In central banking circles, the talk of normalisation has been under way for some time. The Federal Reserve and the Bank of England have both moved interest rates off the lowest levels they hit after the great recession. At the European Central Bank, Mario Draghi has now begun to talk about the path for interest rate increases in the eurozone in order to shape expectations for a moderate tightening process.
“Normalisation” means “going back to normal”. But what, precisely, does that mean? If “normal” means the pre-crisis norm, there are many ways in which that differs from today. So here are four senses in which one could aim to “get back to normal” — and in all four, doing so is either undesirable or meaningless.
First and most simply, people sometimes use normalisation to mean getting back to a “normal” level of central bank interest rates. This is silly. Central banks should aim to set the rates that are most conducive to fulfilling their public mandate. There are many reasons — including persistently benign inflation and a lower “natural” rate (at which desired savings and investment match) — why central bank interest rates should stabilise permanently below where they were before the crisis.
More sensibly, normalisation could refer to a call for a less stimulative monetary policy stance, recognising that this will be achieved at lower policy rates than in the past. But this is problematic too. At the very least, ending monetary stimulus is premature, as I have been at pains to argue. Also, we do not have a very good grip on how stimulative central banks’ current stances are — because we do not know well how far their policy rates are from natural rates, and because other factors (including expectations about future monetary policy) are constantly altering broader financial conditions.
Third, normalisation could refer to the overall policy framework — the objective central banks are trying to achieve. For the past quarter-century, most countries have set their monetary policy to target the consumer inflation rate. Here, central banks have largely stuck to the pre-crisis normal, or at least they say they have. Getting inflation to the target rate remains the official goal — but one that most central banks have undershot for years. If normalisation meant doing more aggressive stimulus to bring inflation higher, that might not be a bad thing. On the other hand, there is a strong case to be made for rethinking the objective.
In high-debt economies, in particular, there is a case to make that monetary policy should aim to keep the expected real burden or value of nominal debts close to what they were expected to be when they were contracted. If the inflation target is missed one year, it is not good enough in this perspective to hit it next year; one needs to compensate for past misses. Take the eurozone as an example. As Martin Wolf shows, its price level is now 7 per cent lower than it was expected to be in 2007 on the assumption that the ECB would do its job. Even if it hit its near 2 per cent target from now on, that shortfall would remain permanently, making the burden of older debts correspondingly heavier.
There is of course a challenge involved in changing from one regime to another — Ben Bernanke has proposed to pre-announce a change from inflation targeting to price level targeting to take place the next time the Fed policy rate approaches zero — but that is no reason not to consider whether another regime would do a better job, in which case any new normal should be distinct from the status quo ante.
Fourth, normalisation could refer to the instruments or levers central banks use to attain their objectives. We should make two observations. The severity of the crisis led central bankers to deploy new tools (or old tools in new ways) when the standard tool of short-term policy rates seemed inadequate. The toolbox now includes large-scale asset purchases (quantitative easing), forward guidance or commitments to future actions, negative short-term interest rates, direct targeting of long-term rates (as the Bank of Japan does) and tiered pricing of reserves (used in conjunction with negative rates to remove the so-called lower bound on interest rates). More radical tools have been proposed but not tried, including targeting stock market prices or issuing “helicopter money”.
Some of these tools have been deployed because we learnt that the old ones did not work well. But there has been reluctance every step of the way away from the old normal. And calls for normalisation today — for example to higher rates — are in part a function of the unwillingness to use the new tools to the full. Much analysis of the challenges to monetary policy (see for example Paul Krugman’s reflections on the 20th anniversary of his famous liquidity trap paper) continue to assume, without justification, that central bank policy rates cannot go (much) below zero. Good policy thinking today requires more boldness in the consideration of “abnormal” policy instruments.
The point is that we should not relearn things we thought we knew but were wrong, nor unlearn the new knowledge that we have discovered through the crisis. If there is to be normalisation, at least let it be to a new normal that is nothing like the norms of old.
