Over the past seven years, the financial system has been severely shaken, in particular one of its major components: the banking system.
The adjustment process is well-advanced, as witnessed by higher liquidity and capital buffers; better crisis management institutions.
The finalisation of the ECB comprehensive assessment will be a key milestone in the near term.
But this will not be the end of the road: looking ahead, we also need to build a financial landscape that is more diversified, more capital market-based for all segments of financial products. It is with this perspective that the ECB has singled out securitisation as one priority; but there are others, such as venture capital.
A note of caution however: the financial system cannot contribute to growth if the underlying fundamentals are not sound.
Contribution of monetary policy and the ECB in particular
Throughout the crisis, the ECB has deployed a number of instruments, conventional and unconventional.
Our approach to crisis management and inflation control in pursuit of our objective was mostly founded on an expansion in size and tenor of temporary credit operations with our counterparties. This reflected the predominant role of banks in euro area financial intermediation and, hence, for the financing of the euro area real economy.
In the post Lehman environment (when banks in the euro area suffered the wave of contagion from the US subprime confidence crisis) and, again, in the face of an escalation in the debt crisis (when banks in selected countries suffered the consequences of a near- loss of market access of their sovereigns) our temporary credit operations were adapted so as to reduce banks’ liquidity and roll-over risks. This meant providing banks with unlimited liquidity at a fixed price and with expanded maturity and against a wider pool of collateral:
In the aftermath of Lehman’s demise, we changed the liquidity allotment system from one where we fixed the quantity of liquidity lent to banks (being confident that the market would distribute that quantity widely at a price that reflected our intended monetary policy stance) to one where we fixed the price of liquidity and let the market decide over the quantity that is needed to support that price.
Also, in the aftermath of Lehman’s default we extended the term at which liquidity was lent to banks from three to six months to one year (in June 2009) and we broadened collateral rules. By allowing banks to liquefy a larger share of their assets and by stabilising the liquidity horizon of those banks in liquidity deficit, which had lost access to the money market, we averted a major disruption in the first stages of the transmission mechanism.
In the face of the 2011-2012 escalation of the debt crisis (when the confidence collapse reached two systemic countries) we concentrated on banks’ loss of access to the wholesale market to address a major roll-over risk. We conducted two three-year operations (LTROs) that helped banks in a wide part of the euro area to do funding substitution. A significant share of the increase in Eurosystem credit to banks in Spain and Italy, for example, was used to reimburse bank securities that were coming due over the next 12 months and would not have been rolled-over by the market. Without this relief banks would have had to shed assets on a large scale (given high leverage), with potentially systemic implications for the euro area economy as a whole.
Finally, when a major tail risk emerged in the securities market in mid-2012, as more and more weight was given in the pricing of market debt to the risk of a euro area break-up, we announced our decision to conduct Outright Monetary Transactions in the market for sovereign securities, conditional on the issuer submitting to strict multilateral surveillance.
In a very difficult environment the ECB succeeded in preserving its credibility, as witnessed by the anchoring of inflation expectations throughout the crisis.
Over the following 12 months or so, the cumulative effects of this action were able to eliminate fears of disaster events, remove tail-risk from the market and thereby reduce fragmentation.
However, with growing evidence since the start of 2013, banks in large parts of the euro area started to respond to the funding challenges and the need to reinforce their capital structure by reducing their activities, in particular their loans to the real economy. The funding crisis had morphed into a trend which could lead to a widespread credit crunch.
Therefore, in May 2013 against the background of subdued monetary and loan dynamics and very weak economic sentiment, and with the aim to support prospects for a recovery, we decided to cut our key interest rates further. This, in conjunction with the abundant liquidity conditions that had been created as a consequence of the 3-year LTROs, helped stabilise the overnight money market rates around levels close to the lower bound of our monetary policy corridor – which had been reduced to zero in July 2012.
However, soon thereafter, market expectations about the likely future evolution of monetary policy worldwide changed abruptly, notably as a consequence of Fed communication about their intention to end its bond-buying programme. And this change of market sentiment introduced heightened volatility in the term structure of money markets rates of the euro area as well, beyond very short maturities. The sharp swings in the term structure that we observed as a consequence starting in late May threatened to lead to a withdrawal of the monetary accommodation that we had introduced in early May.
As a response, the Governing Council decided to clarify its policy orientations going forward, conditional on the outlook for inflation and weak economic activity. Our forward guidance concerning the future likely evolution of our key interest rates which the Governing Council started to provide in July 2013 was a strong signal by the Governing Council on its commitment to fulfil its mandate and anchor market interest rates around levels that were considered consistent with the underlying macroeconomic conditions in the euro area. The forward guidance formulation that was adopted in July 2013 – and substantially reconfirmed subsequently – communicated our assessment of the economic situation, reiterated our two-pillar strategy and the implications on the monetary stance.
It was particularly important to decouple the monetary conditions of the euro area from those prevailing in other major currency areas, in a context where the monetary policy cycle in the euro area was different from that in the US and the UK. Overall our guidance succeeded in securing our accommodative stance and promoting more stable market conditions.
Consistent with our guidance, and with a view to reinforcing its grip on the interest rates in the private interbank market, in November 2013 we further cut our interest rates and we reduced the width of the monetary policy corridor to 25 basis points.
In March 2014 we reinforced our guidance by referring to the degree of economic slack and subdued money and credit creation as conditions for reducing or maintaining rates unchanged even in the face of a recovering economy.
More recently we have been confronted with a situation of successive downward revisions of our expected inflation path. To give you an example, in December 2013 Eurosystem staff projected HICP inflation at 1.1% for 2014. In the most recent projection round of June 2014 – that is, about six months later – inflation was expected to be only 0.7% in 2014. While surprises were initially more in headline (as opposed to core) inflation and stressed (as opposed to non-stressed) countries, very recently the surprises have been more in core inflation figures, and in non-stressed countries. The reasons are well-known:
A mixed bag of supply shocks (energy & food; structural reforms in strained countries) and persistence of underutilised capacity in a wide area of the euro area…
… in a context in which banks’ credit – a critical ingredient to a self-sustained recovery of investment and durable consumption – was withdrawn in many countries and/or made available to companies and households at interest rates probably only partially justified by increasing lending risk.
To address risks and implications associated with a prolonged period of low inflation, we have to ensure that the developments do not become entrenched in inflation expectations. This is why we further cut our interest rate last month in the context of a broader package of mutually reinforcing measures aimed at addressing impairments in lending conditions and the knock-on effects that these impairments are having for the real economy and inflation. Loan growth is still contracting; lending rates have not declined in line with improvements in financial markets.
While the lending rates (for stressed countries) reflect credit risk, borrowers’ credit risk is in part endogenous. High lending rates contribute to higher loan delinquencies, which in turn increase the need for banks to make provisions on past legacy loans by increasing lending rates; so, ex post, higher delinquencies justify high lending rates, and there is a mutually reinforcing spiral of high lending rates, high credit risk and poor macroeconomic conditions.
The TLTROs are designed to contribute to breaking this spiral. They are expected to ease overly tight conditions through several channels.
The first and most important channel is through a reduction in term funding costs for banks. Note that the interest rate on the TLTROs will be fixed over the life of each operation, at the rate on the MROs prevailing at the time of take-up, plus a fixed spread of 10 basis points. So if a bank does well in its lending performance, it enjoys very attractive refinancing conditions for up to four years. Funding relief, however, does not per se guarantee better credit conditions for banks’ customers, unless the supply of loans shifts in parallel and lending mark-ups are kept constant or even pushed down as banks are encouraged to move down the demand curve for loans by chasing good credit with lower lending rates. This is why the targeted nature of the TLTRO is important: by making funding relief conditional on generation of new lending volumes, the TLTRO will encourage a shift outward in the credit supply curve. By simply moving along the demand schedule, this outward shift will reduce the price for lending while increasing new loans. If banks do not manage to exceed a certain benchmark in terms of net lending, they will not benefit from the additional allowances that can be earned under the TLTROs. This shows that the TLTROs are indeed targeted, rather than a broad-based unconditional provision of liquidity as in the case of the earlier 3-year LTROs.
Second, and as a side effect, the TLTROs are expected to increase excess liquidity, which reduces interest rates in the short- to intermediary-term money market. Remember that banks take into account the money market curve when pricing their loans as the pricing basis over which they charge additional lending spreads in order to recover funding and capital costs. By keeping the money market base rate stable at compressed levels, the TLTROs will reinforce the lowering of the ECB’s key interest rates.
Third, by partially replacing term funding in the form of more expensive bank bonds, the TLTROs can create a scarcity of investible assets, which will result in lower yields and easier market funding conditions even for banks that have not taken part in the operations. It may also create spillover effects to other segments of the corporate credit market, as investors in bank bonds will be induced - by a scarcity of supply – to diversify away from that market and re-invest in other market segments. The latter mechanism is not dissimilar to the “portfolio balance channel”. 
The TLTROs are targeted operations. Indeed, there are positive and negative incentives for banks to use the funds for lending, rather than for alternative investments. First, due to the large multiplier that converts the flows of a bank’s new net lending to non-financial corporations and consumers (relative to a specified benchmark) into additional borrowing allowances under the facility, there is a strong inducement for banks to beat their lending benchmark and expand lending beyond their pre-announcement plans. Second, banks that as of April 2016 cannot prove that they have beaten their benchmark will be asked to reimburse in advance funds that otherwise could be borrowed for a period of up to 4 years.
Through these combined effects, we believe that a widespread use of this facility has the potential to halt the vicious circle of constrained lending, weak macroeconomic conditions and elevated loan delinquencies, and re-ignite a positive “credit multiplier” process.
And in fact the combination of monetary policy measures decided last month already led to a further easing of the monetary policy stance. In particular:
We have seen a marked downward shift in the whole term-structure of the money market. For example, in the overnight segment, EONIA fell by 12 bps to 0.03% between 4 June and 1 July. Over the same period, the 3-month EURIBOR declined by 10 bps and the 1year 4year OIS forward rate declined by 23 bps.
To gauge the overall easing effect of our measures one may go back a bit more in time and look at developments since early May, when the Governing Council signalled its readiness to act in June. For instance over the period 7 May [the day before the May GC] and 1 July the 3-month EURIBOR declined by 13 bps and 1y4y OIS forward rate declined by 36 bps.
Since early May we have also seen some compression of liquidity premia [as for example evidenced by a decline in the spread between KfW bonds and German bund yields with the same residual maturity] and government bond spreads.
The monetary operations to take place over the coming months will add to this accommodation.
All measures together should support lending to the real economy, support the economic recovery and – through that avenue – steer inflation rates to levels closer to 2%.
As usual, effects on the real economy will take time as the measures will need to work their way through the economy.
Monetary policy cannot do it alone… and did not do it alone
Although the policy reactions were often too weak and delayed, they have finally been decided and implemented in a number of countries: structural adjustment at national level has proceeded (not sufficiently in all countries). Also institutional reforms at euro area level have been launched. The Banking Union is a key element. It will not only lead to a more effective and resilient banking system but also to better private sector risk-sharing across countries.
The impact on financial markets has been impressive. But let me be clear: the work is not yet done.
Potential implications that a protracted period of monetary accommodation may have on financial stability
Finally questions have been raised on the potential implications that a protracted period of monetary accommodation may have on financial stability.
First of all, not acting forcefully would have had very serious financial stability implications.
Second, looking forward, our monetary policy accommodation is essential to keep medium to long-term inflation expectations well anchored also for financial stability reasons: a fall of expected inflation leads to an increase in expected real interest rate which makes debt sustainability more difficult.
That being said, monetary policy needs to be aware of the potential implications that a protracted period of monetary accommodation may have for financial stability. There are two major risks: first, postponement in bank balance sheet repair; and second, bubbles in asset prices.
The first risk is that cheap funding allows banks to back-load the deleveraging of their balance sheets and encourage so-called “ever-greening” practices which in turn may prolong macroeconomic stagnation. After a financial crisis, banks have to undergo a thorough clean-up. Otherwise, they may continue to roll over legacy loans, thus restricting the resources available for new lending in support of the real economy – so-called “ever-greening”. This risk is being countered through the ECB’s on-going comprehensive assessment of banks’ balance sheets – consisting of the asset quality review and a stress test – which will be completed roughly in parallel with the launch of the first TLTRO. Thus, banks will not be able to benefit from funding relief without also repairing their balance sheets where necessary. Also, macroprudential measures should be used, if needed, to smooth the financial cycle in specific sectors or jurisdiction.
The design of the measures should also limit to some extent the second risk to financial stability, the build-up of bubbles. The TLTROs exclude explicitly loans to households for house purchases. Furthermore, credit developments are always closely monitored when the assessment from the economic analysis is cross-checked with the signals coming from the monetary analysis, the two pillars of the ECB’s monetary policy strategy. Thus, concerns that the measure could stoke a housing bubble seem unwarranted. Nonetheless, it is important to keep in mind the distinction between what monetary policy can and should do, and what is the domain of macro-prudential policy. Monetary policy has the clear primary objective to maintain price stability in the euro area over the medium-term. Macro-prudential measures should be used, if needed, to smooth the financial cycle in specific sectors or jurisdictions.
I would like to thanks Jan Hammermann und Arthur Saint-Guilhem for their contribution in the preparation of this speech.
The existence of a “portfolio balance channel” has been extensively discussed recently in the context of the Fed’s large scale asset purchases. There is a vast literature assessing empirically the portfolio balance channel in that context. See, for example, Gagnon, Joseph, Matthew Raskin, Julie Remache and Brian Sack (2011), “The Financial Market Effects of the Federal Reserve’s Large-Scale Asset Purchases”, International Journal of Central Banking 7 (1), pp. 3-43.