Peter Praet: The low interest rate environment in the euro area
Keynote speech by Mr Peter Praet, Member of the Executive Board of the European Central Bank, at a Pension Funds Conference, organised by the Netherlands Bank, Bussum, The Netherlands, 10 September 2015.
I would like to thank Jonathan Yiangou, John Hutchinson and Katri Mikkonen for their contributions to this speech.
Ladies and gentlemen
Thank you very much for inviting me to address this conference.
Over the last year conditions in fixed income markets have been, by any standards, highly unusual. Policy interest rates in the euro area have fallen to close to 0% - and negative in case of our deposit facility - while we have seen the 10-year German government bond yield drop as low as 0.075%, with at one point up to about 36% of euro area euro-denominated government bonds in negative territory. And though interest rates have retraced some way back up since mid-April, they clearly remain at historical lows.
For investors whose business models are centred around fixed income this is undoubtedly a very challenging environment. I am, of course, aware of the challenges the pension fund sector in particular is facing. What I would therefore like to focus on in my remarks today is the ensuing question: what are the conditions that will allow for rates to rise back to levels that, looking retrospectively, are considered "more normal"?
My answer can be summarised as follows. Very low interest rates are not so much a choice of the central bank as a reflection of economic malaise in the global and euro area economy. Some of the drivers of that malaise are secular, which are more challenging for public policy to affect. But others are related to demand conditions that appear weak even when compared with a long-term potential growth rate that is already disappointingly low.
Such persistent economic slack relative to potential can be reversed through targeted measures. That includes tackling the cyclical headwinds that are dragging on aggregate demand; facilitating the process of balance sheet repair; and removing the structural barriers to higher real growth prospects, which can itself reinvigorate demand through expectation and confidence channels.
The role of monetary policy is mainly to provide countercyclical stabilisation while the economy goes through this transition. When the economy is weak, after all, disinflationary pressures increase, which we have to counter to deliver on our mandate. The monetary policy measures currently in place in the euro area follow from this: they were decided as a direct result of a euro area inflation rate well below comfortable levels and persistent weakness in the economy. And in meeting our objective, we support growth and inflation, which in turn creates the conditions for interest rates to "normalise" once more. Of course, the level that rates will return to in the future is dependent upon the degree to which the secular drivers, currently depressing rates, can be reversed.
Challenges of the low interest rate environment
Let me start, however, by briefly touching on the challenges the current situation poses for the pension fund sector, as well as for the insurance sector given its similar focus on fixed income assets and the prevalence of long-term liabilities. Large euro area insurers show quite reasonable profitability,1 albeit the aggregate picture masks substantial heterogeneity across countries and types of business. Small and medium-sized, non-diversified life insurers with high policyholder guarantees are typically highly exposed to low yields and have found their business models more under pressure. For pension funds the impact of low interest rates is strongest on those maintaining defined benefit schemes.
The challenges for the insurance sector were underscored by the mixed performance in the stress test undertaken in 2014 by the European Insurance and Occupational Pensions Authority (EIOPA).2 While the top 30 European insurers all fulfilled the Solvency II requirements in the baseline scenario, 14% of participants - mainly small firms representing 3% of total assets in the sample - did not do so. The similar exercise currently being conducted by EIOPA for pension funds will provide valuable insights into the strengths and vulnerabilities of the sector in the current market environment, an issue for which, at present, data is relatively scarce.
The challenges, however, are not limited to the sectoral level. How insurance companies and pension funds adjust to the low interest rate environment could also have wider implications. With total assets of over €9 trillion, the strategies they employ to cope with low investment returns could clearly have spillover effects both to financial markets and the broader economy.
One dimension of adjustment that could be relevant here is related to reallocation of portfolios towards higher return assets. "Healthy risk-taking" - to borrow a famous expression coined by Ben Bernanke - is at the very heart of monetary transmission. When central banks lower short-term interest rates in normal times, financial entities respond by shifting their portfolio allocations in precisely the direction that is needed to advance the objectives of monetary policy: they reinforce their duration exposure and step up risk-taking in general. By doing so, they amplify the stimulus that central banks introduce by intervening in an otherwise narrow and relatively unimportant market segment - the overnight interbank market. It is via this transmission mechanism that monetary policy is effective at all in producing a generalised easing of financial conditions.
But that said, the longer low rates persist, the more we have to be watchful that healthy diversification does not morph into under-appreciation of risk that may create fragilities down the road. This could impact a broad swathe of market participants given the importance of insurance and pension funds, in particular as providers of long-term funding. Data for large euro area insurers provide some evidence of re-risking via a gradual rotation of portfolios towards the bonds of vulnerable euro area countries and emerging markets, and a maturity extension, although there are no indications yet of excessive risk-taking.
Another dimension of adjustment that could have broader effects is related to the possible changes in saving behaviour generated by stronger incentives for risk-shifting towards policyholders. While moving towards unit-linked and defined contribution policies might increase the resilience of insurance and pensions companies, it could increase uncertainty over future benefits and thus raise precautionary savings among several cohorts of people. Data suggest, however, that this dimension of the adjustment remains contained. Unit-linked insurance and defined contribution pensions still constitute below 20% of life insurance policies and pensions in the euro area, and the pace of growth remains slow. In addition, survey results by EIOPA indicate that insurers have adjusted largely by focusing on diversification into non-life and asset management businesses or lowering of the guaranteed rates on new policies.3
A third potentially problematic dimension of adjustment is related to incentives to increase concentration to reduce costs, which can bring with it financial stability concerns. In Europe, a handful of large insurers - some part of financial conglomerates - already play a central role in the financial system, although their individual links to banks and to other market participants do not currently seem to be over-concentrated.4 The creation of ever larger champions, however, could change that picture. The work conducted under the Financial Stability Board on Global Systemically Important Insurers (G-SIIs) is thus highly topical, and also important from the European perspective.
Understanding low interest rates5
In this environment, I can appreciate why some in the industry may call for higher interest rates. After all, the simplest solution to the challenges facing the sector would seemingly be for interest rates to rise as quickly as possible. But it is important to understand why interest rates are so low. And a closer examination reveals that the underlying drivers are not so much central bank policies as global and euro area-specific economic factors, some of which are more secular in nature and others which are more associated with the legacy of the post-Lehman financial crisis.
To comprehend this point, it is useful to decompose long-term nominal yields into four components: expected inflation over the term of the asset; the expected path of short-term real rates; the inflation risk premium and real term premium, which together represent the compensation required by investors for holding long-term bonds as opposed to rolling over short-term securities. All these components have contributed to the very low long-term rates we observe today.
From a long-term perspective, nominal yields on long-term bonds have been on a declining trend in all major advanced economies since the 1980s. Improvements in monetary policy frameworks have played a role in this: the commitment to low and stable inflation, and the adoption of explicitly quantified price stability objectives (or targets) by an increasing number of central banks, have brought down long-term inflation expectations and compressed inflation premia, both of which have contributed to lower nominal yields. But the decline in nominal long-term rates is also explained by the real component: the expected real rate as well as the real term premium. Part of this decline in real interest rates can be accounted for by long run, secular forces; the other part is due to more cyclical dynamics.
If one takes the textbook Solow growth model as an organising device for the different forces driving real interest rates in the long run, they ultimately pertain to productivity and population growth, and savings behaviour. The intuition is that these forces determine investment and therefore the demand for loanable funds, which have to be matched by savings. Seen over the long-term, the growth rate of total factor productivity (TFP) has been slowing in the euro area for decades. Population growth has also declined from about 0.7% in the early 1970s to around 0.3% in recent years. And looking into the very distant future, the downward pull from adverse demographics could accelerate: some model-based simulations suggest that demographic trends - notably a rising supply of funds provided by an ageing population - could result in a significant reduction in the real rate in the long-run.6
On top of the gravitational pull of these secular forces, there are cyclical factors weighing on real interest rates that are more directly linked with the global financial crisis. Notably, the euro area is still working its way through a "balance sheet recession", where a situation of severe debt overhang sets the conditions for a sharp downturn, which in turn creates the need for substantial deleveraging and prolongs the length of the slump.
This is the product of a pre-crisis period where, misreading the signs of declining productivity growth, firms, households and in some cases governments accumulated excessive debt on the basis of overly optimistic real income expectations. The result has been excess savings in the post-crisis period caused by the need for all three sectors to repair their balance sheets, while over-indebted firms and households in some countries have been confronted with higher credit premia, further depressing investment. Investment indeed remains well below pre-crisis levels.
The damage caused to the economy by these dynamics has been massive. Youth and long-term unemployment has risen significantly in the euro area, with "hysteresis" effects estimated to have notably reduced labour supply. This feeds back into a slower potential growth rate, and potentially for some time, given the strong sectoral dimension to the crisis and the need for extensive re-skilling and reallocation of workers. Slower potential growth has also coupled with heightened uncertainty, discouraging new investment projects. The upshot has been a continuous decline of long-term growth expectations in the euro area, all of which adds to the downward pressure on real rates.
Such post-debt crisis effects have been visible in many advanced economies, contributing to the low level of global interest rates we see today. But most would agree that the negative impact of the crisis has been stronger for the euro area than for other jurisdictions. The reason for this is that certain features of the euro area have made working through a debt crisis harder and more protracted, namely institutional and structural weaknesses at both national and euro area levels.
The absence of an integrated euro area banking system is one such weakness. The Great Moderation - and a euro-specific misrepresentation of the type of risk that investors were exposed to when accepting to finance imbalances in other members of the monetary union - had encouraged a formidable expansion in cross-border banking activities prior to the crisis. Yet despite an illusion of integration, the euro area still effectively consisted of multiple juxtaposed national banking systems. Large savings and investment gaps within monetary union were therefore financed mainly through short-term banking debt, creating latent vulnerability to a "sudden stop".
When the cycle turned, there was little scope for risks to be reallocated across the union - also due to lack of formal macroeconomic risk-sharing mechanisms in the architecture of our Treaty institutions. Worse, as banks in some of the most vulnerable countries remained heavily exposed to their own sovereigns, the banking sector in fact amplified the crisis via the "diabolical loop" between banks and sovereigns. In these ways the structure of the euro area banking sector interacted with the incomplete construction of our institutions to aggravate financial fragmentation and credit constraints, thereby magnifying the impact of balance sheet adjustment on the economy.
Other mechanisms which could have cushioned the shock also turned out to be unavailable. The possibility for a sequenced macroeconomic deleveraging, where the public sector leverages while the private sector deleverages, was hindered by fragile public finances entering the crisis and a weak fiscal framework that could not maintain market confidence in public debt. Fiscal policy has therefore been largely pro-cyclical, and all sectors have had to deleverage simultaneously in a context of weak nominal growth.
Ways to accelerate balance sheet restructuring while minimising its fallout on the economy were also inoperative due to shortcomings in the legal frameworks of many countries. In particular, weak national insolvency frameworks and slow judicial systems made write-offs and resolution of bad loans a more negligible source of debt reabsorption in the euro area, especially compared with other post-crisis experiences, such as that of the US or the UK. All this has contributed to continuing savings and investment imbalances.
The final component affecting long-term interest rates is the term premium. Here too longer-term forces have mingled with more cyclical dynamics connected to the financial crisis to push nominal interest rates down. Term premia in the euro area have been suppressed - including before the crisis - by supply and demand imbalances for safe assets at the global level, especially as increasing demand from foreign investors has interacted with a shrinking supply of highly-rated securities denominated in euro, owing to crisis-time downgrades of certain sovereign issuers and progress in fiscal consolidation in some other AAA issuers. More recently, regulatory changes under Basel III have boosted demand for safe assets even further. Safe-haven flows in periods of financial unrest have also exacerbated these effects.
The role of monetary policy7
Monetary policy has certainly contributed to reducing interest rates. Yet our policy should be understood as a response to this challenging context. Monetary policy cannot affect the secular forces weighing on the euro area economy, nor can it provide an answer to our institutional and structural questions. What it can and must do, however, is respond to the weakness in aggregate demand and the disinflationary pressures that creates.
That is essentially the course we have been steering with our policy since 2008, and it is in fact remarkable that, given the strength of the headwinds, inflation has remained relatively stable. We have always addressed signs of a de-anchoring of inflation expectations with determination. This is testament to the role of monetary policy in preserving a nominal anchor.
Given the depth and length of the crisis, however, that anchor could not be provided through conventional monetary policy alone. The short-term interest rate has eventually run into its effective lower bound, reducing the potential for further rates cuts to support the economy. And even before the short-term rate became constrained in this way, we have also, unlike other economic areas, had to deal with serious disruptions in the transmission of our policy caused by the financial fragmentation of the euro area. We have therefore had no choice but to resort to unconventional measures to ensure price stability in this environment.
Operating in a bank-centric economy, the main focus of our unconventional measures in the early phase of the crisis was ensuring the transmission of our policy across banking institutions, in the context of a generalised collapse in the interbank market. Then, when fragmentation became more pronounced along national lines in the post-debt crisis phase starting in 2011, our focus shifted more to ensuring adequate liquidity to banking sectors. By the middle of last year, however, it became clear that the measures we had adopted to provide a liquidity backstop for banks were becoming less effective, in terms of their pass-through into credit conditions for the real economy, as banks were actively deleveraging. We therefore adopted the Targeted Long-Term Refinancing Operations (TLTRO), as part of our credit easing package, in order to create incentives for banks to lock-in low long-term central bank funding and originate cheaper loans to the real economy.
After we launched this credit easing in the summer 2014, the inflation picture materially worsened. Throughout the second half of the year macroeconomic outlook deteriorated and underlying inflation dynamics weakened, which began to spillover into long-term inflation expectations. That fall in inflation expectations also corresponded to a rise in real interest rates - an effective monetary tightening. Faced with growing disinflationary risks, and with the scope for rate cuts now exhausted, we at that point moved to asset purchases as the main tool to expand our monetary policy stance. This was the context in which we launched our expanded Asset Purchase Programme (APP).
We are generally satisfied with how these measures are working so far, and we will continue our monthly asset purchases until the end of September 2016, or beyond if necessary, and in any case until we see a sustained adjustment in the path of inflation that is consistent with our price stability objective. Notably, our asset purchases have contributed to a compression of term premia, which are currently negative according to several model-based estimates.
Our measures are also having a positive impact on transmission. Banks' intermediation capacity is strengthening via the incipient improvement in their asset quality and increased profitability. Bank lending rates for both households and non-financial corporations (NFCs) are converging downwards across euro area countries, with those banks that participated in at least one of the TLTROs reducing lending rates to NFCs - including SMEs - more than non-participating banks. And the July 2015 Bank Lending Survey showed improving lending conditions, which we expect to support the recovery in loan growth.
Banks have so far been able to buffer borrowers from the higher rates that they now face in the market following the recent re-pricing in financial markets. The improving macroeconomic environment has lowered other costs faced by banks', such as the amount of provisioning required for non-performing loans, and has thereby allowed them not to pass-through their higher cost of financing to borrowers. Still, credit is starting from a low base and volumes on aggregate remain relatively weak. The justification for our credit easing therefore remains very much in place. Moreover, the Governing Council will remain vigilant that recent volatility does not materially affect the broad array of financial conditions and therefore lead to an unwarranted tightening of the monetary policy stance.
On the real side, while our measures are feeding through to domestic demand, we do observe a deceleration in the pace of economic recovery. The latest ECB staff projections see lower growth in 2015, 2016 and 2017 as well as a slower increase in inflation rates compared with the previous exercise in June. The weaker growth outlook primarily reflects slower growth in emerging markets, while the downward revisions to inflation are largely explained by lower oil prices. On account of the external environment, risks to both euro area activity and inflation are on the downside, and these risks have intensified since the cut-off date of 12 August for the technical assumptions which underpin the projections.
In this context, the Governing Council will closely monitor the risks to the inflation outlook over the medium-term. It has emphasised its willingness and ability to act, if warranted, by using all the instruments available within its mandate. In particular, the asset purchase programme contains sufficient flexibility to adjust its size, composition and duration.
To sum up, low interest rates are ultimately a consequence of weak secular trends, coupled with the cyclical consequences of a complex debt crisis, exacerbated by a monetary union with institutional and structural flaws. "Normalisation" is therefore not about the central bank raising policy rates or removing the accommodation provided through its unconventional policy measures - this would only result in too tight a monetary policy stance and make the situation worse - it is about addressing each of these underlying drivers that are holding real rates and inflation expectations down.
Addressing structural and institutional challenges
Monetary easing alone cannot achieve this. Certainly, our policy creates the pre-conditions for rates to rise back towards more normal levels. When expected real rates are below expectations of economic growth - as we see at present - there should be incentives for entrepreneurs to capitalise on this differential to invest in new projects, increasing the demand for funds. Incomes also rise faster than debts, supporting a faster deleveraging and reducing the supply of funds. But monetary policy is only a necessary condition for "normalisation"; it is not a sufficient one.
If monetary policy is to have its full effects on the economic cycle, it needs to be supported by policies that help dislodge negative sentiment about the euro area's prospects and induce higher investment. It is surprising, for example, that businesses currently perceive themselves to have limited excess capacity, despite a large estimated output gap. Expectations of long-term prospects are also depressed. If such pessimism were to persist, we would risk that low investment and high unemployment enter a feedback loop, increasing the likelihood of the economy settling into a form of "underemployment equilibrium". Cyclical drags would then become structural.
In my view addressing this situation has three parts.
The first is boosting the euro area's long-term growth prospects, which given our demographics comes down chiefly to raising TFP. The determinants of TFP growth are complex, but broadly speaking it is a function of within-firm and across-firm efficiency. The former relates to how well firms innovate and adopt new technology, while the latter relates to how well capital and labour reallocate to the most productive firms. Structural reforms are decisive in both cases. OECD research finds that structural factors produce considerable differences in productivity diffusion from frontier firms8 as well in the responsiveness of labour and capital to innovative activity.9
The second part is accelerating the process of private sector debt workout within the euro area. Raising productivity growth would certainly support this, insofar as higher long-term growth increases future debt service capacity. But reductions in numerator of the debt ratio - nominal debt - must also be part of the equation. The provisioning for non-performing loans as part of the Comprehensive Assessment takes us halfway towards a solution. What is now key is that written-down loans can be quickly restructured through more efficient insolvency regimes10 or worked out privately through deepening markets for distressed debt. The Capital Markets Union project could be an important catalyst here as it should ultimately induce convergence in both these areas.
The third part is addressing the institutional incompleteness of our monetary union. There are several dimensions to this, as laid out in the recent Five Presidents' Report, but a priority is completing Banking Union and the integration of the euro area banking sector. This could have a direct impact on investment, insofar as it encourages financial reintegration and reduces credit frictions. Institutional reform is also necessary, and would act as a statement of intent on behalf of members of their lasting commitment to the euro area.
Progress on these three fronts would not only go a long way towards removing the factors preventing higher and sustainable growth. It would also accelerate the normalisation of interest rates away from low levels.
Let me conclude.
What I have tried to explain today is that the central bank policies should not be interpreted as a cause of low interest rates but rather as a consequence of economic malaise and disinflationary pressures, which the central bank aims to forestall. That is to say, they are simply the mirror image of global and euro area specific factors that are pulling both long-term nominal and real interest rates down to very low levels. Addressing those factors with determination can accelerate the process of "normalisation".
That implies on our side the full implementation of the asset purchase programme, on which the cyclical recovery hinges. And it implies on the side of other policymakers addressing the structural and institutional factors that are preventing our monetary policy from producing a stronger and more sustained recovery. In other words, a lasting return to normalcy depends on all policymakers playing their part - and the longer others wait, the greater the hurdles that the euro area will ultimately have to climb.
1 See ECB (2015), "Financial Stability Review", May.
2 EIOPA (2014). "EIOPA Insurance Stress Test 2014", available at https://eiopa.europa.eu/.
3 EIOPA (2014), "Low interest rate environment stock taking exercise 2014", available at https://eiopa.europa.eu.
4 Alves, I., J. Brinkhoff, S. Georgiev, J.C. Héam, I. Moldovan and M. Scotto di Carlo (2015), "Network analysis of the EU insurance sector", ESRB Occasional Paper No. 7 (July).
5 For this section see charts 1-12.
6 Kara, E. and von Thadden, L., INTEREST RATE EFFECTS OF DEMOGRAPHIC CHANGES IN A NEW KEYNESIAN LIFE-CYCLE FRAMEWORK, Macroeconomic Dynamics, forthcoming, (Published online: 14 July 2014.)
7 For this section see charts 15-17.
8 OECD, "The Future of Productivity", OECD Policy Note, 2015.
9 Andrews, D., C. Criscuolo and C. Menon (2013), "Do Resources Flow to Innovative Firms? Cross-Country Evidence from Firm-Level Data", OECD Economics Department Working Papers.
10 For more on this point see IMF Global Financial Stability Review, April 2015.