The European Central Bank did what most people expected on Thursday and announced it would halve the rate at which it buys securities - and hence the rate at which it "prints" new central bank money to inject into the economy - to €30bn a month. There are many good things to say about this. In contrast to the Federal Reserve's "taper tantrum" in 2013, the ECB's market communication was perfect. And this compromise between hawks and doves keeps the purchase programme in force indefinitely, leaving the central bank room for manoeuvre rather than signalling a firm end to purchases at some future point.
As Free Lunch argued on Wednesday, however, keeping the current €60bn-a-month purchase rate unchanged would have been better. A decent growth rate does not indicate inflationary pressure when the economy is still well below its potential capacity. And just look at the ECB's own survey of professional forecasters published today: even by 2019, inflation is expected only to reach 1.6 per cent, well below the ECB's 2 per cent ceiling. It is too soon to withdraw monetary stimulus.
Virtually all the talk now is nevertheless of when and how to "normalise" - ie tighten - the ECB's policy. So it may seem an odd time to think of new unconventional policies. But I want to give some more attention to the suggestion I brought up earlier this week, following a comment by Steve Donzé: that the ECB could follow the Bank of Japan by targeting long-term interests rates directly.
Note that long-term rate targeting - also known as "yield curve control" because when you control both short-term and long-term rates, you pin down the whole term structure of interest rates (the yield curve) - is a substitute for "quantitative easing". QE targets a quantity of long-term securities to purchase in order to ease long-term borrowing conditions by driving down the yields on the purchased assets, driving investors into other securities. Direct rate targeting simply specifies the desired yields on the relevant assets and buys (or sells) securities to achieve it. It is similar to how many central banks fix short-term interest rates.
How would the ECB do this? Since it's the central bank of many countries, it would have to target a composite rate - the average rate on a specified bundle of all its member countries' government bonds. But this is no more complex than what it already has to do in its QE programme. A very useful first step would be to construct and regularly report an official synthetic long-term interest rate - say the average 10-year yield on governments' bonds weighted by either the size of their economies or their total debt stock. Once the measure was in place and followed by markets, the ECB would be in a position to influence the rate through signalling or market operations.
There are two strong reasons why this would be a particularly useful step for the ECB to take. One is that worries remain that it may run out of assets to buy under its self-imposed cap on how much of each bond it is willing to own. That is not just a potential operational challenge; it also means that arithmetic alone makes markets expect a hard limit to monetary accommodation. By halving the purchase rate, the ECB has gained some time, but not removed the built-in expectation of tightening. By announcing a long-term interest rate target instead, the Japanese experience suggests that standing ready to buy and sell as required would mean the ECB did not actually have to do so. A bonus would be to calm the nerves of those worrying about a large central bank balance sheet, even as the ECB could credibly say it had a tool for further easing should this be required.
The other reason is that direct long-term interest rate targeting would be useful when policy has to be tightened as well as loosened. At some point monetary stimulus will have to be withdrawn, and some of that tightening will aim to raise long-term rates. Under the current regime, the ECB will have to copy the Federal Reserve's strategy for reducing its bond holdings. A rate-targeting regime would offer an alternative. Since the policy would involve targeting a rise in the synthetic common interest rate, the obvious way to achieve that would be for the ECB to package (securitise) the bonds it holds (or create new, asset-backed securities based on them), and sell the bundles that exactly correspond to the constructed rate. That would not only achieve the desired monetary policy, it would also create the common safe asset many are asking for without having to go through the political minefield of debt mutualisation which exposes different eurozone countries' taxpayers to one another.
A safe normalisation of eurozone monetary policy in due time, with the legacy of creating a permanent foundation for financial stability - what's not to like?
