We welcome Rafael Repullo’s insightful comment on the report on the ECB strategy review that CEPR recently published and that we co-authored. Repullo provides a sharp discussion of some technical features of central banking that aficionados of monetary policy will enjoy – the transmission of monetary policy from short to long rates, yield curve control, the corridor system, variable versus fixed tender auctions, and the link between prices and quantities in the Friedman rule. Here we briefly reply to his more pointed criticisms of our report.
As a way of introduction, the ECB strategy review involved hundreds of economists at the ECB and the national central banks, who produced many pages discussing almost every single issue facing any central bank today. The ECB governing board decided what it found most important and released it as a 2-page “monetary policy statement” together with a 16-page “overview of the monetary policy strategy”. In our independent and parallel work, we had ourselves written a report on what we thought were some of the most important issues: a clearer statement of the inflation target and the use of make-up strategies; a normalisation of the operational framework, as too many frequently used tools were called unconventional; the need to clarify the role of the central bank’s balance sheets; the uncomfortable topic of fiscal-monetary interactions that now take multiple shapes; and the way in which dealing with climate risks could be made operational within an inflation-targeting framework. As the ECB had either not highlighted these, or done so with a different perspective, we felt that our report was a good complement to the ECB’s strategy and an invitation to keep the discussion open on future strategic reviews.
When it comes to the tools of monetary policy, which Repullo focuses on, the ECB’s monetary policy statement covered them in its point 8: “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.” Chapter 2 in the report devoted much more attention to this topic. It discussed the need to define a new norm and gain credibility, how to communicate the use of multiple tools, the central role of the deposit rate, the desirability of keeping the demand for liquidity satiated, the role of credit agencies in affecting collateral, the dangers posed by emergency liquidity assistance, how financial crises require the ECB to become market maker and lender of last resort, and how to consider risk premia in the different sovereign debts. More specifically, in the use of its multiple tools, the ECB today discusses its policy rates in terms of interest rates, its forward guidance in terms of dates of expected take-off, its asset purchases in terms of billions bought of each asset, and its long-term refinancing operations through a mix of rates and quantities. The report proposed focusing all tools by stating what their likely impact on interest rates would be at different horizons. This would give a common denominator to all policies, making the complementarity between different tools more transparent.
Repullo raises three relevant concerns in his first set of comments. First, that it will not be feasible to achieve a particular term structure of risk-free interest rates even if the ECB uses all the tools that it has. We agree. Doing so would be akin to yield curve targeting. As the report writes: “Yield curve control, introduced in 2016 in Japan, involves committing to whatever set of purchases and sales are needed to attain a particular value, or even a range, for a specific long maturity in the yield curve. Doing so in the euro area would be difficult, and is probably a bad idea.” Second, Repullo notes that it would be difficult to characterise the optimal combination of “tools that would deliver the target yield curve”. Again, we agree. The report does not propose this. Rather, it writes: “Using the risk-free yield curve, or the long rate, as the unifying principle is a much more modest step than yield curve control”. The report proposes that, when the ECB communicates a new policy, it states how it expects that it will move the yield curve in what particular direction by what approximate amount. It does not propose that the central bank announces a desired long rate and then states what mix of policies will achieve it. Repullo’s third and final comment defends that it is “better not to tie your hands with a commitment to a particular yield curve goal”. Emphasising again that nowhere did the report defend a yield curve goal, this is an important point in the classical discussion of discretion versus rules. Repullo defends a situation where the ECB is not clear about what it is trying to achieve with each of its different tools so that it can preserve discretion; the report sees the need for transparency and some commitment.
Next, Repullo criticises the report’s discussion of the separation principle because the ECB abandoned it already through its policies of the last decade. But, as Repullo implicitly acknowledges by referring to these policies as “unconventional tools”, this has not been normalised. The point of the report is that the ECB should use a strategy review to affirm that this is the new normal. Repullo confidently asserts that “[s]ince 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”. We disagree. Since the ECB has not stated this, there is such a risk. The strategy review made a point of stating, as it should, that forward guidance and asset purchases are now part of the strategy. The ECB did not do so for the separation principle. Strategy reviews are about putting on paper what is the strategy, whether that has been the status quo or not.
The third set of comments by Repullo is more puzzling. After a clear discussion of floor and corridor systems, he concludes that “the main refinancing rate largely irrelevant, being replaced by the deposit facility rate as the signal of the stance of monetary policy”. This is precisely what the report says, going further by suggesting that the ECB should clearly state this in the strategy. Because of the irrelevance of the main refinancing rate, the report describes a world in which it is identical to the deposit rate. There is no disagreement between the report and Repullo’s comment on this. The same applies to the discussion of the Friedman rule. The report writes that “[t]he ‘Friedman rule’ states that the opportunity cost of social means of payment should be zero, since the cost of producing them is approximately zero. Satiation of reserves is, therefore, the Friedman rule applied to reserves.” Note that the report had defined these two terms on the previous page: “The difference between the interbank rate and the interest rate on the reserves gives this opportunity cost. Once the supply of deposits grows large enough, the opportunity cost falls to approximately zero, as it did in the euro area some years ago. Economists would say that the demand for reserves was satiated, or that the market for reserves was saturated.” Repullo writes that “the Friedman rule is about eliminating the opportunity cost of holding assets that provide liquidity services”. Precisely as we wrote: same content, different words. When reserves are below satiation, an increase in the supply of reserves reduces the cost of financial intermediation. Therefore, increasing reserves to the point of satiation is an example of the Friedman rule. Note that efficiency requires the differential between the policy rate and the interest on reserves to be zero, but this does not require that interest rate be zero. With a positive interest rate on reserves, the central bank is free to set any rate consistent with this condition. So, with interest paid on reserves, there is no disagreement between us.
To conclude, the substantive discussion is on the first two points. On those, Repullo argues against the report’s proposal of transparent communication policies using the yield curve because this would remove discretion and might lead to trying to target the yield curve. He further argues that the separation principle is already dead, making the arguments of the report against it redundant. This discussion ultimately highlights that the recent ECB strategy review did not fully succeed to transparently establish a new normal for the tools of monetary policy. Thankfully, there will be future reviews, and we hope that those in charge of them in the future will find this exchange useful.