There is a broad consensus that the intraday interest rate should be set to zero on efficiency grounds. In this work we document that, while in normal times money market rates are roughly in line with this efficiency criterion, they may deviate by a large extent in a situation of liquidity tension, like the one taking place at the outset of the subprime financial turmoil. We provide an analysis of the European electronic interbank market (e-MID) with high frequency data, showing that the hourly interest rate ? implicitly defined by the intraday pattern of the overnight rate ? jumped by more than ten times (from 0.2 bp to 2.2 bp) in the reserve maintenance period starting on August 8th 2007. This finding has no straight foward explanation, since the Eurosystem supplies intraday liquidity at no cost and without limit, except for the collateral requirement.
We attribute this result to the sudden increase of the liquidity and credit risks taking place at the outset of the financial turmoil, with two likely consequences. First, in times of liquidity crisis the intraday credit provided by the central bank is an imperfect substitute for an early delivery of overnight funds. Then the market price of intraday liquidity incorporates a liquidity premium, making it deviate from the cost of daylight central bank overdrafts. Second, the widening of the spread between unsecured and secured interbank interest rates implies an increase of the cost of collateral, making it more costly to get intraday credit from the central bank. A stronger demand of collateral, in order to guarantee a larger amount of available funds from the Eurosystem, has presumably contributed to making the collateral requirement more costly.
Several recent contributions in monetary theory, focussing on the role of money as a medium of exchange, point to the optimality of a zero intraday rate. For example, Zhou (2000) distinguishes between "consumption/investment debt" and "settlement debt": since the latter does not affect the inter temporal allocation of resources, the intraday rate is just a transaction cost which should be minimized. In Martin (2004) and Bhattacharya et al. (2007) a zero level of the intraday rate is optimal since it provides an insurance for consumers against liquidity shocks. Other works, more focussed on the mechanics of the payment systems, stress that a positive cost of intraday liquidity may induce individual banks to delay payments, putting a negative externality onto the banking system (see Angelini 1998, Bech and Garratt 2003, Mills and Nesmith 2008, Martin and McAndrews 2008). The issue of "delayed payments" have raised the concern of policymakers for its impact on the operational risk in the payment systems (see FED 2006, 2007).
Overall, this literature shows that the market equilibrium may be inefficient, since a positive cost of intraday credit may emerge as a market outcome. This creates a role for the active intervention of central banks, who are able to provide intraday liquidity at no cost. Our work contributes to this literature by showing that during a liquidity crisis the ability of the central bank to curb the (implicit) market price of intraday liquidity may be limited.
While the theory of the intraday interest rate is well developed, the empirical evidence is scarce, due to the absence of an explicit market for intraday credit. The implicit intraday rate has been estimated by Furfine (2001) for the federal funds market in the US and by Baglioni and Monticini (2008) for the e-MID market in Europe: both point to a very low level of the hourly rate (0.9 bp and 0.4 bp respectively) ? although statistically significant. While the cost of intraday liquidity may be negligible under normal conditions, we show in this work that it may gain economic significance in times of liquidity tension, leading to an inefficiency of the money market.
In the next section we provide the empirical evidence: after defining the implicit intraday interest rate, we document the striking rise taking place at the outset of the financial crisis. In section 3 we discuss this result and we suggest a rationale for it. Some concluding remarks are given in section 4. Finally, the Appendix extends the analysis to a longer time span, in order to check for the robustness of our results.
Download
PDF Ebook The intraday interest rate under a liquidity crisis: the case of August 2007
