Yves Mersch: The European Central Bank and the Federal Reserve - an ocean apart?
Speech by Mr Yves Mersch, Member of the Executive Board of the European Central Bank, at Harvard University, Boston, Massachusetts, 12 October 2016.
Interest rates on both sides of the Atlantic seem to be trapped near zero. At first glance, looking only at the short term policy rates, one might think that both the Federal Reserve (Fed) and European Central Bank (ECB) are facing very similar circumstances and challenges. Below the surface, however, the differences are pronounced.
In the euro area monetary policy continues to be highly accommodative, aiming to steer inflation back to levels in line with our definition of price stability. In the US, which is at a different stage of the business cycle, interest rate normalisation has so far been slower than expected.
It is worth taking a closer look at the details to understand the unprecedented actions of central bank on either sides of the ocean in recent years.
To do so, it is useful to divide the crisis into three separate phases. The first phase was the global financial crisis after the collapse of Lehman Brothers just over eight years ago. Central banks worldwide faced a largely common challenge to avert a collapse in aggregate demand. The second phase centred on sovereign debt, a problem which only affected the euro area. In the third phase, which is still in place today, stubbornly low global inflation and anaemic growth are compounding domestic challenges.
The first phase: combating a systemic shock
The first phase of the crisis had its origin in the US. In 2006, many experts considered the subprime crisis to be a minor incident with only local impact, but then it escalated into a financial crisis of systemic proportions that propagated to the whole advanced world. Many banks experienced a sudden stop in their access to money markets, and the ensuing liquidity squeeze in interbank markets affected credit provision across the United States and euro area.
The effects on the real economy were immediate and large. Both regions went through severe recessions. But the respective central banks and fiscal authorities responded promptly and decisively.
The Federal Reserve quickly cut its target for the fed funds rate to close to zero, then embarked on a series of large-scale asset purchases of US government bonds and mortgage-backed securities. Likewise, the ECB cut its main refinancing rate by more than 300 basis points between October 2008 and May 2009. To ease the liquidity constraints of the banking system we altered the mode and maturity of our operations and widened the range of eligible collateral.
These responses were, in many ways, an extension of the existing operational frameworks. The asset side of the Fed's balance sheet traditionally consisted mainly of holdings of US government bonds. By contrast, the ECB, lacking a single fiscal counterpart, injected liquidity mainly through repos with banks. This mode of operation reflected in turn the predominance of bank-based intermediation in the euro area.
Coupled with a sizeable fiscal impulse, both types of actions were successful in delivering the stimulus needed to counteract the shortfall in aggregate demand, stabilise output and initiate the economic recovery. Unemployment in both jurisdictions, which had spiked in the early phase of the crisis, levelled off and began to decrease.
At this point, however, the euro area entered the second phase of the crisis, causing its growth trajectory to diverge substantially from that of the United States. It experienced a sovereign debt crisis in 2011-12 which was unique among advanced economies.
Fears about the sustainability of the sovereign debt of some euro area Member States triggered a sharp widening of sovereign spreads and introduced the fear of redenomination risk: that some countries would revert to national currencies, invariably weaker ones, and thus leave the euro area.
Moreover, banks in some countries came under intense scrutiny over their domestic sovereign holdings and some national banking systems completely lost access to wholesale funding markets. Private investment contracted as the cost of capital rose and uncertainty spiked. The fiscal stance in the euro area shifted from expansionary to contractionary as governments acted to restore market confidence. The euro area entered a renewed recession.
The second phase: unique institutional challenges
The ECB responded to the sovereign debt crisis with various new monetary policy measures, such as our long-term lending operations, and even began purchasing sovereign bonds under the Securities Markets Programme. All these measures aimed to support bank funding markets and ensure that the flow and pricing of credit to the real economy was in accordance with our intended monetary policy stance.
But a question often asked is why we did not resort to large-scale asset purchases at this point, as the Fed was already doing. Bear in mind that inflation developments in the euro area at the time did not suggest a tight stance. Headline inflation was rising steeply from its 2009 trough, exceeding 3% in October 2011, partly driven by an increase in underlying inflation. Inflation expectations were well anchored around our objective.
Hence our diagnosis of the problem in this period focused more on monetarytransmission than on the policy stance. Our stance was not being transmitted evenly across all parts of the euro area. And so our measures in this period were specifically designed to remove those impairments in transmission.
The main issue we faced was the perverse interaction between the euro area banking sector and institutional incompleteness of our monetary union, which had fundamental implications for our policy in a system of bank-based monetary transmission. While the first phase of the crisis can be interpreted as a severe recession amenable to monetary policy, this second phase was the consequence of institutional challenges that monetary policy could only aim to mitigate, but could not overcome.
Despite swift efforts to set up a European Banking Union there were in fact three major differences between the euro area and US in terms of the environment facing monetary policy: the sluggish recovery of the banking sector, the fragmented fiscal framework of the euro area and the structural rigidity present in many euro area economies.
The necessary process of banking sector repair proceeded much more slowly in the euro area than in the US. Indeed, here in the US it was more or less completed by end-2011, but in some euro area countries it is still ongoing, even today. US authorities exerted pressure on banks (and non-financial corporations) to recapitalise with public money, via the Troubled Asset Relief Program. In the euro area, there was a preference for recapitalisation through organic growth.
Recapitalisation in the United States was bolstered by the early and effective use of stress tests in the US to restore confidence in solvent banks, reinforced by a credible backstop. By contrast, early attempts at stress tests in the euro area were hampered by splits among national supervisors and, with only national backstops available, the tests proved unsuccessful at restoring confidence in the banking system. US banks were also encouraged to recognise non-performing loans (NPLs) more quickly, freeing up capital for new lending.
The third difference was the ability to efficiently resolve insolvent banks and prevent the emergence of "zombie lenders". The US had a trusted process through the Federal Deposit Insurance Corporation, backed by a credit line from the US Treasury. By contrast, prior to the crisis many member states of the euro area did not even have a national resolution framework. Since 2008, the FDIC has resolved over 500, mostly smaller banks in the United States. While determining the exact definition is tricky in the euro area, the number appears less than a hundred.1
All this explains why the recovery of the banking sector took longer in the euro area, which our monetary policy had to work to mitigate. And what is more, when public authorities in the euro area did intervene to backstop their national banking sectors, the fiscal implications naturally differed: the onus fell entirely on national budgets. This in turn contributed to the second area where the euro area and the US diverged: the emergence of the so-called "vicious loop" between banks and sovereigns, which further delayed the financial repair process.
Importantly, this vicious loop not only further impaired monetary transmission, but it also had implications for the stance of monetary policy. The need for accelerated fiscal consolidation to quell the sovereign debt crisis constrained the ability of fiscal to carry out macroeconomic stabilisation in the euro area and hence the overall policy-mix. Whereas the US could consolidate its budget against the backdrop of a firming recovery, some euro area governments were forced into a pro-cyclical fiscal stance. That in turn worsened the macroeconomic environment in which monetary policy had to operate.
Institutions are again central to explaining the difference. While the US operates in a consolidated union - with a federal Treasury acting as the counterpart of the central bank - the euro area comprises 19 sovereign governments with separate national budget constraints, making sovereigns more vulnerable to a loss of market confidence. This is particularly valid amid a no-bail-out clause for member states without any debt-restructuring mechanism below the federal level.
Accordingly, national fiscal policies have to be sound ex ante to build fiscal buffers, as is the case for states in the US. Yet fiscal rules were only weakly enforced before the crisis, permitting public debt levels at the national level that were too high for a monetary union that is not complemented by a fiscal union.
That said, even countries with low starting debt levels - such as Ireland and Spain - ended up embroiled in the bank-sovereign loop, showing that the institutional framework was incomplete in other ways, too. In terms of sovereign risk, there was no fiscal backstop of sufficient capacity for sovereigns until the European Stability Mechanism was established in late 2011, leaving the euro area vulnerable to contagion. Likewise, there remains no mechanism for sovereign debt restructuring.
And in terms of bank risk, there was no backstop on the European level as one might expect for a European banking sector.
Still, it would be too simple to lay the challenges faced by monetary policy solely on the European supranational institutional framework; national characteristics were in many ways inadequate too. Indeed, the third major difference between the US and the euro area can be found in the structural flexibility of their respective economies.
Many euro area economies entered the crisis with long-standing structural rigidities. In a more flexible economy, output tends to be less affected by shocks as relative prices adjust more quickly, and the recovery in output is faster since the economy can reallocate resources in a more efficient way.
Unemployment has decreased more slowly in the euro area than in the US, not only due to a weaker demand environment, but also because of labour market rigidities. Indeed, every time since the 1970s that there has been a downturn in economic activity, unemployment in the US has quickly readjusted when the economy picked up, while in euro area it has taken much longer to do so. And once the unemployment rate starts falling it rarely returns to the previous low. The result is that euro area unemployment has risen steadily for 40 years.
So in sum, in this second phase of the crisis, monetary policy in the euro area, which works primarily through banks, had to contend with a protracted process of banking sector repair, exacerbated by a sovereign debt crisis and a lack of fiscal space, and compounded by sub-optimal institutions to tackle structural issues and support the transmission process.
The measures we adopted were specifically aimed at addressing the impairments of the transmission mechanism. Our two 3-year lending operations, in particular, were effective in securing term funding for banking systems cut off from wholesale markets, and initially had a strong easing effect on sovereign spreads. Our interventions in sovereign bond markets under the Securities Markets Programme have also been found to have reduced yields and volatility in the countries targeted.2
And when redenomination risks spiked, the Outright Monetary Transactions announcements were extremely successful in removing unwarranted fears about a break-up of the euro area. Importantly for the credibility of the ECB, the legality of OMTs was confirmed on the highest judicial level by the European Court of Justice and also by national constitutional courts.
Fortunately, around the same time as these measures were introduced, euro area leaders called on the Presidents of major European Institutions to develop a roadmap towards deepening Economic and Monetary Union.3 This has led to a marked improvement in the institutional environment, which has contributed both to an improved transmission of our policy and to a stronger recovery.
In 2014, the Single Supervisory Mechanism was launched under the aegis of the ECB, which directly supervises the 129 significant banks in the euro area. In contrast to the earlier attempts to build confidence in the euro area banking sector, its founding act was to carry out a thorough and credible asset quality review and stress test of bank's balance sheets. The result has been an improved acknowledgment and provisioning for NPLs, while bank capitalisation has also increased - the banks participating in the Comprehensive Assessment had built up additional capital of over €200 billion between the onset of the crisis and the Assessment.
In terms of repairing the banking sector, the new Bank Recovery and Resolution Directive provides a harmonised EU framework for resolution with a requirement of bail-in. And the new Single Resolution Mechanism ensures that decisions on resolution are made consistently across the euro area, with resolution financing given at the union level and then recouped from the banking sector. This is a vital step to make resolution more effective, while it should also go some way towards severing the link between banks and their sovereigns by shifting the financial burden onto the private sector, or in the interim, onto the union.
But there is more still to be done. For the bank-sovereign nexus to be truly broken it is clear that Banking Union needs to be completed. That means moving towards genuine risk-sharing over deposit guarantee schemes, coupled with genuine risk reduction on banks' balance sheets.
European institutions need to be strengthened in other areas, too, notably those governing fiscal policies. To my mind this must entail moving from a framework of decentralised rules to more effective central institutions; it hardly helps credibility when we take years drafting new rules only to break them at the first test. And there is still an urgent need for more structural reforms - because third phase of the crisis has been, above all, structural.
The third phase: the re-emergence of structural challenges
This third phase - which we are still in today - has been characterised by two main challenges: low inflation and low growth. Just as in the initial phase of the crisis, these are common challenges for advanced economy central banks. But what has become increasingly clear is that the challenges today have a much stronger structural element. Low global inflation is partly a cyclical phenomenon - linked to falling energy prices and the slowdown in major emerging market economies - but it is also a result of structural changes brought about by globalisation. And the secular slowdown of global growth has brought with it declining long-term interest rates across the advanced world.
To be sure, the intensity of these challenges has varied across jurisdictions. The euro area, still recovering from its double dip recession, has faced a more prolonged period of low core and headline inflation, which the ECB has reacted to by launching three new unconventional tools: a set of targeted long-term refinancing operations with in-built incentives for banks to lend, a negative interest rate policy, and an asset purchase programme encompassing public and private sector securities. These measures aim both to smooth transmission and to expand the policy stance to combat excessive disinflationary pressures. And they have been effective in easing financial conditions and supporting the recovery.
Yet it is clear that, for all central banks, the structural backdrop has made monetary policy normalisation slower. Consider lower growth. It is now clear that the long-run growth potential of major economies has been falling for a number of years. For example, the Consensus forecasts for long-run growth real GDP for the US - that is the average growth six to ten years - stood at 3.1% in early 2003. The latest figure is just 2.2%. Long-run growth forecasts for the euro area were revised down from 2.2% to 1.4% over the same period. These lower forecasts represent in part the impact of ageing populations, but also reflect the falling trend in total factor productivity.4
This affects monetary policy because slower growth in potential output reduces what is often referred to as the equilibrium interest rate.5 While not directly observable in the real world, this widely used concept is organised around the notion of an equilibrium level of the rate of interest where there is neither upward nor downward pressure on inflation. If the equilibrium rate falls, the margin above the effective lower bound becomes compressed, reducing the space in which central banks can operate to provide accommodative policy.
One might reasonably ask why central banks did not identify this trend sooner. The answer seems to lie in an overreliance on pro-cyclical estimates of potential output. Indeed, the revisions to estimates of potential growth have mostly occurred after the crisis, whereas it is most likely that the slowdown in productivity had already startedbefore the crisis but was masked by the credit-fuelled boom. For example, in 2007 the euro area's output gap was estimated by international institutions such as the IMF, the OECD and the European Commission to be negative - somewhere between -0.2% and -0.6%. The most recent estimates put the output gap in the region of positive2.7% to 3.4%, a significant downward revision to views of potential.
At the same time as potential growth has been declining, the share of global factors driving inflation seems also to have been increasing. In large part, the common global trend can be explained by the decline in price of energy and other commodity prices. But low global inflation goes beyond weakness in domestic demand and falling energy and commodity prices. In recent years there has been a growing understanding of the role of global demand factors on domestic inflation.6
For example, it has been observed that changes in the degree of slack in OECD economies are having a smaller impact on inflationary pressures than it has in the past.7 In part this changing correlation is a result of the success of monetary policy in anchoring inflation expectations.
But structural factors are also playing a role: higher import volumes as a result of advanced globalisation have increased the importance of international prices relative to domestic prices, forcing domestic mark-ups to be less sensitive to the state of the domestic economy. This may imply that monetary policy has to react more aggressively to counteract shocks.
So in combination, central banks may be facing a two-pronged challenge to their mandates: a situation where, due to the role of global inflation, more stimulus is needed than in the past to deliver their domestic mandates; and where, due to the falling equilibrium interest rates, their ability to deliver that stimulus is more constrained. That would certainly offer one explanation for why monetary policy has needed to stay accommodative for so long across advanced economies.
This does not make monetary policy ineffective. As we have seen in the euro area, even with a low equilibrium rate central banks can still steer conditions in the wider economy by targeting directly the constellation of interest rates that matter for borrowers. But one nonetheless cannot deny that the fall in the equilibrium rate poses a challenge for central banks. This challenge for monetary policy has been recognised, with a number of innovative solutions proposed by academics. Let me mention two but also say upfront that I am not particularly convinced by any of them.
One option that has been mooted is to allow nominal interest rates to go deeper into negative territory. Yet there is an effective limit to lowering rates since ultimately this conflicts with the ability of citizens to hold cash. Therefore some scholars have called for the imposition of electronic money and the abolition of cash. As I have discussed at length elsewhere, I see serious problems with such an approach.8 It risks gambling with central banks' credibility at a time when credibility is paramount. And perhaps even more importantly, it denies the crucial social function of cash.
Another option proposed by some observers is for central banks to raise their inflation objectives and therefore to lift equilibrium nominal rates. Leaving aside the debate about the costs of higher inflation, I am not sure that this idea is entirely convincing. The reason is that the mechanism through which it works is the signalling channel - people expect higher inflation, so inflation rises. But at a time when inflation is low everywhere, and when central banks are running very accommodative monetary policies just to achieve that, one can question whether such signals would really be credible. I see a risk, in fact, that changing our strategy to rely on fuzzy concepts and unobservables would only end up damaging our credibility.
So what we need today, in my view, is to shift the debate away from its narrow focus on monetary policy. Rather than concocting more and more innovative solutions for central banks to get around the lower bound, we should focus instead on what other policy areas can contribute. In a world of very low interest rates, monetary policy cannot be the only game in town - and we should not pretend it can be.
In particular, we urgently need to press ahead with structural reforms to improve long-run economic growth, which will consequently lift equilibrium interest rates. In my view it is much better to provide monetary policy with greater room with a process that increases output and wealth than by ones that increase social costs. At the same time, fiscal policy needs to become more effective at complementing monetary policy in managing cyclical fluctuations. In the current euro area context that means those countries with fiscal space should use it better, and going forward fiscal rules should be more rigorously enforced so that countries enter downturns with greater ability to carry out active stabilisation and invest into education and innovation, to name but a few.
For individual jurisdictions the exact policy mix will of course differ. In some cases the emphasis may be on increasing public investment, in others on more growth-friendly tax policy. Likewise structural reforms that aim at raising productivity have to be tailored towards domestic priorities. But given the common nature of the challenges for central banks, all countries would benefit from prioritising it. And the positive spill-overs will clearly be larger if all act in unison.
Roughly eight years of extremely accommodative monetary policies on both sides of the Atlantic might lead to the notion that the ECB and Fed have been confronted with similar or even homogenous challenges. This would be a misapprehension.
Major institutional and structural differences at the onset of the financial crisis explain why the clean-up efforts in the US were faster and more effective and the recovery was more sustainable.
After the institutional short-comings of Monetary Union had been painfully exposed, much progress has been achieved on the road to completion and further integration. However, this institutional reform package took time. Now, economic recovery, although still timid, is under way in the euro area.
But this finding cannot be taken as an excuse to rest on our laurels. More determined structural reforms need to be implemented to lift the growth potential and overcome the ultra-low interest rate phase. A more determined approach to tackle the problem of non-performing loans is necessary to tackle the remaining legacy of banking crisis.
When the macroeconomic ship is engulfed by a major storm, all hands - not just monetary policy - need to man the pumps. This, indeed, holds true on both sides of the ocean.
1 Lang, J. H., Peltonen, T. and Sarlin, P., "A framework for early-warning modeling with an application to banks", Working Paper Series, ECB, forthcoming.
2 Ghysels, E., Idier, J., Manganelli, S. and O. Vergote, (2014), "A high frequency assessment of the ECB Securities Market Programme", ECB Working Paper No. 1642.
3 Junker, J.-C., Tusk, D., Dijsselbloem, J., Draghi, M. and M. Schulz, (2015), "Completing Europe's Economic and Monetary Union", European Commission, Brussels.
4 See, e.g. Gordon, R. (2016) "The Rise and Fall of American Growth: The U.S. Standard of Living since the Civil War", Princeton University Press.
5 Constâncio, V., (2016), "The challenge of low real interest rates for monetary policy", Lecture by Vítor Constâncio, Vice-President of the ECB, Macroeconomics Symposium at Utrecht School of Economics, 15 June 2016.
6 Inflation as a global phenomenon has been documented e.g. by M. Ciccarelli and B. Mojon (2010), "Global Inflation", The Review of Economics and Statistics, 92:524-535, although there is less evidence of the phenomenon in developing countries: Parker, M (2016) "Global inflation: the role of food, housing and energy prices", RBNZ Discussion Paper 2016/05. Measures of global economic slack are good predictors of national inflation in advanced countries, as shown empirically e.g. by C. Borio and A. Filardo (2007), "Globalisation and inflation: New cross-country evidence on the global determinants of domestic inflation" BIS WP no. 227; and in New Keynesian open economy models e.g. by R. Clarida, J. Gali, and M. Gertler, (2002) "A Simple Framework for International Monetary Policy Analysis," Journal of Monetary Economics 49: 879-904.
7 IMF (2013), "The dog that didn't bark: has inflation been muzzled or was it sleeping?", World Economic Outlook, April.