Ultimi link in questa sezione
Bce: un comportamento razionale, ma folle
A number of analysts are calling for the ECB to act as the lender of last resort in the Eurozone government bond market (see for example Wyplosz 2011 on this site). Up to now it has resisted. But why should the ECB refuse to take up the role of last-resort buyer?
A rational, non-dogmatic explanation
Here is a possible explanation that has the merit of being based on rational behaviour. Other explanations that have been popular are based on a belief that the persons deciding about this issue are driven by dogmatic thinking preventing them to see the need to act. This may be the case, but it remains interesting to try to explain the ECB’s behaviour assuming that its decision-makers behave rationally.
When a central bank is called upon to be the lender of last resort it has to evaluate costs and benefits of its actions. Let us rephrase the problem in terms of the costs and benefits of inaction, ie of not providing the last-resort buying service.
Why central banks act as last-resort buyers for banks
Picture central bank facing a banking crisis. (Later we will do the same analysis for the case of the government bond market.)
- The cost of inaction arises from the risk that inaction will lead to a collapse of the banking system.
If the latter collapses the central bank will most likely be made responsible.
- The benefit of inaction is the avoidance of future moral hazard risk which is beneficial to maintaining a stable banking system in the long run.
When evaluating cost and benefit, the time horizon over which these costs and benefits materialise matters a great deal.
- When the central bank faces a banking crisis the cost of inaction is likely to be realised very quickly.
When banks are close to collapsing, the cost of not providing the lender-of-last-resort service is almost instantaneous. This has to do with the fact that the banks’ liabilities typically have very short maturities (demand deposits, interbank deposits).
- The benefits of inaction, however, will be realised in the future, possibly far in the future.
It is even likely that only the successors will reap the benefits, and they may not even be aware of this.
Asymmetric timing of costs and benefits
This asymmetry in the timing of the realisation of costs and benefits goes a long way towards explaining why even the most conservative central bank is likely to wish to avoid the immediate cost (collapse of the banking system) even at the cost of foregone future benefits, even if these benefits are very large. This asymmetry explains why the ECB did not hesitate for a moment to provide last-resort buyer support to Eurozone banks in 2008, despite the fact that in doing so it created moral hazard risk in the future.
What about government bond markets?
We can now apply this cost/benefit analysis to the government bond market. Here we have a striking difference with the banking sector. The sovereign debt crisis occurs at a snail’s pace compared to banking crises. When investors sell government bonds and push the interest rate upwards, they affect the cost of borrowing of governments with some delay because the maturity of the bonds is typically of the order of five to seven years. As a result, there is not the imminent threat of a rapid collapse as there is with a banking crisis.
The result is that when the central bank faces a sovereign debt crisis the lack of immediate danger has the effect that a conservative central bank, such as the ECB, will attach more weight on the long-term benefits of reducing moral hazard. The central bank will therefore wait far longer to take action.
Note that this does not mean that moral hazard risk is more important in sovereign bond markets than in the banking sector. Bankers are just as likely to take additional risk when they know that in times of crisis the central bank will provide liquidity, as governments are. In addition, there is no reason to believe that the risks bankers take on is less dangerous than the risk taken on by governments. The only difference is that the imminence of a collapse is higher during a banking crisis than during a sovereign debt crisis. As a result, a central bank is likely to reduce the weight on moral hazard risk.
A forecast of action
The previous analysis leads me to the following forecast.
- The ECB will only act when the cost of inaction is immediate and clear.
As a result, the ECB is likely to wait until the sovereign debt crisis has degenerated into a full-scale banking crisis.
- There can be little doubt that the sovereign debt crisis will lead to a banking crisis.
The reason is that the continuing decline in the price of sovereign bonds will hammer the banks’ balance sheets to such an extent that the losses become unbearable.
In addition, sovereign debt crises lead to funding problems for banks and a risk of being shut out from the interbank market. Thus there is a moment when the sovereign debt crisis inevitably triggers a banking crisis. This will be the moment when the timing asymmetry between costs and benefits is such that the ECB will see the merits of being a lender of last resort. Only then will the ECB come to action.
Concluding remarks
All this is quite depressing for two reasons.
- First, the amount of liquidity the ECB will have to inject in the banking system is likely to be higher than the amount that is necessary to stabilise the government bond markets.
This assertion is based on a simple fact. Total liabilities of the Eurozone banks are more than three times the liabilities of Eurozone governments (De Grauwe 2011).
- Second, the banking crisis will also trigger a deep and long-lasting recession (see Reinhart and Rogoff 2009).
All this may in the end endanger the Eurozone itself.
References
De Grauwe, P (2011), “The ECB as a Lender of Last Resort”, VoxEU.org, 18 August.
Reinhart, C and K Rogoff (2009), This Time is Different, Princeton University Press.
Wyplosz, Charles (2011), “An Open Letter to Dr Jens Weidmann”, VoxEU.org, 18 November.
This article may be reproduced with appropriate attribution. See Copyright (below).