Bienvenue à Paris and to the Banque de France (BDF) … I’ve switched to English so as to welcome you all to this new Conference Center created by you, dear Christian. What better way to wish you farewell after 12 years as Governor of the BDF and six years as President of the Bank for International Settlements (BIS)? And what better way for me to welcome for the first time this unique community of actors, academics and observers who are together genuinely contributing to a better understanding of monetary policy and financial stability?
To quote a few numbers, some 200 VIPs are gathered here, representing more than 80 central banks (CBs), governments and international organizations, as well as about 50 universities, think tanks and financial institutions. It’s no surprise, as Christian enjoyed meeting all of you at international gatherings and research conferences.
The Symposium title, chosen jointly by the BDF and the BIS, relates to the prevalence of very low interest rates for the past 7 years in advanced economies and, by contagion, around the world. Even in the USA after the FED lift-off, interest rates still qualify as historically low and the tightening cycle has been announced as “gradual”.
Now why should we, as central bankers, care about ultra-low rates? For many reasons, notably because they can lead to a misallocation of resources or bubbles, and because they are currently associated with persistently low inflation or ‘lowflation’ – shall I even say ‘noflation? The latter is of particular concern given the costs resulting from price or wage rigidities and distortions, as well as the risk that we might fall into deflation.
My role as “opening speaker” is modestly to stress some puzzles and challenges we face. You will then have the whole day to provide the best answers. To kick off, I’ll focus on where we stand, with two puzzles, and then on what monetary policy should/can do, with four challenges.
I Where do we stand? And to start with, what is ultra-low?
A lot of what matters for CBs is low or ultra-low compared to history, notably in advanced economies (AEs): first, inflation, be it headline or core, realized or expected; second, interest rates on safe assets, whether short or long, nominal or real, ex post or ex ante; third, term premia in spite of measurement uncertainties. And all variables have been decreasing and are now low at the same time which is new.
Spreads and volatility have also been lower for longer, although there is some market discrimination between agents and between emerging market economies (EMEs), as well as periodic spikes in volatility, especially recently.
Next, why are inflation or interest rates so low and how does this impact policy?
First puzzle: inflation remains low in countries where unemployment is close to the NAIRU, like the USA, or where the output gap has been sharply reduced, like Germany; this may imply that the Phillips curve has a lower slope too. Yet, various studies by the IMF or the BDF show that, even if slopes have tended to flatten since the early 1990s, there has been no clear-cut change over the last years.
There are two possible explanations. The benign one is that transitory disinflationary shocks, e.g. due to commodity prices, temporarily shift the Phillips curve downwards. The worrying one would be a disanchoring of long-term inflation expectations, threatening the objective of price stability; such a disanchoring is suggested by the surprising correlation between spot oil prices and market expectations derived from inflation-linked swaps up to 5 years in 5 years. People can reasonably differ in their interpretation; hence the scope for healthy policy debates.
As regards low policy or market rates, these are mainly seen as reflecting lower equilibrium, or neutral, interest rates. This leads us to the second puzzle: does the lower neutral interest rate result from cyclical or structural factors?
On the one hand, the factors may be cyclical (or mainly demand-driven); indeed, we are currently seeing low levels of global demand and trade. After a financial crisis, cyclical factors may persist for a long time: for example, with the deleveraging of agents and hence the reduction in their spending. In this case, aggregate demand policies, including monetary policy, have an important role to play.
On the other hand, the factors may be structural (or mainly supply driven): e.g. lower productivity growth, higher inequalities or population ageing which pushes the savings rate higher. The other usual suspects behind secular stagnation include globalization, which boosts price/wage competition, or the savings glut. And the latter may be involuntarily compounded by liquidity regulations which can permanently raise demand for safe liquid assets, driving down risk free rates. In this case, the neutral rate may have become so negative that monetary policy alone may not suffice to raise inflation.
The jury is still out; but faced with this uncertainty, CBs’ mandates push them to avoid the risks of the first type, related to inaction, rather than those of the second type, related to pro-action; in other words, be
ready to risk doing too much rather than too little. Moreover, lower supply may well stem from persistently lower demand, e.g. with some hysteresis resulting in disqualified workers and higher structural unemployment.
II Now let’s turn to what monetary policy should/can do
I would like to mention four challenges:
First challenge: if the factors at play are global, what can domestic policies do?
CBs have domestic mandates whereas the prevalence of lowflation and low rates suggests that some common or external factors may be at play. This is especially problematic for smaller open economies. And the existence of a “global financial cycle” à la Rey highlights the limits of flexible exchange rates in insulating those economies.
Ultra-accommodative policies in AEs have definitely had spillover effects on EMEs, including the often-forgotten –albeit welcome- reduction of spreads in the latter, as shown by IMF research. Nevertheless, the formulation and stance of policies in one country can take account of feedback effects from others. And it goes without saying that the world at large would have been even worse off if AEs had fallen into deflation.
Aside from building and using forex reserves, domestic policies should therefore contribute towards making the domestic real economy and its financial system more resilient to external shocks. In this regard, prudential policy aimed at building buffers or protecting financial institutions may complement monetary policy, provided these measures do not conceal a form of protectionism or an attempt at deglobalization.
Second challenge: what should the inflation target and its horizon be?
Some economists have called for adjusting the inflation target, although in opposite directions: a few have argued for a higher target to avoid testing the Zero Lower Bound often (ZLB); others have called for a lower target, assuming lowflation is the New Normal. In both cases, hasty changes may only affect the CBs’ credibility and the expectation channel of monetary policy.
We have more flexibility on the target horizon. This is the medium term and should not be confused with the 3-year horizon of the forecasts published by the ECB for instance, which may fall short of the target level; what matters is being on a path that is consistent with this medium-term mandate. Forward guidance and the announcement that rates will be “lower for longer” should provide reassurance over the CB’s determination to do “whatever it takes” to fulfill its mandate within this mandate.
True, uncertainty about the causes of lowflation and low rates leads to uncertainty over the neutral rate and the need for accommodation; but the more uncertain our environment, the more important and useful it is that we be clear and predictable in order to reduce this uncertainty. According to simulations carried out within the Eurosystem and the BDF, clarifying that the future stance will remain accommodative for many years enhances the efficacy of Quantitative Easing (QE).
Third challenge: how effective can non-conventional monetary measures be?
The ZLB has proved not to be the effective lower bound. Going negative has sharply lowered and flattened the yield curve in the euro zone, which is especially relevant given the importance of long-term financing. And so far the substitution for cash has also remained negligible. Yet, there are limits on how far one can go. Thus lower key interest rates should be combined with other non-conventional measures.
In short, CBs have innovated a lot, both in terms of rates and quantities. We knew CBs could do a lot and they have actually done a lot, even if they cannot do everything.
In the main advanced economies, the precise macroeconomic effect of large-scale asset purchases is still under discussion. But I can summarize the Eurosystem’s shared analysis and Mario’s recent statement: our purchases will add around half a point to inflation in 2016 and almost the same to growth. In the US case, the QE2 launched in late 2010 is the most similar to our QE. Controlling for its size, roughly half of ours, it has had a broadly comparable impact, even if published estimates differ somewhat.
Fourth challenge: how to deal with the risk of a mismanaged exit from the ZLB and financial instability?
Staying too low for too long would be counterproductive: ultra-low margins may discourage financial intermediation and lending to the real economy, while addiction to forward guidance may silence contrarians and hamper market information on fundamentals. Yet the exit may be bumpy and, hence, postponed or even aborted. Worse, as regularly pointed out by the BIS, staying too low for too long may fuel the next financial crisis as liquidity addiction and excess risk taking eventually lead to bubbles and financial instability while macro-prudential policy may be too slow to help.
Even if monetary and macro-prudential policies are two distinct instruments serving two distinct objectives, they impact each other; but it is hard to integrate one when formulating the other: e.g. to integrate a sustainable “output gap à la Borio” (BIS) into a monetary rule, insofar as the business and financial cycles do not often coincide.
On the other hand, an ill-timed exit or its premature expectation may also put some financial institutions at risk, depending on their maturity mismatch. But as the major CBs will not all exit simultaneously, the impact should be mitigated. Thus, what is sometimes depicted as the “great divergence” between monetary policies on the contrary reflects, in my view, a “great wisdom”: as domestic cycles differ, exits will be spread over time. This moment of exit is obviously further away for the euro zone. Yet when it does come, we shall all be as innovative, as ready to learn from each other and as decisive when acting to exit as we have been when entering into non-conventional territory.
For the sake of time, I will not elaborate on whether ultra-low rates alter the monetary transmission channels and their interactions with financial stability. As I already said, we have the whole day to discuss these tricky issues thanks to our speakers who are not only prominent but also open and humble; I noticed and appreciated these qualities in their recent speeches exploring the new territory we are in.
Instead, let me conclude on the limits of the scope for action by CBs. Indeed, I prefer to stress that central banks cannot be the only game in town and, more precisely, that monetary and macro-prudential policies are not conducted in a vacuum. Structural problems require structural answers; in particular, only structural reforms can raise potential growth; aggregate demand policies cannot provide a substitute. What they can do –and it’s precious— is to help better manage time.
Speaking of managing time, I think mine is up now and I shall give the floor to Christian who can give us all the benefit of his experience and wisdom.
- R. Baldwin, C. Teulings, 2014: Secular Stagnation: Facts, Causes and Cures, CEPR Press; in particular, Chapter 8 by O. Blanchard et al., Chapter 9 by R. Caballero and E. Farhi, Chapter 11 by R. Koo.
- BIS Annual Report, 2014: “Debt and Financial Cycle” (Chapter IV).
- O. Blanchard, E. Cerutti and L. Summers, 2015:“Inflation and Activity – Two Explorations and their Monetary Policy Implications”, IMF Working Paper No 15/230.
- N. Chatelais, A. D. Gaye and Y. Kalantzis, May 2015: “Low inflation in the euro area: import prices and domestic slack”, Rue de la Banque 6, Banque de France (showing no clear-cut change in the slope of the Phillips curves in the Eurozone and France over the last 10 years).
- M. Draghi, the Economic Club of New York, Dec. 4, 2015, based on work by a Eurosystem task force, including BDF staff: http://www.ecb.europa.eu/press/key/date/2015/html/sp151204.en.html.
- E. Engen, T. Laubach and D. Reifschneider, 2015: “The Macroeconomic Effects of the Federal Reserve’s Unconventional Monetary Policies”, Finance and Economic Discussion Series 2015-005. Board of Governors of the Federal Reserve System (and references therein).
- S. Guilloux-Nefussi, March 2015: “Globalization, Market Structure and the Flattening of the Phillips Curve”, Banque de France Working Paper No 539, (proposing a theoretical model explaining a possible flattening of the Phillips curve since the 90s).
- IMF Global Financial Stability Report, April 2013 (Chapter I).
- H. Rey, Nov. 2014: “Monetary Policy and the Global Financial Cycle”, BDF International Symposium on “Central Banking: The Way forward” and Jackson Hole Symposium Proceedings, Sept. 2013.