The CEPR report on the ECB strategy: A comment on the tools of monetary policy
VoxEU Blog/Review

The CEPR report on the ECB strategy: A comment on the tools of monetary policy

Rafael Repullo comments on some recommendations and statements on the tools of monetary policy at the ECB from the recent CEPR report on the central bank's strategy.

The CEPR report on the ECB strategy brings together a leading set of European economists (Lucrezia Reichlin, Klaus Adam, Warwick McKibbin, Michael McMahon, Ricardo Reis, Giovanni Ricco, and Beatrice Weder di Mauro) to review some of the most important issues central bankers are dealing with in the current environment. As they state in their introduction, their contribution should not be read as a commentary to the new ECB strategy published in July 2021. But, inevitably, many parts of the report read as such.

In this blog, I focus on some recommendations and statements in the second chapter of the report, entitled “The tools of monetary policy at the ECB: A new normal”. In particular, I have two main concerns. 

The first one is about the proposal to use the yield curve to describe the stance of monetary policy. In their words: “we recommend summarizing the monetary policy stance in terms of prices rather than quantities, and describe the intended policy impulse as the expected price effects onto the risk-free yield curve at all maturities”. 

The second one is about the working of the corridor/floor system for monetary policy implementation, including the role of the separation principle, the tender procedures for monetary policy implementation, and the connection with the Friedman rule

Monetary policy stance and the yield curve

Moving from a policy stance described by the short-term interest rate to a stance described by the expected effects on the entire yield curve might bring some benefits in terms of communication, given the complex interaction between the different unconventional tools, but in my view has three problems. 

First, it is not clear that it is feasible to achieve a particular term structure of risk-free interest rates using the unconventional tools of forward guidance, lending programmes, and purchases of public and private assets (quantitative easing). I can think of circumstances in which the size of the required interventions would hit the issue share limit of 33% that currently applies to the purchases of securities (or any other such limit). In this case the target would not be feasible, compromising the credibility of the central bank. 

Second, even if it were feasible to target the yield curve, given the state of our knowledge on the effect of unconventional tools (and their interactions), it would be tricky to characterise the optimal combination of (both conventional and unconventional) tools that would deliver the target yield curve. Of course, this does not mean that we should try to advance in the understanding of the impact of the different unconventional tools, as in Rostagno et al. (2021). 

Third, even if it were feasible to target the yield curve it is not clear that it would be desirable. Monetary policy operates through many channels and has the advantage of “getting in all the cracks” (Stein 2014). But to do this, it has to be used with flexibility considering the conditions of many debt markets. From this perspective, it would be better not to tie your hands with a commitment to a particular yield curve goal. 

Monetary policy tools

One first issue to note here is that some recommendations of the report do not have much bite, since the ECB implemented them some time ago. In particular, the report says that “the ECB should abandon the rhetoric of the ‘separation principle’”, and that “the ECB should use the deposit rate as the main short-term policy rate.” 

The separation principle states that one can separate the decisions on the (short-term) policy rate (taken by the Governing Council of the ECB) and the decisions on liquidity provision designed to keep very short-term interest rates (such as EONIA) close to the policy rate. The separation principle made sense in the ‘conventional’ monetary policy regime, but it ceased to operate once the ECB started using unconventional tools, because their activation also corresponded to the Governing Council. 

Moreover, providing ample liquidity through quantitative easing implied that short-term rates quickly approached the deposit facility rate. Since the current size of the excess reserves is likely to stay with us for some time, there is no risk that the separation principle will be reinstated in any form and that the deposit facility rate will not continue to be the main short-term policy rate. 

Similarly, the report states that “the set of innovative policies that the ECB put in place since 2007 should now be considered as part of the new ‘normal’ operational framework”. Well, this is what the ECB monetary policy strategy statement said on this matter: “The primary monetary policy instrument is the set of ECB policy rates. In recognition of the effective lower bound on policy rates, the Governing Council will also employ in particular forward guidance, asset purchases and longer-term refinancing operations, as appropriate.” 

The corridor system of monetary policy implementation

To dig deeper into this discussion, it is useful to review the implementation of monetary policy with a corridor system, like the one used by the ECB. Such system involves a monetary policy rate (the main refinancing rate) and two other rates: one above (the marginal lending facility rate) at which banks can borrow reserves from the central bank, and one below (the deposit facility rate) at which banks can place their excess reserves. 

As I noted in Repullo (2011), there are three ways in which conventional monetary policy can be implemented. First, as in the case of the ECB until the deployment of unconventional tools, one may have a structural liquidity deficit and implement monetary policy with a regime under which the central bank lends reserves to the banks. In this case, the policy rate would be the central bank’s lending rate (like the ECB’s bid rate in the main refinancing operations). Second, as in the case of the Federal Reserve prior to the crisis, one may have an approximate liquidity balance and implement monetary policy by conducting open market operations designed to compensate the daily movements in the autonomous liquidity creation factors. In this case, the policy rate would be the target short-term money market (Federal funds) rate. Finally, one may have a structural liquidity surplus and implement monetary policy with a regime under which the banks place their excess reserves at the central bank, in which case the policy rate would be the interest rate paid by the central bank on these excess reserves (the deposit facility rate). 

In the period with a structural liquidity deficit, the ECB injected most of the required liquidity through the weekly main refinancing operations. The policy rate was the main refinancing rate placed in the middle of the corridor between the deposit facility rate and the marginal lending facility rate. During this period, the ECB used a number of tender procedures, starting with a fixed rate rationed quantity tender, then a variable rate tender, and moving at the time of the global financial crisis to a fixed rate full allotment tender. In this last tender procedure, the ECB provides (against collateral) unlimited amounts of liquidity at the policy rate (see Ayuso and Repullo 2003 for a model of the different tender procedures). 

There has been a fair amount of confusion about the fixed rate full allotment tender procedure. The confusion is partly due to the fact that the ECB announced the change as part of the unconventional policies designed to deal with the global financial crisis, probably because it was fighting against a monetary policy tradition based on controlling the money supply. But there is nothing unconventional about fixed rate full allotment tenders, as evidenced by the fact that they were used by, for example, the Bank of Spain prior to the monetary union. 

This confusion shows in the report. For example, the authors write: “With variable tenders auctions at the main refinancing rate to satiate the demand for reserves, the central bank is operating what is sometimes called a ‘floor system’ when setting interest rates.”

Variable rate tenders at the main refinancing rate deliver interest rates at the main refinancing rate, not at the lower deposit facility rate, because banks would refrain from borrowing at the main refinancing rate to end up with excess reserves placed at the deposit facility. 

The move to a structural liquidity surplus, following the implementation of asset purchase policies, made the main refinancing rate largely irrelevant, being replaced by the deposit facility rate as the signal of the stance of monetary policy. It should be noted that almost 20 years ago, Goodfriend (2002), with amazing foresight, advocated implementing monetary policy by flooding the market with reserves and making the policy rate the interest on reserves. So the only striking element of using a floor system was the huge amount of excess reserves, coming as a response to hitting the effective lower bound and having to resort to quantitative easing. 

Enter the Friedman rule

Finally, I would like to comment on the connection between the implementation of monetary policy with a structural liquidity surplus and the Friedman rule. The report states that “the Friedman rule would be achieved when the market for reserves was saturated”. But the Friedman rule is about eliminating the opportunity cost of holding assets that provide liquidity services. Assuming that reserves provide liquidity services to banks (and indirectly to depositors), then the focus should be on the policy rate, not on the quantity of reserves. Since you can operate a floor system with a policy rate at any level – negative, zero, or positive – there is simply no connection with the Friedman rule. 

In this regard, it is important to stress that the market for reserves is unique in that the central bank can set independently the price and the quantity. But this is nothing new. As pointed out by Goodfriend (2002), this regime “would make full use of two monetary policy instruments – open market operations and interest on reserves – to enable a central bank to simultaneously pursue interest rate policy and an independent objective for aggregate bank reserves.”


Ayuso, J and R Repullo (2003), “A Model of the Open Market Operations of the European Central Bank,” Economic Journal 113: 883-902.

ECB (2021), The ECB’s Monetary Policy Strategy Statement. 

Goodfriend, M (2002), “Interest on Reserves and Monetary Policy,” Federal Reserve Bank of New York, Economic Policy Review 8. 

Reichlin, L, K Adam, W McKibbin, M McMahon, R Reis, G Ricco, and B Weder di Mauro (2021), The ECB Strategy: The 2021 Review and its Future, CEPR Press. 

Repullo, R (2011), “Monetary Policy Operations: Experiences during the Crisis and Lessons Learnt,” in M Jarocinski, F Smets, and C Thimann (eds.), Approaches to Monetary Policy Revisited - Lessons from the Crisis, European Central Bank. 

Rostagno, M, C Altavilla, G Carboni, W Lemke, R Motto, and A Saint Guilhem (2021), “Combining Negative Rates, Forward Guidance and Asset Purchases: Identification and Impacts of the ECB’s Unconventional Policies,” ECB Working Paper No. 2564.

Stein, J (2013), “Overheating in Credit Markets: Origins, Measurement, and Policy Responses,” Federal Reserve Bank of St. Louis.