Benoît Cœuré: Monetary policy in a low-growth environment

Speech by Mr Benoît Cœuré, Member of the Executive Board of the European Central Bank, at the 4th Frankfurt Conference on Financial Market Policy, organised by the Sustainable Architecture for Finance in Europe Policy Centre, Goethe University, Frankfurt am Main, 28 October 2016.

Since the onset of the crisis, the ECB has introduced a wide range of unconventional measures in order to meet its inflation objective, including negative deposit facility rates, targeted long term refinancing operations, and asset purchases. These measures were - and remain - necessary, and have been effective. But in the public debate, there is an increasing focus on the potential negative side effects of these measures and the rationale behind their deployment. This triggers two questions: why did we launch unconventional monetary policy in the first place? And how can we mitigate the potential risk of negative spillovers from monetary policy over the medium term?

The answer to the first question is clear - these measures have been taken to react to the large negative shock caused by the global financial crisis, which has exacerbated a number of structural factors that have been driving down real interest rates even before the crisis. The unconventional measures are a reaction to the challenge of low global interest rates, not their cause.

In my remarks today I will set out the challenges for monetary policy caused by structural factors such as low interest rates and low productivity growth. Understanding the structural nature of these challenges is key to answering my second question on mitigating potential spillover risks from monetary policy in the future.

The impact of low productivity growth on monetary policy

To understand the structural challenges for monetary policy, it is useful to consider two interlinked concepts for the setting of monetary policy - potential output and equilibrium interest rates. Understanding the interplay between these two concepts is at the heart of the current policy settings of the European Central Bank. How they evolve will determine the future path for inflation and at what stage monetary policy can eventually be normalised.

Potential output represents the level of activity where capital and labour are sustainably used. Equilibrium nominal rates are the level of nominal interest rates where there is neither upward nor downward pressure on activity, and are composed of two parts: the real equilibrium rate and inflation expectations. Actual levels of activity above potential put upward pressure on inflation and levels below potential put downward pressure. In similar fashion, setting actual interest rates below equilibrium rates stimulates economic activity, whereas setting interest rates above equilibrium restrains activity.

Over recent years, economic activity in the euro area has been below potential - in other words there has been a negative output gap - which has put downward pressure on inflation. The classic textbook prescription is simple: loosen monetary policy to drive interest rates below equilibrium, stimulating output to bring it back to potential and generate inflationary pressures to achieve the inflation objective. In general terms, the ECB's monetary policy since the crisis fits that textbook prescription.

But the finer details also matter. Simple models in textbooks assume that the growth of potential output and the rate of equilibrium interest rates are fixed and immutable. In reality, productivity growth has slowed in the euro area and other advanced economies. Combined with unfavourable demographics, the slowdown in productivity has depressed potential output growth and lowered equilibrium interest rates. The productivity slowdown has been particularly pronounced in the euro area, where both total factor productivity and capital per employed person have markedly slowed down.

These changes matter for the conduct of monetary policy, and they create a dilemma. The slowdown in growth below potential has increased the need for macroeconomic stabilisation. At the same time, the overhang of public and private debt and their interaction through the bank-sovereign nexus impairs the use of fiscal policy and places a greater burden on monetary policy to carry out that stabilisation. Yet the fall in equilibrium rates reduces the margin of operation for traditional monetary policy instruments to stimulate the economy. Let me address these three elements in turn.

The greater need for macroeconomic stabilisation

Shocks constantly hit economies, driving activity away from potential and inflation away from the central bank's objective. Recent examples include the global financial crisis and large fluctuations in oil prices. In principle, these deviations from potential output can be managed by appropriate macroeconomic stabilisation. But cyclical deviations can have more lasting effects on potential output through a process termed hysteresis.

For example, following a prolonged period of output below potential, the public may become pessimistic about future growth and income prospects and consequently reduce consumption and investment. In the short term, lower consumption and investment worsen the negative output gap and are deflationary. The lower investment also reduces longer-term capacity and hence future potential output. In a similar fashion, there is a risk of hysteresis. This means that people made unemployed due to the cyclical downturn may remain out of employment for too long, losing valuable human capital and finally becoming structurally unemployed, a tragedy for them and their families, and a loss for society.

There is some evidence that the prolonged period of negative output gap in the euro area since the crisis has weighted on potential. Potential growth is estimated to have fallen from 1.6% a year over the period 2000-07 to 0.7% over the first five years of the crisis.1 Skill mismatch has increased in a number of countries, in part arising from the shift in sectoral composition as the construction sector shrunk rapidly.2 The high rates of youth unemployment are also likely to result in labour market "scarring".

The financial crisis and the bank-based nature of finance in the euro area amplified the slowdown in potential growth. The transmission of monetary policy in the euro area has been impaired by fragilities in the banking system, compounded by the overhang of public debt, on which I will shortly say more. Loss of access to finance can have a marked impact on business performance and in aggregate lower potential output.

To put some perspective on the problem, let me highlight recent research carried out for the ECB looking into the impact of loss of finance on firm performance in the euro area.3 The research looks at thousands of loan applications made by small and medium-sized enterprises to a major German bank between 2009 and 2012. In particular, it focuses on firms that are around the cut-off line for loan acceptance. These firms are in general very similar in terms of balance sheet and profitability metrics, but some lie slightly above the line and some lie slightly below.

The study maps the outcomes for those firms in the year following the loan rejection. On average, a rejection leads to a 9% fall in assets, and lower investment and employment. The effects are amplified in firms with low liquidity, with such firms seeing a 5.6% fall in investment and a 7.2% fall in employment. Yet these are not unprofitable or inefficient firms - the average return on assets for firms just below the cut-off line is a little over 14%.

The long-term risk from the prolonged period of sluggish growth also extends beyond real activity into prices. A key part of our monetary policy is to anchor inflation expectations around our - price stability aim of a rate of below, but close to two percent over the medium term. Although our objective is forward looking, there is a risk that continued undershooting of our inflation aim will cause households and firms to revise down their inflation expectations. And lower inflation expectations directly lower the equilibrium nominal interest rate, requiring a lower interest rate to create monetary stimulus. Thanks to our measures, inflation expectations have remained anchored. According to the ECB's latest survey of professional forecasters, expected inflation at a five-year horizon is 1.8%. But market-based expectations have been lower and more sensitive to current levels of inflation, and without our measures, the risk of a de-anchoring would have been very serious.

The increased burden on monetary policy

Taken together, a long period of growth below potential requires robust policy action to prevent the low growth from becoming entrenched into the long run. But at the same time, falling potential growth also increases the burden on monetary policy to carry out that macroeconomic stabilisation.

The policy mix in the euro area is already heavily tilted towards monetary policy given the decentralised nature of fiscal policy by 19 national governments and the limited success of European fiscal rules in creating incentives to build up national fiscal space. The available fiscal space in most euro area countries is in addition constrained by the overhang of public debt inherited from the crisis. Lower potential growth and weaker inflation inhibit the reduction of public debt ratios by acting on both sides of the fraction. Higher expenditure arising from higher structural unemployment and lower tax revenues from weaker inflation restrict the ability of governments to save, i.e. reduce the numerator. And low inflation increases the real savings required to serve a given level of nominal debt. Similarly, weaker potential growth lessens the ability to increase the denominator and grow out of the overhang.

Since the crisis, the overhang of public debt has not just reduced the ability of fiscal policy to share the burden of macroeconomic stabilisation with monetary policy, but it has also impaired the transmission of monetary policy through the sovereign-bank nexus.

Recent research by the ECB shows that during the sovereign crisis, banks in stressed countries increased their holdings of domestic sovereigns.4 This was particularly marked for publicly owned banks and recently bailed-out banks. As the price of stressed sovereign bonds fell, these banks were forced to deleverage, reducing their lending. This drop in lending was not just confined to the domestic market - foreign subsidiaries of banks with head offices in stressed countries also reduced lending. The notion that sovereign bond purchases by domestic banks acts as a useful stabilising mechanism in the face of financial stress therefore needs qualification.

The reduced margin of operation for monetary policy

As I have just mentioned, falling growth rates had increased the need for macroeconomic stabilisation, and put greater burden on monetary policy to carry out that stabilisation. But it has also affected the capacity of monetary policy to stimulate the economy. That is because the long-run growth in output is an important determining factor of equilibrium interest rates, which have been falling in major advanced economies for the best part of the past three decades.

As I noted above, in order to generate inflationary pressure in the economy, central banks have to set interest rates below equilibrium. As equilibrium interest rates have declined, so has the interest rate that is stimulatory. Estimating equilibrium rates is tricky, and estimates are subject to bands of uncertainty. But a range of estimates puts the euro area real equilibrium rate, which is to say the equilibrium rate minus inflation expectations, at close to zero, with some measures even negative.5

To provide the required monetary accommodation to bring inflation back to our objective, the ECB has needed to bring the main refinancing rate to zero, and our deposit facility rate has been negative since 2014. These low and negative rates are a direct consequence of the decline in the equilibrium rate. If that rate were higher, so would our interest rate settings. Failing to follow the move lower in the equilibrium rate would have created the risk of remaining trapped in a low growth, low inflation equilibrium or even of falling in a deflationary spiral.

But there are limits to how low interest rates can go. First of all there is the physical lower bound, where households and businesses disintermediate the banking sector and hold their money in cash. Experience of other central banks shows that the physical lower bound is below where we currently have our deposit facility rate. But there may well exist an economic lower bound at which the negative impact on monetary policy transmission through banks outweighs the positive benefits of low rates for economic activity.6 Low interest rates have implications for bank profitability in both the short and long term and can weigh on credit provision and financial stability.

Such negative effects have not materialised so far, partly because lower rates have initially generated capital gains on banks' fixed-income portfolios and lowered banks' funding costs, and partly thanks to the overall positive impact of low rates on the volume and riskiness of bank loans. But over time, falling rates will squeeze bank interest margins.

In the longer run, there are also risks to financial stability if low interest rates result in asset price bubbles, which are vulnerable to sharp reverses once interest rates rise. Similarly, if banks "search for yield" by increasing lending to lower quality borrowers or if they roll over non profitable, "zombie" loans to ailing companies, there may well be a higher aggregate default risk.

ECB staff estimates suggest that recent monetary policy actions in the euro area have so far been net positive for bank profitability, relative to a scenario of no policy action,7 and we don't see today asset price bubbles which would threaten the euro area's financial stability. But if rates are low for too long, the negative effects may well dominate and impair the effectiveness of our measures.

Of course, low interest rates are not the only factor affecting bank profitability in the euro area. Europe as a whole is overbanked, and a number of jurisdictions have suffered from low profitability and high cost-to-income ratios for several years. With interest rates likely to remain low for the foreseeable future, even once monetary policy normalises, banks will have to revisit their business models to ensure continued profitability over the medium term.

The monetary policy reaction

Faced with activity below potential and inflation below its objective, the ECB has sustained accommodative monetary policy since the onset of the crisis. We have responded to the fall in equilibrium rates and the negative output gap by cutting our main refinancing rate to zero and our deposit rate into negative territory. Our targeted long-term refinancing operations have encouraged lending by banks and helped to mend the transmission mechanism.

More recently, our asset purchase programme (APP) has sought to influence a broader range of interest rates. This recognises that there is in fact not just one equilibrium rate, but a constellation of rates across a range of maturities and credit conditions that are important for the borrowing decisions of firms and households.

It is clear that the accommodative monetary stance is having the desired effect in the euro area. GDP growth continues, albeit still at a sluggish pace. According to the European Commission's most recent estimates, the negative output gap has more than halved since 2014, from -2.5% to a projected -1.1% this year. The euro area unemployment rate fell from 11.2% at the start of 2015 to 10.1% now. Headline HICP inflation has risen by a percentage point over the same period, although mostly through the removal of energy price base effects. Credit standards have been loosening for firms and households, loan demand is increasing and banks are using the additional liquidity from the APP to grant more loans.8

There is, so far, little evidence of negative effects from our unconventional measures. Bank deposits continue to grow, and there is little sign of a disproportionate increase in holdings of cash, so the low interest rates are not causing disintermediation. Similarly, there is little sign of excessive increases in real estate prices across the euro area as a whole, although some local markets have begun to see some stronger increases.

Taken together, the ECB's current policy settings are appropriate and are clearly providing the support required to sustain the recovery. In providing that support, we are minimising the risk that the sluggish growth becomes entrenched as enduring stagnation and that inflation becomes de-anchored. Monetary support for the recovery will continue until inflation is sustainably adjusting to our objective.

Ensuring sustainable policies for the future

But there is a risk that prolonged use of unconventional measures brings about greater risks. Absent structural changes to potential growth and equilibrium interest rates, so-called unconventional measures will become conventional measures in the euro area.

Because monetary policy has shouldered the lion's share of the burden of post-crisis stabilisation, the debate has naturally centred around what the ECB should do. However, structural problems require structural solutions and other actors need to shoulder some of the burden. The public debate needs a renewed, wider focus.

In particular, fiscal policy has to shoulder a greater share of the burden of macroeconomic stabilisation by providing better support to investment and by achieving a growth friendlier composition. Equilibrium interest rates have to rise from their current low levels through structural reforms improving the efficiency of our economies.

Other actions are needed to ensure euro area resilience over the medium term, and may take some time to have their effects. This includes steps to complete banking union, complement it with a capital market union and address in a credible way the bank-sovereign nexus. This may also include steps to create a common fiscal capacity, backed by credible rules for national budgets and under appropriate democratic scrutiny.

As our political fabric is under severe strains, these are not low hanging fruits. But postponing the necessary reforms is not a valid option anymore. Procrastination and forbearance have not served the euro area well. Our financial system is a case in point. If our banks had been cleaned up and strengthened early after the crisis, our growth path would today be higher.

Building a more resilient euro area economy through better rules and structural and financial sector reforms will protect the potency of monetary policy to perform its stabilising role in the presence of structurally low equilibrium interest rates. But the benefits extend beyond short-term macroeconomic stabilisation. Higher potential output recreates fiscal space for governments and shifts equilibrium interest rates upwards. If the reform path is credible, part of these benefits can materialise today. Higher equilibrium rates, and a smaller share of the stabilisation burden, will reduce the need for monetary policy in the future to use the full range of its unconventional measures. But by far the best dividend from these reforms will be to improve the welfare of euro area citizens and their trust that Europe can make their lives better. This in itself is a matter of urgency.

1 Anderton, R. et al. (2014), "Potential output from a euro area perspective", ECB Occasional Paper Series No. 156.

2 Anderton, R. et al., ibid.

3 Berg, T., (2016), "Got rejected? Real effects of not getting a loan", ECB Working Paper Series No. 1960.

4 Altavilla, C., Pagano, M. and S. Simonelli, (2016), "Bank exposures and sovereign stress transmission", ECB Working Paper Series No. 1969.

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 Cœuré, B., (2016), "Assessing the implications of negative interest rates", Speech at the Yale Financial Crisis Forum, Yale School of Management, New Haven, 28 July 2016.

7 See Rostagno, M. et al., (2016), "Breaking through the zero line: the ECB's negative interest rate policy", Brookings Institution, Washington DC, 6 June 2016. Presentation available on the Brookings Institution website.

8 ECB (2016), "Euro area bank lending survey", Third quarter 2016, European Central Bank, October.