriate role of a LOLR.1 Following the definitions, in Section 3 we proceed with a discussion of the tools that central banks have at their disposal and how they might be tailored to address each type of liquidity shortage. Section 4 provides a brief description of how recent actions by major central banks can be interpreted from this perspective and the last section concludes. It should be noted at the outset that the focus of this paper is on central bank liquidity operations and not on their interest rate responses.
2. Liquidity Shortages and the Lender of Last Resort
Apart from the conduct of monetary policy, a vital responsibility of central banks in most countries is to perform the role of LOLR. At its core, the LOLR function is to prevent and mitigate financial instability through the provision of liquidity support either to markets or individual financial institutions. The underlying premise is that shortages of liquidity, by which we mean the inability of an institution to acquire cash or means of payment at low cost, can lead to otherwise preventable failures of institutions that then result in spillover and contagion effects that may ultimately engulf the financial system more broadly with significant implications on the real economy.2 By signaling its willingness and ability to act decisively, the central bank’s actions are intended to restore confidence in the system by avoiding fire sales of assets and supporting market functioning.
The “classical” doctrine on the LOLR as attributed to Thornton (1802) and Bagehot (1873) is commonly interpreted to imply that such lending should be extended freely without limit, but only to solvent institutions at penal rates and against good collateral (for example, see Rochet and Vives, 2004). This set of principles has been subject to substantial debate for much of the last 30 years with many issues as yet to be resolved.3
1 We do not enter into the debate over whether the LOLR takes the place of a deposit insurance system. Recent events, especially the retail bank runs that accompanied the nationalization of Northern Rock in the UK, would appear to have settled the matter in favour of the importance of rule-based deposit insurance system.
2 This definition of LOLR is quite broad and can, in principle, encompass any injection of central bank liquidity, including routine ones. That said, the focus will be more on extraordinary liquidity provisions that are driven by unanticipated events.
3 See Davis (2008) and Rochet (2008) for a detailed exposition of the various views.
1
At the most basic level, the underlying principles of Bagehot’s original dictum have been subject to various interpretations. Goodhart (1999), for example, emphasized that Bagehot’s criteria for lending was not conditioned on the individual borrower, but on the availability of good collateral. As such, the distinction between illiquidity from insolvency was never an important issue. Similarly, while the imposition of a penal rate has traditionally been judged relative to the prevailing market rate, it can be argued that Bagehot only advocated that lending take place at a rate higher than the pre-crisis level (Goodhart, 1999). Following the logic that the LOLR should try to achieve the good – panic-free – equilibrium, the penalty ought to be relative to the interest rate during normal times rather than relative to the surely higher rate at which institutions would lend to each other in the market during a panic. Indeed, in practice LOLR lending has frequently taken place at prevailing market rates (Giannini, 1999).
At a more practical level, the distinction between illiquidity and insolvency has been largely dismissed on the grounds that banks generally face illiquidity when solvency is in question (Goodhart and Schoenmaker, 1995). Indeed, an individual bank will only seek assistance from the authorities when it cannot meet its liquidity needs in financial markets. Since the wholesale interbank money market is the first stop for most banks, this almost certainly means that there are significant doubts about the institution’s ultimate solvency. The proposition that central banks only lend against good collateral is also undermined by the fact that a bank that is unable to raise funds in the market must, almost by definition, lack access to good security for collateralized loans. As such, emergency lending assistance from the central bank will likely be against collateral of questionable quality. In addition, the imposition of a penal rate has been criticized on grounds that such a policy could compound the problem if it imposes a substantial burden on the troubled institution.
At the same time, another facet of the debate has focused on the appropriate way in which the LOLR should be implemented. Some argue that in an advanced financial system, LOLR should be exclusively through open market operations. Insulating the real economy from financial sector disturbances means keeping interest rates near their target. Open market operations that meet system-wide changes in demand for reserves do this, after which the market can direct reserves to those most in need, thereby avoiding the mispricing that administrative mechanisms might create (Schwartz, 1992; Kaufman 1991; Goodfriend and King, 1988). Such an approach was clearly successful in the case of operations associated with the spikes in liquidity demand during the Y2K episode and in the aftermath of the stock market crash of October 1987, for example. However, others argue that LOLR may require direct lending, not open market operations, as the market may fail to deliver liquidity to banks in distress whose failure threatens the financial system (Rochet and Vives, 2004; Freixas et al., 2000a, 2000b; Goodhart, 1999).
2.1. Three Kinds of Liquidity Shortages
Rather than getting bogged down in the theoretical debate on the design and role of the LOLR, we take a more pragmatic approach and outline the broad conditions under which central banks’ provision of liquidity is undertaken in practice. From this we derive some general principles that apply depending on the specific situation. Indeed, once it is recognized that the nature of the LOLR differs across circumstances, many of the issues that have been at the center of the theoretical debate fade.
It will be useful at the outset to distinguish between three types of liquidity: central bank liquidity, market liquidity, and funding liquidity. Central bank liquidity is deposits of financial institution at the central bank – often known as reserve or settlement balances. These reserve balances are held by financial institutions to meet reserve requirements, if any, and to achieve final settlement of all financial transactions in the payments system. Individual institutions can
đang được dịch, vui lòng đợi..